<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Rik Lomas on Medium]]></title>
        <description><![CDATA[Stories by Rik Lomas on Medium]]></description>
        <link>https://medium.com/@riklomas?source=rss-9ee66467202a------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*ceOXE0jQrPbLQqpkagcjoA.jpeg</url>
            <title>Stories by Rik Lomas on Medium</title>
            <link>https://medium.com/@riklomas?source=rss-9ee66467202a------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 16:59:39 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@riklomas/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Announcing SuperHi Studio]]></title>
            <link>https://medium.com/@riklomas/announcing-superhi-studio-d84a9c4bd5f7?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/d84a9c4bd5f7</guid>
            <category><![CDATA[coding]]></category>
            <category><![CDATA[design]]></category>
            <category><![CDATA[mentorship]]></category>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Thu, 18 May 2017 21:55:49 GMT</pubDate>
            <atom:updated>2017-05-18T21:55:49.111Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*yXud-3IeYrowYhBGzRDmHw.png" /></figure><p>We’re launching a new beta service called SuperHi Studio for our students but also people who are looking to get digital things made.</p><p>We’ve seen a lot of our students want to do real client work after the course but are unsure where to start and are worried about taking on client work and not delivering good work.</p><p>As a solution to this, SuperHi Studio will let clients work with junior coders who are mentored by our expert team. Each student will get paid for the time they work on the project so they could earn their tuition cost back and more.</p><p>For clients, it means superior, professional work, expertly managed at a lot less than agency and freelancer cost.</p><p>For students, it means real world experience that they get paid to make, examples to put on a portfolio plus the opportunity to be mentored by the SuperHi expert team (Lawrence, Rosie and me).</p><p><strong>This trial will only be available to current students and anyone who signs up to </strong><a href="https://www.superhi.com/online"><strong>our 8-week online course</strong></a><strong> before the end of May 2017</strong>. If successful, we’ll extend the service to future students too, but if you want to be part of the trial, sign up now!</p><p>If you’re looking to work with either junior coders or wanting to get a digital product/website made, contact us at <a href="/service/mailto:hi@superhi.com">hi@superhi.com</a> with details.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d84a9c4bd5f7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Long time no update]]></title>
            <link>https://medium.com/@riklomas/long-time-no-update-794902a9a761?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/794902a9a761</guid>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[coding]]></category>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Thu, 26 Jan 2017 23:47:38 GMT</pubDate>
            <atom:updated>2017-01-26T23:47:55.868Z</atom:updated>
            <content:encoded><![CDATA[<p>Hey Medium friends!</p><p>I just wanted to keep you in the loop with what I’m up to.</p><p>My company, SuperHi, has just launched an <a href="https://www.superhi.com/online">online coding course</a> — it’s 8 weeks long and starts on March 20th. If you want to learn how to code with me, sign up!</p><p>If you want to know the first steps about learning to code, I’ve written a guide called the <a href="https://www.superhi.com/first-steps-to-learning-to-code">First Steps to Learning to Code</a> — you’ll get one email every morning between 7–10am about how to learn to code. It’s not technical and we’ve tried to keep it as light as possible.</p><p>I moved to New York a year ago and I’ve written <a href="https://www.superhi.com/blog/how-to-move-to-new-york-city">a guide about how to move to New York City</a> featuring all the tips and tricks I learned the hard way.</p><p>Next week, there’s a 2017 update to my How to Start a Startup that will be posted on the SuperHi blog. <a href="http://twitter.com/riklomas">Follow me</a> or <a href="https://twitter.com/superhi_">SuperHi on Twitter</a> to get the updates. Or if you’re not a Twitterer, <a href="https://www.facebook.com/superhidotcom/">like us on Facebook</a>. Like us even if you are a Twitterer too!</p><p>Rik</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=794902a9a761" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Bootstrapping a startup sucks (and here’s why you should do it)]]></title>
            <link>https://medium.com/make-stuff-happen/bootstrapping-a-startup-sucks-and-heres-why-you-should-do-it-67031340b60e?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/67031340b60e</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[bootstrap]]></category>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Wed, 01 Jun 2016 11:35:01 GMT</pubDate>
            <atom:updated>2016-06-01T20:16:35.804Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kDzxu0rDO2CB2TauO6VpeA.jpeg" /></figure><h4>Let’s go back in time</h4><p>Back in 2012, my old co-founder and I had an idea to start a brand new code school called Steer where we taught people on intensive 5-day courses. We’d met at a General Assembly course (I was the teacher, she was a student) and we felt we could make coding courses a lot better.</p><p>We raised investment money quickly. Too quickly. We were total newbies to investment and we gave away <em>far</em> too much equity (55% of the company for £100k!) to the wrong type of investors. To be fair, the investors cared about us personally but didn’t really understand the business vision. They weren’t tech- or education-focused, they just wanted a return on their money, Shark-Tank-style. Again, there’s nothing with this, but looking back, it wasn’t the right decision for us.</p><p>When I left Steer about 2 years ago, I was determined to not make the same mistakes again. No giving away too much equity and yes to finding the right investors if I did want to take investment.</p><p>I had one rule that I wanted to stick to before I raised any kind of money.</p><blockquote>“Make something awesome that people love.”</blockquote><p>I generally try and stick to that rule for most things — side projects, client work, etc. — but for the main project, it’s something you need to completely commit to. Now that quote might seem like some startup bullshit that <a href="https://www.wework.com/">WeWork</a> would put on their walls, but if some of your first users aren’t saying “holy shit”, then you need to check if it’s something people <em>love</em> and not just <em>like</em>.</p><h4>How to bootstrap</h4><p>Currently, I’m working on a startup called <a href="https://www.superhi.com/">SuperHi</a> — it’s a combination of an online code school, a beginners code editor and a web hosting service. I’ve been working on it really hard for months and months. It’s not launched yet but you can sign up to the waiting list and I’ll let you know when it’s ready (hopefully in the next few weeks!).</p><p>I launched a site called <a href="https://www.superhi.school/">SuperHi School</a> on Friday. SuperHi School is different than <a href="https://www.superhi.com/">SuperHi</a>. The schools are in-person whereas the dot-com version will be fully online. If you’re in London in late June and you want to learn how to create websites with me teaching you IRL, please come along! It’s only £100 for a day. Total bargain.</p><p>There’s two reasons for wanting to do in-person classes as well as the online classes. Firstly, the in-person classes make the online classes better — the more I see going well, the more I can improve the classes going forwards. I’ve been trying out this material for around a year now, constantly improving it and it’s really good.</p><p>Using SuperHi, I can teach the same type of material 3–4 times as quickly compared to how I used to teach at General Assembly or Steer, without any drop in understanding. I can teach what was a 3 day course in around 6 hours now.</p><p>Secondly, the in-person classes help fund the online classes. A lot of people think that SuperHi is more than one person (I often say the royal “we” in copywriting) and others think that we’re a funded startup. I currently bootstrap, which essentially means I pay for it myself using freelance earning. The branding… I paid for. Freelancers… I paid for. Server costs… you get the idea.</p><h4>The downsides of bootstrapping</h4><p>It’s expensive to bootstrap. I’m lucky and privileged that I have a job where I get paid well for my skills and time. Not everyone has that same kind of income to get to the point where I am at with SuperHi (and I haven’t even launched).</p><p>However, the worst thing about bootstrapping is the slow progress. I remember saying to people that SuperHi would be ready in March… of last year. When you’re a team of one or two working on an idea, it can be frustratingly slow progress, especially if you’re doing freelance work to pay your bills. You can see why people raise investment money so they can hire staff and ramp up the process a lot more.</p><h4>Should you raise investment?</h4><p>Probably not just yet.</p><h4>Hey! Why not?</h4><p>Last year, I mentored around 50–60 different early stage startups at <a href="http://fi.co/">Founder Institute</a> and <a href="http://www.escapethecity.org/">Escape the City</a>. 80% of them had imminent investment plans. Of those 80%, most were without real proof that their business worked or had the potential to work. I would say that it would have been a better use of their time to work on their product and make their startup the best it can be rather than focus on raising money.</p><p>Imagine you were an investor…</p><p>Someone pitches you with an idea. No working prototype. No users. No revenue. No testing of the idea.</p><p>Now imagine someone pitches you and they have a working prototype, a small amount of users but enough to test the idea, and a ton of feedback. Even if the design sucks. Or the name sucks. Or it’s buggy. Or you don’t have any revenue just yet.</p><p>Investors are risk-adverse. They don’t want to lose money. The less risky you can prove your startup to be, the more likely you’re not wasting their and more importantly your time by pitching.</p><p>If you are a non-technical person trying to start a tech startup, the best advice I have is to do anything you can that isn’t the tech. Do all the wireframes, work out how you can market it, <a href="https://milanpevec.files.wordpress.com/2013/06/lean-canvas_blank.png">fill in your lean canvas</a>, write your business plan, <a href="https://medium.com/@riklomas/minimizing-the-minimum-viable-product-6c47deb15b77#.kf88c8nsb">use tools that are out there already</a>. Anything but the tech side until you’re completely ready to do that.</p><h4>When will you be ready to take investment?</h4><p>July last year, I decided to go and pitch an investor — one of the best ones in London too — <a href="https://twitter.com/eileentso">Eileen</a> from <a href="http://www.passioncapital.com/">Passion Capital</a>. SuperHi was a lot rougher and way less feature-complete than it is today but it was working. It existed. I had a plan for launch, a plan for what I would do with investment and a plan for the future. As the meeting went on, I got so much useful feedback from her and essentially it was the nicest “no” I could have received. I won’t go into details but I was very happy leaving the meeting.</p><p>A friend of mine is doing the opposite. I went to an investment meeting with her once as she wanted some back-up. She just had an idea of what the startup was. No prototype, no design work, not even a name. The investor ripped her to shreds. I sat there feeling very awkward and I felt so bad for her but the investor was right — she wasted his time.</p><h4>Do you need to take investment?</h4><p>Maybe not! There are plenty of companies out there who have never taken any kind of money at all. <a href="https://m.signalvnoise.com/reconsider-41adf356857f">Basecamp are probably the most vocal of all the tech startups</a> but companies like <a href="http://blog.mailchimp.com/investor-relations-at-mailchimp/">Mailchimp have never taken money</a> either:</p><blockquote>MailChimp is weird in a lot of ways. One thing that makes us different from many other tech companies is that we don’t have investors. We’re privately owned, bootstrapped, and proud of it. We chart our own course, guided by our core values and the needs of our users.</blockquote><p>You can always think about taking money later too. The awesome <a href="https://twitter.com/destroytoday">Jonnie</a> from <a href="http://cushionapp.com/">Cushion</a> talks about this on a <a href="http://cushionapp.com/journal/funding-cushion/">recent blog post</a>.</p><blockquote>When I started Cushion, if someone were to tell me I would one day take investment, I wouldn’t have believed them. I also wouldn’t have believed that I would hire a team. Over the past two years, however, Cushion has grown from an idea into something real, and I’ve grown with it. Considering where Cushion is now and where I want it to go, I know this is the right decision, and most importantly, with the right people.</blockquote><p>Always think about how you can fairly charge for your service. That money can be used to continually improve it for your users.</p><h4>Looping it around</h4><p>A close friend of mine is a personal trainer. He was showing me how to do pull-ups one day and I told him I can’t do them very well. The technique to do it is every time you go to the gym, grab the bar and just hold yourself as long as possible. First few times, you’ll drop off after a few seconds. Later, you’ll be able to hold for around 10 seconds. Further down the line, you’ll be able to finally do them properly.</p><p>Startups are like that. You can’t be instantly good. Having someone hold your feet at the start doesn’t help you in the long term. Later, you may need help to go to the next level. But you may not even need to.</p><p>This is the reason for me putting on <a href="https://www.superhi.school/">SuperHi School</a>. SuperHi (online version) isn’t live yet but I do have a few customers from the 200 beta testers but it’s not enough to pay the bills just yet. Putting on paid-for, in-person code courses helps me grow the startup and means I can continually make it better and better for everyone.</p><p>Bootstrapping can suck and you may feel like you’re travelling up-stream very slowly. Features might take weeks instead of days. You’re busy with paid work. However, think of this as a good sign — it means that you’re learning from your mistakes early without the financial pressure of investors plus you give yourself breathing room to really think about how your startup is and how you’re solving the problem you set out to fix.</p><p><em>As always I’m happy to give out advice if you’re running a new startup, just email me at rik@superhi.com. Sign up at </em><a href="https://www.superhi.com/"><em>superhi.com</em></a><em> to be the first to know when I launch!</em></p><p><em>Thanks to </em><a href="http://elliotbentley.com/"><em>Elliot Bentley</em></a><em> for his help with this article.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=67031340b60e" width="1" height="1" alt=""><hr><p><a href="https://medium.com/make-stuff-happen/bootstrapping-a-startup-sucks-and-heres-why-you-should-do-it-67031340b60e">Bootstrapping a startup sucks (and here’s why you should do it)</a> was originally published in <a href="https://medium.com/make-stuff-happen">Make / Stuff / Happen</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Minimizing the Minimum Viable Product]]></title>
            <link>https://medium.com/@riklomas/minimizing-the-minimum-viable-product-6c47deb15b77?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/6c47deb15b77</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[mvp]]></category>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Tue, 23 Feb 2016 19:28:24 GMT</pubDate>
            <atom:updated>2016-02-23T19:28:24.531Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ekpjUJ5vu3aYqBoMq1Zk7Q.jpeg" /></figure><p>Over the past few years and especially since I wrote <em>“</em><a href="https://medium.com/makers-gonna-make/how-to-start-a-startup-without-ruining-your-life-af45eec09406#.b2u7meqok"><em>How to start a startup without ruining your life</em></a><em>”</em>, I’ve received hundreds of emails from people wanting to start a new company. Lots pitch their <strong>big</strong> idea to me — what the business will look like in 3–5 years time — that’s great but it skips out the detail of how to actually get to that point from now. In this article, I’m going to talk about how to simplify your Minimum Viable Product (the first version of your business that your customers will interact with — I’ll call this MVP from now on) down into something a lot more attainable and a lot less expensive. This may not work for some ideas but it may for yours.</p><p>Most people’s MVPs are not minimums. They add in too many features or think too big. But the main problem is they need people from outside the startup to help them do the things they want to do. For instance, in tech startups these people are usually developers or designers. A lot of these startups could start without needing a developer but they need to reduce what they’re trying to achieve with the MVP.</p><p>Remember that a MVP is to <strong>test the business idea</strong>. You want to do this as quickly and cheaply as possible. You don’t have to look like a 30-person company at launch and if anything your <a href="https://en.wikipedia.org/wiki/Early_adopter">early-adopter customers</a> will appreciate that one-on-one feel you get with a scrappy, brand-new company.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QK9hVg4tT52So2O6yhOlmg.jpeg" /></figure><p>One of the emails I received was from a man who wanted to create a marketplace for supper clubs — “Airbnb for food with new people”. He had already planned out what payment platform to use and how all the pages on his webapp would look. However, he had never tried to organize a supper club before.</p><p>What I suggested was to drop the plans for the whole platform until he knew <em>exactly</em> how to do this — how would he know what his customers need if he’d been through the same problems?</p><p>What I suggested was to put up a quick holding page using something like <a href="http://squarespace.com/">SquareSpace</a> with a link to a <a href="http://typeform.com/">Typeform</a> form to collect more information from users, then email those users directly inviting them to his own supper club. Figure out why those users want to come to a supper club and ask them if they’d like to host their own. If not, why not? If so, would they do it on his site if he organised it?</p><p>This approach is lo-fi but it means that you’re testing your idea for next-to-nothing and you don’t need a nerdy coder. The man was prepared to spend thousands on this idea without testing what users wanted. Most potential startups don’t need to build their own tools — they can start with tools already available to them, get that MVP out there quickly, test and then move towards that end goal.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E8knSy5tTjltrUmz4A7xaw.jpeg" /></figure><p>Another email was from a woman whose startup was to create an app where users review restaurants for credit. Cash-back for reviews essentially. Great idea but do users want it? She had saved money for years to be able to afford an iOS developer to make her app and she was now ready to spend that tens of thousands without truly knowing if it was going to work.</p><p>My suggestion was similar to before. Don’t hire an iOS developer and instead, use the tools available until you know it works.</p><p>I told her to find 20 people on <a href="http://www.craigslist.org/">Craigslist</a> in her area (this could be <a href="https://www.gumtree.com/">Gumtree</a> if you’re in the UK, etc), get them to review restaurants via a <a href="https://www.typeform.com/">Typeform</a> and pay them manually for their reviews via bank transfer.</p><p>It’s harder work than automating the whole process but it will save her tens of thousands she was willing to spend automating before she knew it was the right thing to automate. You can use tech without having to build your own tech.</p><p>If those people don’t review, she will have saved thousands or something in her idea needs to be fixed. If those people do review and continue to review, that’s a great sign, she can iterate on the idea based on the feedback and then consider spending her money.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y7ZgwZF83EOzfNPNijHdMQ.jpeg" /></figure><p>This doesn’t work for all startups. Some startups do have to start by spending a lot of money, but if you can get by without having to hire other people to do urgent parts of your business, then you should. This system works for not just the tech industry but many others too.</p><p>I’m guilty of not testing quick enough too. When starting <a href="http://superhi.com/">SuperHi</a> (which still hasn’t launched at the time of writing), I didn’t test it for 10 months. I was working on it a lot but no user had even seen it. To get over this, <a href="http://formfiftyfive.com/2015/06/superhi-school/">I gave away free code lessons</a> to try out what I’d been working on. For the customer, they received free lessons. For me, I received a ton of feedback. Win win. It was the <strong>most valuable thing</strong> I’d done in the first year of making the site.</p><p>That exposure to real people meant that I could see what wasn’t working and gave me deadlines to fix it by. If something wasn’t working well, I had 7 days to change it. Then in the next lesson, does that change fix the original problem? I had 16 guinea pigs to try out changes with, which is infinitely better than 0 guinea pigs.</p><p>This continual exposure to users has meant that the MVP is continually getting stronger. One of the reasons I haven’t launched SuperHi yet is I saw fixable problems within the MVP. I could have launched without testing and had all the problems open to the public — instead I can now fix obvious problems before I launch.</p><p>I doubt there’s many startup founders that regret getting their ideas sooner than they did. The quicker you can get your idea out of your head and give it exposure to real people, the better it will be. That exposure to real life use-cases and iteration will make your MVP way stronger than keeping your initial plan and hoping for the best. Get your idea out there as quickly as possible and learn from your potential customers.</p><p><em>The ideas in this article have been changed and tweaked to protect the owners identities. If you have any idea you’d like to run past me, email me at rik@superhi.com or tweet me at </em><a href="https://twitter.com/riklomas"><em>@riklomas</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6c47deb15b77" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to make a site in 30 minutes that even Kanye West can enjoy]]></title>
            <link>https://medium.com/@riklomas/how-to-make-a-site-in-30-minutes-that-even-kanye-west-can-enjoy-b0c04ba7ae68?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/b0c04ba7ae68</guid>
            <category><![CDATA[kanye-west]]></category>
            <category><![CDATA[ember]]></category>
            <category><![CDATA[javascript]]></category>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Mon, 22 Feb 2016 19:43:18 GMT</pubDate>
            <atom:updated>2016-02-22T22:29:47.081Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CRoDBLYE-lKQ2ij0rMzt7Q.png" /></figure><h3>The Life of Pablol: How to make a site in 30 minutes that even Kanye West can enjoy.</h3><p><em>This is a quick tutorial about </em><a href="http://emberjs.com/"><em>a Javascript framework called Ember</em></a><em> (wait, don’t run!) and how to build a dumb site using it. Not sure how to code at all? I’m working on a new code school called </em><a href="http://www.superhi.com/"><em>SuperHi</em></a><em> — we’re launching soon!</em></p><p><strong>“wana make a generator of kanyes album cover?”</strong><em> </em>said <a href="https://twitter.com/whybray">Simon</a>, popping up in Facebook Messenger. This wasn’t the first time we’d made an album cover generator together. We’d previously made one for Drake’s If You’re Reading This It’s Too Late a few months earlier called “<a href="http://www.ifyouretypingthisitstoolate.com/">If You’re Typing This It’s Too Late</a>” that had a few million hits and coverage in mainstream media, despite spending 45 minutes creating it one Friday evening. It was almost as if the less effort we put into things, the more popular it became.</p><p>After launch, again we have a ton of coverage (example: <a href="http://www.avclub.com/article/make-your-own-version-kanye-wests-life-pablo-album-232176">1</a>, <a href="https://www.thefader.com/2016/02/12/generate-cover-art-life-of-pablo">2</a>, <a href="http://www.telegraph.co.uk/music/news/how-to-make-your-own-cover-art-for-kanye-wests-the-life-of-pablo/">3</a>) and best of all, Kanye himself used the site <strong>twice</strong>!</p><h3>KANYE WEST on Twitter</h3><p>https://t.co/mOps0CD20f</p><h3>KANYE WEST on Twitter</h3><p>pic.twitter.com/CGJ5PYLq67</p><p>I also now have Kanye’s IP address, passwords and log-in details for sale. (I am joking, Kanye’s legal team. No. Really.).</p><p>So how did we make it?</p><p>I had 55 minutes to create this new Kanye cover creator before I had to leave for a Javascript meetup (yes, I am that cool) about a framework called <a href="http://emberjs.com/">Ember</a>. I’ve been using Ember for around a year now, it’s a great library for building complex, interactive websites, and while the general use case isn’t for a dumb album cover generator, it was my first thought to use the power of Ember for the short timeframe we had.</p><p>Simon and I thought let’s keep this super simple, or in his words “pure”. We wanted this to be instantly usable without any user interface — just type and get what you want. From a code point of view, all we need is an input field and when you type in that, it changes the text.</p><p>I already had the latest version of Ember installed on my laptop but it’s a fairly quick thing to do. Just go to <a href="http://emberjs.com/">http://emberjs.com/</a>, open your command line (Terminal on Mac or Command Prompt on Windows) and follow the instructions.</p><p>First thing you need to do create a new application in your usual code folder</p><pre>ember new thelifeofpablol</pre><p>Quick aside, we called it The Life Of Pablol because we were looking through the domains list and saw there are now .lol domain names. Who knew? We felt it fit well enough for a quick joke so it stuck. You can call your project whatever you like as long as you do “ember new [whatever]”.</p><p>You should get something happening that looks like this.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ha3-dutd2GISgRXTyRfoLA.png" /></figure><p>Next, move your command line into the folder you’ve just made by doing:</p><pre>cd thelifeofpablol</pre><p>Open up your folder in a code editor (I use <a href="https://atom.io/">Atom</a> personally) and let’s explore what’s just happened. Ember is set out in a particular structure which makes it very useful if you’re working with other people — everyone knows where everything is.</p><p>The main folder we’ll be concentrating on is the /app folder. This has seven different folders:</p><ul><li>Components: for reusable bits of code across your webapp — if a particular area of code appears multiple times across different pages, it’s probably best as to make it as a component.</li><li>Controllers: the bit that hooks up the routes to the data it needs. Don’t worry about this too much as we’ll talk about this later.</li><li>Helpers: for use in simple data formatting, e.g. turning a date into “two hours ago” would be done in a helper. Think of them as mini-components.</li><li>Models: for talking to your database — we don’t need it for this project but if you have any kind of stored data that may change or update, you’ll probably need to use models in your code.</li><li>Routes: for preparing which URL (e.g. /about) you’re about to go to into. They work closely with controllers. Again more on this later.</li><li>Styles: where your CSS lives</li><li>Templates: where your HTML lives</li></ul><p>For our site, we need one page only — <strong>let’s call this “home”</strong> as it’s the first and only page we’ll hit on this site.</p><p>When I go to the root domain (e.g. “www.thelifeofpablol.com” without any “/about” or “/contact”), I need to handle this particular URL to go to a router first of all. This router will then set up a controller to talk to and all our fun stuff will happen in this controller.</p><p>Let’s generate a “home” route first of all.</p><pre>ember generate route home</pre><p>This will generate all the files we need for the route.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gufLulCOPrfvgYOe5Llm_g.png" /></figure><p>We also will need a controller called “home” too. Let’s generate this too.</p><pre>ember generate controller home</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LLagocdnZzRK6Mx_7JeGpg.png" /></figure><p>Okay, great. Next we need to run the Ember server so we can see what we have in our browser so far.</p><pre>ember server</pre><p>You should get something that looks like:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6fxPQnmRVBYOtmyFawpXdA.png" /></figure><p>If you look, it says “Serving on http://localhost:4200” — in your browser, go to that URL and you should get:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nmfQgWj4crw0-gZ1wTGvtw.png" /></figure><p>Phew. Okay, so we’ve done all the commands that we’ll need for a while. At least we can concentrate on making some cool stuff happen now.</p><p>At the moment, the home route isn’t hooked up with the root URL. By default, it works when you go to “/home” but not the normal “/”. Let’s fix that. Go to your code editor and find the file in <strong>/app/router.js</strong></p><p>There will already be some code set up for one “route”, we just need that route go to the correct place. We want the code for our home route and home controller to work when we go to “/”. Find the line:</p><pre>this.route(&#39;home&#39;);</pre><p>And change it to:</p><pre>this.route(&#39;home&#39;<strong>, { path: &quot;/&quot; }</strong>);</pre><p>Okay, this means our home route and controller will work when we go to this particular URL <em>but </em>we don’t have any code yet. You’ll still get the same thing when you look at your site.</p><p>Next let’s get rid of that “Welcome to Ember”. We don’t want our users seeing that. Go to <strong>/app/templates/application.hbs</strong> and remove the line:</p><pre>&lt;h2 id=&quot;title&quot;&gt;Welcome to Ember&lt;/h2&gt;</pre><p>We don’t need that. Leave the {{outlet}} bit as it is though.</p><p>If you notice in the folder, there’s a application.hbs file and home.hbs file — the application.hbs file is for the entire site, so if you go on “/about” and “/contact” they’ll both use this file, but home.hbs is specific for that route.</p><p>If you want a header across your site, you’d put it in application.hbs whereas if it’s page specific, you would put it in the correct template. For our generator, we only want to do things in the home.hbs so our controller can talk to it.</p><p>Open home.hbs and let’s start using The Power of Ember<strong>™</strong>.</p><p>At the moment, there’s just {{outlet}} in there. Delete that and replace everything with:</p><pre><strong>{{input value=say autofocus=&quot;autofocus&quot;}}</strong></pre><pre><strong>&lt;p&gt;{{say}}&lt;/p&gt;<br>&lt;p&gt;{{say}}&lt;/p&gt;</strong></pre><p>The first line of this is an Ember input helper. Basically it will make a variable in your code called “say” and update it based on whatever you’re typing into that input field. The second attribute (autofocus=”autofocus”) will make your browser focus onto the input field when you load the page</p><p>Below it we have two paragraphs with {{say}} inside. This won’t actually say “{{say}}” when you view it but it will hook up with the variable “say” in your code and replace it.</p><p>Try going to your site and typing in the input field.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jAklTS6LYX0GnFkOQ6Y_8g.png" /></figure><p>Whatever someone types into the input box, they’ll get in the two paragraphs below that. Simple, right? That’s pretty much most of the functionality of the album generator! Let’s add some more things into it.</p><p>Notice at the moment that when we have a empty input field, we have no copy. Wouldn’t it great to have a default value. We can’t do this directly in our template home.hbs so we’ll have to write some code to do this for us.</p><p>Go to your code editor and to the file <strong>/app/controllers/home.js</strong></p><p>At the moment you should have an empty controller that looks like:</p><pre>import Ember from &#39;ember&#39;;</pre><pre>export default Ember.Controller.extend({<br>});</pre><p>Let’s add some code in to give us a default value when our input is empty.</p><pre>import Ember from ‘ember’;</pre><pre>export default Ember.Controller.extend({<br> <br><strong>  sayWithDefault: function () {<br>    if (this.get(&quot;say&quot;)) {<br>      return this.get(&quot;say&quot;);<br>    } else {<br>      return &quot;Start typing&quot;;<br>    }<br>  }.property(&quot;say&quot;)</strong><br> <br>});</pre><p>I’ve created a new function called “sayWithDefault”. Inside the function, I’m looking to see if “say” is not empty:</p><pre>if (this.get(&quot;say&quot;))</pre><p>If it isn’t empty, return back the variable “say”</p><pre>return this.get(“say”);</pre><p>If it is empty, then return back “Start typing” or whatever default you’d like.</p><p>This function is also dependent on the variable “say” — it will change depending on what the user types in so we want to keep this whole function up to date whenever the variable say changes — so we make the function dependent on that.</p><pre>}.property(“say”)</pre><p>Now to use this new function we’ve just made in our template, we need to go back to <strong>/app/templates/home.hbs</strong> and update our code. We can leave our input alone as this is where people are typing in, but the output paragraphs will change.</p><pre>{{input value=say autofocus=&quot;autofocus&quot;}}</pre><pre>&lt;p&gt;{{<strong>sayWithDefault</strong>}}&lt;/p&gt;<br>&lt;p&gt;{{<strong>sayWithDefault</strong>}}&lt;/p&gt;</pre><p>So now when we view our site, by default it will have “Start typing” in the paragraphs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0o4236TKAIk-Civ7k0j-DA.png" /></figure><p>Next, wouldn’t it be great if the URL changes so that it reflects what you’re typing. I want to be able to share my hilarious comment about how Kanye likes fish-sticks that <em>definitely</em> hasn’t dated since <a href="https://en.wikipedia.org/wiki/Fishsticks_(South_Park)">2009</a>. At the moment, if I share the URL it doesn’t pre-fill the input box. Let’s update the URL as we type so we can share our finished result.</p><p>In Ember, this is fairly easy to do. It’s built in to the framework itself. Go to your home controller in <strong>/app/controllers/home.js</strong> and add in the following line:</p><pre><strong>queryParams: [&quot;say&quot;],</strong></pre><p>just before the sayWithDefault function to get something that looks like:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*33H34q-7raic4992ZLaYwQ.png" /></figure><p>Watch out for the comma at the end! This queryParams bit is part of Ember’s Controller classes. It helps keep the URL up to date with that array of variables. Notice the URL change when you type now.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2H3j41kFkIUBEyVNFJ8OQQ.png" /></figure><p>You might wonder why there’s the funny <em>%20</em> between hello and there in the URL. That’s the browser converting our data into a safe string — spaces aren’t allowed in URLs but can be converted to “%20” (URL-speak for gimme a space please). You can now share your URL and this will be the text when your friends see it.</p><p>I’m going to skip the CSS and styling part of the generator as it’s a bit more code and specific to just one album cover. The way we styled up the page for the Kanye generator was to hide the input off the top of the page so no one could see it, then we repeated the text and placed it where it should go on the album cover.</p><p>Let’s add one more thing in before we finish. When you look at the generator on a phone, it doesn’t autofocus. That’s because most mobile browsers correctly think that a keyboard popping up on page load would annoy users. So let’s make something a user can click that would focus them into the input box.</p><p>To do this we’re going to add in an Ember action — something that will happen when a user interacts with it — in this case, when a user clicks on something, it will focus the input box. We could have used this for instance to change the photo on Kanye’s album or to change the color of the background for instance.</p><p>Let’s go to <strong>/app/templates/home.hbs</strong> and at the bottom of the file, add in.</p><pre><strong>&lt;span {{action &quot;focusOnInput&quot;}}&gt;Focus on input&lt;/span&gt;</strong></pre><p>We’ve created a span tag with the text “Focus on input” but with this, added in an Ember action called “focusOnInput” — I’ve just made up that name, I could have called it “doThisDumbThing” instead.</p><p>If you click the span tag right now, it won’t do anything and will give you an error. That’s because it’s not hooked up to anything yet. Let’s make it focus on that input field instead.</p><p>Go back to your home controller in <strong>/app/controllers/home.js</strong>.</p><p>First thing is we’re going to add a comma at the end of the function for sayWithDefault:</p><pre>}.property(&quot;say&quot;)<strong>,</strong></pre><p>Watch out for this, it’s something I forget a lot. It means that there’s more code to your controller class after this.</p><p>Okay, all good. Let’s add in some actions. Below that line, write in:</p><pre><strong>actions: {<br>  focusOnInput() {<br>    Ember.$.find(“input”).get(0).focus();<br>  }<br>}</strong></pre><p>What this means is our controller has some actions (as a Javascript object). You can have as many action functions as you like here too. We’ve just got the one called focusOnInput.</p><p>Inside this focusOnInput function, we have a tiny line of Javascript that runs when the user interacts with the span tag. It will first of all, ask Ember to talk to <a href="http://jquery.com/">jQuery</a> (the dollar sign bit). Then jQuery will select the first “input” in its HTML, and then make that input have focus. Fairly straightforward!</p><p>In our CSS, I made this something that covers the entire page so whenever a mobile user clicks anywhere, they automatically get the keyboard pop up.</p><p>So once you’re finished with your beautiful generator, how do you get it online?</p><p>The Kanye and Drake generators are both hosted on <a href="https://www.heroku.com/">Heroku</a> which handled the insane amount of traffic we got without any problem whatsoever. Simon and I didn’t have to do anything apart from get it online. Sign up for an account if you don’t have one already! Heroku give you a lot of command line tools you can use by installing their <a href="https://toolbelt.heroku.com/">Heroku Toolbelt</a> so that it’s very easy to get the generator live.</p><p>Once you’ve got all the Heroku tools set up, quit out of your Ember server by pressing Ctrl+C (yes, Ctrl even for you Mac users, not Cmd).</p><p>First of all, we’re going to add all our new files to Git by doing:</p><pre>git add .</pre><p>Then we’re going to commit them all to the version control by doing:</p><pre>git commit -am &quot;First version complete&quot;</pre><p>Now we’re going to set up Ember on Heroku’s server (we’ve already have Ember on our laptop).</p><pre>heroku create --buildpack https://github.com/tonycoco/heroku-buildpack-ember-cli.git</pre><p>Put this all one line (it’s pretty long) and press Enter. This will create your Heroku app ready to receive your code.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Zmw_9NQ6Lts6_PP69tGwiA.png" /></figure><p>Next, let’s put the code we’ve written online!</p><pre>git push heroku master</pre><p>This might take a few minutes to setup but once it finishes you should get something that looks like:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OUc7nFY0YXomZglHwtughQ.png" /></figure><p>Done! It says on mine that <a href="https://salty-dusk-42233.herokuapp.com/">https://salty-dusk-42233.herokuapp.com</a> was deployed to Heroku. Yours will be something similar to that. Go to your URL by copy and pasting the URL or by typing “heroku open” into your command line and there you have it! Your own online generator!</p><p>Of course, this is a very simple example of what you can do with Ember and Javascript in general. Hopefully it shows off the power of Ember and why I’ve been using it heavily for the last year or so.</p><p>Little sites won’t change the world but they might just make the world’s biggest ego notice you.</p><p><em>If you liked this article, please press recommend. If you’re interested in what I’m doing with SuperHi or about learning to create websites, sign up at </em><a href="http://www.superhi.com/"><em>http://www.superhi.com/</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b0c04ba7ae68" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How exactly do you beta test an online code school?]]></title>
            <link>https://medium.com/@riklomas/how-exactly-do-you-beta-test-an-online-code-school-fe69038231ee?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/fe69038231ee</guid>
            <category><![CDATA[code]]></category>
            <category><![CDATA[beta-testing]]></category>
            <category><![CDATA[online-education]]></category>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Fri, 10 Jul 2015 11:34:25 GMT</pubDate>
            <atom:updated>2015-07-10T11:34:25.317Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EHBjqNO0q67-rBqhtMu1og.png" /></figure><p><strong><em>TL:DR? I’ve launched a side project called </em></strong><a href="http://superhi.school/"><strong><em>SuperHi School</em></strong></a><strong><em>!</em></strong></p><p>Over the last few months, I’ve been working on a new startup called <a href="http://www.superhi.com/"><strong>SuperHi</strong></a><strong> </strong>— an online coding platform that teaches beginners how to easily create professional websites — and I’m really excited to launch it really soon.</p><p>I’m a big fan of beta testing, especially when it comes to things like education. When I first started teaching people how to code, what I thought would work, just didn’t. People would find it too complicated or didn’t get particular concepts. <strong>It’s tricky to unlearn what you already know</strong> and come at things from a complete beginner standpoint.</p><p>When you’re a classroom, it’s easy to change course or answer questions if students don’t understand something, but with online courses there’s an inherent distance between student and teacher. A lot of my research before starting SuperHi said this is one of that most difficult problems when using online courses. So how do you shorten that knowledge gap?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ks-iDRMqKp_4Ha_BuXHE7w.png" /></figure><p>At the moment, I feel both the SuperHi platform and the material are incredibly strong but I have to make sure users think that too. I was initially planning to beta test with friends of mine but the problem with testing with friends is that they are more likely to give good feedback and more importantly, they’re less likely to give criticism of the things you’ve been working on for months.</p><p>How do you get around this? <strong>Open it up to the public and test it with people who are likely to be your final customers</strong>.</p><p>There’s a lot of people who want to learn to code and willing to pay thousands of pounds for the opportunity. Just look at the plethora of code schools popping up all over the world — I should know as I used to run one of those!</p><p>I’ve decided to give away free, in-person coding lessons to try out the material and the SuperHi platform. If you’re available on Wednesday evenings in August and September and in London, you can apply to <a href="http://superhi.school/"><strong>SuperHi School</strong></a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*v-8l0G3_SrEYZvNAcfEVNw.png" /></figure><p>There’s no catch to the lessons (apart from a couple of potential bugs) and it’s aimed at total beginners. It’ll be a one-off series of six lessons, so if you want to do it, don’t delay! There’s limited spaces so make sure any application is as good as possible.</p><p>The material we’re teaching has been designed by some of the best designers, illustrators and startups in the world, I won’t announce them yet but it’s a fantastically varied selection. Applications shut on the 26th of July.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ks-iDRMqKp_4Ha_BuXHE7w.png" /></figure><p>If you can’t come to the lessons for whatever reason, SuperHi Online will be launching in the next few months (depending on feedback from SuperHi School of course). I honestly think it’ll change how a lot of people learn to code and make websites, and I can’t wait to launch!</p><p>If you’d like to be a beta tester, there’s a short form to fill in on <a href="http://superhi.com/">SuperHi.com</a> and you’ll be the first to try it out.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/700/1*tnPX12a_LnVRTL_yplPX7w.png" /><figcaption>Coming soon!</figcaption></figure><p>There’s more about SuperHi School over at <a href="http://formfiftyfive.com/2015/06/superhi-school/">FormFiftyFive</a> and <a href="http://www.ycn.org/news/1364-learn-to-code-at-superhi-school">YCN</a>. We’re on <a href="http://twitter.com/superhi_">Twitter at @superhi_</a> and you can email me about anything at rik@superhi.com.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fe69038231ee" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Crossing out ideas: the beginning of a new startup]]></title>
            <link>https://medium.com/@riklomas/crossing-out-ideas-the-beginning-of-a-new-startup-2b32742b914e?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/2b32742b914e</guid>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Mon, 03 Nov 2014 15:44:11 GMT</pubDate>
            <atom:updated>2014-11-03T15:44:11.422Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mmjURygU-r63SAjZ0pRDSA.png" /></figure><p><em>Recently I published an article called “</em><a href="https://medium.com/@riklomas/how-to-start-a-startup-without-ruining-your-life-af45eec09406"><em>How to start a startup without ruining your life</em></a><em>” and from today, I’ll be forced to follow my own advice.</em></p><p>Six months ago, I left the code school that I co-founded and since then, I’ve been freelancing at <a href="https://www.picfair.com/">various</a> <a href="https://www.checkframe.com/">London</a> <a href="https://www.provenance.org/join">startups</a>. It’s been fun to get my hands dirty by returning back to design, code and product strategy.</p><p>But recently, I’ve missed teaching how to code. It was something that I got great pleasure from it and even years later, I still love getting emails from old students showing me what they’ve made. If you’ve never taught anyone before, I would highly recommend it. I haven’t been able to start a new startup due to a non-compete clause that ended today but over the last few months, I’ve been back and forth on quite a few ideas.</p><p>I wanted to share my ideas to talk about the reasoning behind them and why I settled on a particular one over others.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oAkQLgd-Pw8bg_EWcBC-NQ.png" /></figure><h3>Dropped idea 1: Superscript</h3><p>My first idea was to make a better and easier to learn Javascript. Whenever I taught, the hardest thing for new coding students seemed to be Javascript. The brackets, the semicolons, the parenthesis and the weird nature of the language made it a difficult language to grasp instantly. Personally, I love an alternative version called <a href="http://coffeescript.org/">Coffeescript</a>, but for beginners Coffeescript is still difficult to learn. I wanted to be make something that was fundamentally easier.</p><p>I dropped this idea due the lack of a business model. I would have loved to spent time making the idea into a reality but it was hard to justify. It would have been hard to bootstrap and investors wouldn’t be into the idea.</p><p>For advanced programmers, it wouldn’t offer significant improvements over Javascript or Coffeescript as it wasn’t designed for them. That made me doubt whether coders would will be willing to fund it via Kickstarter or not. I still think it’s a strong idea so I’ve put it into the pet project folder for now.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L9YwJwvXp_cRwOr3K7tTXA.png" /></figure><h3>Dropped idea 2: New code school</h3><p>My second idea was one I was quite sure that I wanted to do for a good few months. Starting a new code school would have been relatively simple for me — I’ve done it before, I have a good teaching reputation and there’s a decent demand for it in London.</p><p>The idea for it was strong too and it would fix the main two problems that all current London code schools have: make attending more affordable and more flexible to people’s schedules. It was attractive to investors too as it would make revenue from day one pretty much. All it would need is a space and equipment and it was ready to go.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lf8teTg-sqDMfgktd8MMuA.png" /></figure><h3>Getting rid of the nagging feeling</h3><p>For weeks during my freelance period, I was convinced that I would be doing this second idea. 100% sure. This was it. The winning idea. As soon as it hit November, I was going to drop everything and work on this and nothing was going to stop me.</p><p>After I wrote my “<a href="https://medium.com/@riklomas/how-to-start-a-startup-without-ruining-your-life-af45eec09406"><em>How to start a startup</em></a>” article, I had 100+ emails from people all over the world, asking me questions from “is this idea good?” to “I’m not sure if my startup is taking the right direction?”.</p><p>One email in particular was from a teenager in New York asking me how to learn to code. He was really interested but couldn’t afford to go to one of the schools in New York and he found most of the online schools frustrating.</p><p>He reminded me a little of myself when I was that age. I learnt to code a lot later than most of the coders I’ve worked with — most started in their early teens (or younger) and I was in my late teens. <a href="http://www.bbc.co.uk/news/business-23510961">I was brought up in much poorer family</a> than most of my peers and if I was 18 today, I wouldn’t have been able to afford the in-person schools either.</p><p><strong>The feeling of my younger self not being able to afford my current self’s startup idea frustrated me.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1Lp6vep6ZZrx9YxMidNQvA.png" /></figure><h3>The new idea</h3><p>Only two weeks ago, the combination of my subconscious and the ending of my non-compete forced me into my current thinking —<strong> digital education should be as affordable and as accessible as possible</strong>.</p><p>I’ve been scared to take on online learning. I feel the whole online education industry is fundamentally broken and unengaging. Previously I’ve felt it’s easier to engage students in person.</p><p>My thinking was “it’s broken, let’s do it in person instead”, rather than “it’s broken, <strong>can I fix it?</strong>”. As someone who’s taught a lot of people face-to-face, all the current online services make me cringe in one way or another, from how they teach to what they teach. It can be done better.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fG86qhtKUOydw9G1YTCHTg.png" /></figure><p>Starting today, I’ll be working on a new startup. A lot is still yet to be revealed and I will write all my plans along the way. I do have initial ideas about what I want to do and I’m very, <em>very</em> excited.</p><p>I’m inspired by various people and businesses who are open about the way they work — <a href="http://hearstartup.com/">Alex Blumberg’s Startup podcast</a> and <a href="http://blog.bufferapp.com/">everyone over at Buffer</a> particularly — so over the next few months, I’ll try to be as open as possible too. I’ll talk about branding, business models, raising money, hiring and more.</p><p>Every startup begins with a first step and today is the day I take it. Wish me luck!</p><p><em>Want to run your startup idea past me? Email me at rik@riklomas.com or tweet at </em><a href="http://www.twitter.com/riklomas"><em>@riklomas</em></a><em>. Follow me on </em><a href="http://www.medium.com/@riklomas"><em>Medium</em></a><em> or </em><a href="http://www.twitter.com/riklomas"><em>Twitter</em></a><em> to keep track of my story.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2b32742b914e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to start a startup without ruining your life]]></title>
            <link>https://medium.com/superhi/how-to-start-a-startup-without-ruining-your-life-af45eec09406?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/af45eec09406</guid>
            <category><![CDATA[long-reads]]></category>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Wed, 15 Oct 2014 10:30:40 GMT</pubDate>
            <atom:updated>2017-01-31T13:56:59.069Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*A5xTHHgRyXs355S3Gzzp5w.png" /></figure><p><strong>Note: This article is from 2014. It’s a little bit out of date now. There’s a updated version for 2017 over on the SuperHi blog here:</strong></p><p><a href="https://www.superhi.com/blog/how-to-start-a-startup-without-ruining-your-life">How to start a startup without ruining your life</a></p><h4>Startups are emotional roller coasters that one minute make you feel like you’re changing the world and the next minute, that everything is falling apart.</h4><p>For 6 years of my professional life, I worked only at startups. I saw myself as “the coder” — the guy that the brave hired to turn their ideas into real businesses. I never had the inclination to start my own… why would I risk a fairly comfortable life to pursue an idea that may not work?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*7kN90xGlUBawYYf5Qm3vjg.jpeg" /><figcaption>Me in full-time coder mode, reluctantely looking at someone else’s code</figcaption></figure><p>But then two years ago, I dived headfirst into running a startup — a code school in London — without really knowing what would happen. It was a fascinating, stressful and occasionally scary journey. In a few weeks time, I’ll be doing it all over again with a brand new startup.</p><p><strong>I am not a startup expert</strong> and I don’t think anyone is as everyone’s startup experiences are different. However, there are a lot of people who are new to both startups and the tech scene, and I’ve put this guide together for anyone who might be interested in dipping their toes into the startup water.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mGPPDb_PwRmfMiNmCCnEkA.png" /></figure><h3>There’s not always money in the banana stand.</h3><p>The tech industry is the only industry where you don’t need a business model. Imagine running a cake shop that gave away free samples but had nothing customers could pay for? Putting adverts on the wall wouldn’t save your bakery.</p><p>Tech businesses can get away with this as they have the one thing most industries don’t: scale. The internet gives tech businesses an opportunity to reach potential customers thousands of miles away from their base.</p><p>However this can be an illusion. If your business plan is based around having millions of users that don’t pay for things (or “eyeballs” as it’s called), there’s generally two things that may happen.</p><p>Number one is that your startup makes it big and everything will be okay. Number two is that you’ll slowly struggle and spend all your money. The first option is a lot less likely to happen than the second. For some startups, it might be worth pouring your money into buying lottery tickets — it’s a lot quicker, less painful and has more chance of making your money back.</p><p>There’s a balance to achieve with your market. Too niche and you’re in danger of having too few paying customers, and too broad and you might not be solving a particular problem. Watch <a href="http://www.ted.com/talks/simon_sinek_how_great_leaders_inspire_action">Simon Sinek’s TEDx talk</a> about starting with “why?” and think about your idea starting with why you’re doing it. Be passionate about your idea or at least be comfortable with the fact you’ll be spending <em>at least </em>a year with that idea. If you’re not comfortable being Mr Baby Product Shop or Ms Taxi Hiring Service, reconsider your idea.</p><h3>Don’t keep your idea to yourself.</h3><p>It’s tempting once you have your idea to keep it to yourself. Other people might steal it, but if you’re not confident in your idea <strong><em>or </em></strong>that you can assemble the best team to execute the idea, then maybe that idea isn’t right.</p><p>I don’t expect you to tweet your idea but if you’re making freelancers or advisors sign <a href="http://en.wikipedia.org/wiki/Non-disclosure_agreement">non-disclosure agreements</a> (NDAs) before you tell them your idea, you’re not confident in your idea.</p><p>When I was freelance, I refused to sign NDAs as it could possibly affect any future work that I may get.</p><p>Let’s say you NDA everyone so they can’t even talk about it and you launch without your idea leaking, what is to stop someone seeing your idea and making a better version? Google was not the first search engine and Facebook was not the first social network. Your startup needs to be the best-in-class, not the first to market.</p><p>There is one great piece of advice in Tim Ferriss’s <a href="http://www.amazon.com/The-4-Hour-Workweek-Anywhere-Expanded/dp/0307465357">Four Hour Work Week</a> which is to validate ideas by using Google Adwords. Set up a simple one-page site, check how many visitors it gets through Google Adwords and see how many of them sign up to a mailing list or click through to a fake buy page. It’s a quick, cheap way to see if people are interested in what you’re doing.</p><h3>The first idea probably won’t be your final idea.</h3><p>A lot of successful companies start out as different ideas. For example, Flickr was originally a photo-sharing chat room for gamers and on the extreme end of the scale, Nokia<strong> </strong>was originally a paper mill in Finland. My first startup was originally a careers advice website that mutated into a code school when we realised we couldn’t survive with the original business model. It’s common to change direction. Again, it comes down to what problem you’re trying to solve.</p><h3>Naming your baby.</h3><p>Naming things is hard — children, cats, dogs, plants — but naming a startup is particularly difficult. With web domains in ever shorter supply, the current process seems to be (1) think of a great name (2) check <a href="http://iwantmyname.com/">iwantmyname.com</a> (3) see that all the best domains are taken (4) go back to step 1.</p><p>Make sure that your customers can spell your startup within 3 tries (think <a href="http://en.wikipedia.org/wiki/Sensational_spelling#Music">Led Zeppelin or The Beatles</a>) and if possible, it’s worth getting the .com version at the start if it’s cheap enough as domain sellers may put the price up further once you become successful.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hnwLBc2KY0gvMRuxPpcisQ.png" /></figure><h3>Sketch everything.</h3><p>Once you have an idea, the next thing to do is sketch everything. If you’re making a site, draw all of the pages, and if you’re making an app, draw all the screens. Don’t skip any interaction — if you have a link or a button, what happens when you click it? Draw that too.</p><p>This process gets the idea out of your head and on to paper. It also helps anyone else working with you, be it co-founder, investor, designer or developer, understand what you’re creating. For co-founders, it means that you can come to a consensus about what the idea is. For investors, they can picture what they’re giving you money for. For designers and developers, sketching helps them picture what they need to and how long it will take.</p><h3>Find co-founders you trust.</h3><p>People are the important part of your startup. More than your idea. Have an uncommited or irresponsible co-founder and your startup may as well not exist. Good founders know how to pivot and change things when they’re not working. Ideas are worthless and execution is key.</p><p>You can do things as a solo founder but it’s harder. <a href="http://paulgraham.com/founders.html">Lots of investors don’t like investing in just one person</a> for various reasons but there are <a href="http://www.picfair.com/">examples</a> of going it alone. If you are considering it, make sure you get decent support on the areas you’re weaker on. For example, if you’re a businessperson, find design and tech advisors.</p><p>It’s always easier to work with your friends rather than finding someone to be your co-founder. You trust them, you know them and you have faith in their abilities. However, startups will strain even the most solid of friendships. Startups can be stressful and you may take your stress out on the nearest person.</p><p>If you and your co-founders do the same role — for example, you’re all businesspeople or all designers or all developers—then you have to make your roles <em>very </em>clearly defined in the business. If you don’t then you will fall out because one of you won’t want to do the shitty work that needs doing.</p><p>Co-founder fallouts are more common than most people realise. I left my first startup for that reason— it happens and you all just have to move on. There’s two very good articles on <a href="https://twitter.com/msuster">Mark Suster</a>’s blog<strong> </strong>Both Sides of the Table on <a href="http://www.bothsidesofthetable.com/2011/05/09/the-co-founder-mythology/"><em>The Co-Founder Mythology</em></a> and <a href="http://www.bothsidesofthetable.com/2014/01/04/the-perils-of-founder-fighting/"><em>The Perils of Founder Fighting</em></a><em>. </em>Read them both.</p><h3>How to find a tech co-founder.</h3><p>When I was freelance, I was asked weekly if I could join a startup as the technical co-founder. I generally didn’t know the person asking so I always said no.</p><p>Coders can be arrogant and they believe that they can build the whole idea on their own. Which is true… to a point. Coding the idea is only a tiny part of doing a startup, there’s design, sales, marketing, admin and plenty more. But at the moment, you need them more than they need you.</p><p>How do you differentiate yourself from other “<em>idea people</em>”? First is do a lot of homework. Do lots of research, draw sketches, get branding done, anything that a developer doesn’t do.</p><p>Secondly, learn some skills to sell yourself to developers. Learning the basics of design and/or coding, so that you know what you’re talking about. You wouldn’t start a cake shop if you didn’t know how to cook — same thing applies to starting a tech company.</p><h3>How to find investors.</h3><p>I said earlier than I’m not a startup expert: I’m definitely not an expert in investment. However, <a href="http://en.wikipedia.org/wiki/Paul_Graham_%28computer_programmer%29">Paul Graham</a> from <a href="http://en.wikipedia.org/wiki/Y_Combinator_%28company%29">Y Combinator</a> is and you should read all of his articles, but <a href="http://paulgraham.com/fr.html"><em>especially </em>this one</a>.</p><p>You need a “runway” — the amount of time you need from starting your startup to it being profitable. This is going to be an estimate as most startups never lift off the runway. Your runway probably will be at least twice as long as you first predict so raise money accordingly. It’s easier and time-saving to do one big raise than several small ones.</p><p>Raising money will also take a lot longer than you plan too. I’ve heard of startups taking a year to raise a round of money. Prepare for the worst.</p><p>The best investors are advisors with money. What sort of investors will depend on what you’re creating. If you’re creating a tech product in a particular area, you can look for tech investors and investors in that particular area too.</p><p>For early-stage startups, you will be looking to raise seed investment. Don’t bother trying the bigger companies like <a href="http://indexventures.com/">Index</a> and <a href="http://balderton.com/">Balderton</a> just yet. For instance, in London, investors like <a href="http://passioncapital.com/">Passion Capital</a> are great for early-stage investors (<em>disclaimer: I’ve freelanced at companies funded by Passion).</em></p><p>How much to raise again depends on your runway, your investors and also tax schemes. In the UK, look into <a href="http://www.seis.co.uk/">SEIS</a> and <a href="http://www.hmrc.gov.uk/eis/">EIS</a> as investors will probably want these tax breaks as incentives for themselves. A finger-in-the-air figure for an average early-stage raise would be between £150k and £300k as 3–5 employees plus operational costs can be expensive (see next section for more on this).</p><p>As always, talk to people who’ve done it. Make contacts at startups who can advise you and talk you through the stages.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Xm6_ftA0pHIqz90amUEzmg.png" /></figure><h3>There are many ways to create a cat.</h3><p>There are a lot of different ways to build your tech startup if you’re not technologically skilled enough to do it yourself. You can either hire full-time staff, you can hire freelancers or you can outsource it to an agency.</p><p>Without stating the obvious, on the <em>giving-a-shit scale, </em>agencies are going to give 0–3 shits out of 10, freelancers 3–8 shits out of 10 and full-time staff are going give 7–10 shits out of 10.</p><p>Agencies survive on getting things out of the door. You will be one of many companies in their queue and their aim is to get you out of the door as quickly as possible. Agencies are also going to be the most expensive out of the options.</p><p>I used to teach a 5 day long Ruby on Rails course. On the course’s last 2 days, I taught how to create a marketplace site where people could place ads and others pay for it. Think like AirBnb, Etsy and eBay. I anonymously contacted a few London agencies about how much they’d charge for all the features we do on the project. Three agencies replied and I received quotes between £35,000 and £60,000 (around $100k or €75k). Now, these figure were for a project I showed beginner coders how to make in 2 days. That version wasn’t all bells and whistles version but it wasn’t far off a <a href="http://en.wikipedia.org/wiki/Minimum_viable_product">minimum viable product</a>.</p><p>Once you’re out of the door with most agencies, that’s it, no more help unless you’re on an expensive retainer. I’ve heard of horror stories of agencies charging £10,000+ a month in retainer fees.</p><p>Freelancers are a way better option than agencies. They’ll be cheaper, care more and be a lot more receptive to you. Rough quotes for the project I mentioned earlier from freelancers were between £8,000 and £25,000 ($40k, €30k). They will also be more likely to have the flexibility to help you out longer term if you need them.</p><p>Last, and definitely not least, are full time staff. They’re the hardest to find out of the options but probably the cheapest in the long term. They’ll be a big part of your team and <em>really </em>care about what you’re making.</p><p>Junior designers in London will cost very roughly £18–30k, senior designers £30–50k, junior developers £20–35k, senior developers £30–60k (where junior means less than 5 years experience). Salaries vary widely on location and experience.</p><p>There’s also equity (shares in your company) to consider too. Most equity vests (gets given) over a couple of years — i.e. the staff member gets it based on how long they stay or if the company is sold. How much to give depends on a lot of things, how old the business is, how qualified that staff member is, how much money you’ve raised. There’s <a href="http://www.scribd.com/doc/55945011/An-Introduction-to-Stock-Options-for-the-Tech-Entrepreneur-or-Startup-Employee">a great guide to stock and equity in tech startups here</a>.</p><h3>Find good people.</h3><p>Why would I want to work at your <em>shitty </em>startup? Okay, I am being facetious but the point still stands.</p><p>When someone works for a startup, they’re giving up a lot of stability and possibly pay to work there. However, good people are not solely motivated by money, otherwise everyone would be working at banks and management consultants. Good people want to be part of a cause and a belief that what they’re doing will make an impact. It’s your job to convince them that your startup is going to make that impact.</p><p>The best way to find people is as always by networking. Ask friends or make friends in the tech scene by going to events. For instance, in London there are events like <a href="http://silicondrinkabout.com/london">Silicon Drinkabout</a>. You may not find your staff there but you may find someone who knows others looking for opportunities.</p><p>Another way is to advertise. Sites like <a href="http://techjobsin.london/">techjobsin.london</a> <em>(disclaimer: I created this site)</em>, <a href="http://workinstartups.com/">WorkInStartups</a>, <a href="http://www.hackerjobs.co.uk/">Hacker Jobs</a>, <a href="http://jobs.smashingmagazine.com/">Smashing Magazine Jobs</a> and <a href="http://www.authenticjobs.com/">Authentic Jobs</a> are great places to find good staff.</p><p>Startup job fairs can be great — I met a previous employee at <a href="http://siliconmilkroundabout.com/">Silicon Milkroundabout</a> — but if you do exhibit, make a <em>big </em>effort. Don’t just turn up with <a href="https://www.google.co.uk/search?q=banner+stands&amp;source=lnms&amp;tbm=isch&amp;sa=X&amp;ei=VRMyU9WvK8mqhQfriYGABQ&amp;ved=0CAcQ_AUoAg&amp;biw=1251&amp;bih=822&amp;dpr=2">those awful banner stands</a>. Stand out to get attention away from every other startup there. Be honest with the attendees though, they will appreciate it if you say that you don’t know what you’re doing over pretending you do and finding out later.</p><h3>Design is as important as development.</h3><p>A lot of soon-to-be founders ignore a big part of startup process — design. If you look at any popular site or app, you will notice how much effort has gone into not only the branding of the site but the whole user experience.</p><p>If you’re not planning to hire a designer, I would definitely suggest getting a freelancer to create a brand identity for you. Having a strong brand is essential to make it successful in a crowded market. There are only a couple of tech companies I can think of that don’t have strong brands — <a href="http://www.craigslist.org/">Craigslist</a> probably being the main one. Everything else has a memorable brand.</p><p>I would advise against using something cheap like <a href="http://www.99designs.com/">99designs</a>. Pay peanuts, get badly-kerned logo. Suggest trying a skill-swap with a designer if you can’t afford them.</p><p>Next, read up on user experience. Read <a href="http://www.amazon.co.uk/Dont-Make-Me-Think-Usability/dp/0321344758"><em>Don’t Make Me Think</em> by Steve Krug</a>, browse <a href="http://littlebigdetails.com/">Little Big Details</a> and scrutinise everything about your site. Make sure everything works as simply and flawlessly as possible for your users. Use services like <a href="http://www.usertesting.com/">UserTesting</a> to check your prototype when you have one.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bRoNbqqOdoxGHmwQ0uFCnA.png" /><figcaption><a href="https://twitter.com/paulg/status/522069193345093632">https://twitter.com/paulg/status/522069193345093632</a></figcaption></figure><h3>Simplify.</h3><p>First versions of products always have a tendency to be too complicated. Be very wary of adding more and more features. Keep it as minimal as possible. Read (or re-read)<em> </em><a href="http://www.amazon.co.uk/The-Lean-Startup-Innovation-Successful/dp/0670921602"><em>The Lean Startup</em> by Eric Ries</a>.</p><p>As an example, at my previous startup, we launched with 7 different courses, including 4 that competed against each other. We quickly learnt that customers were getting confused between the difference so trimmed it down to just 2 courses which sold far better than having 7. Keeping it simple works. In the words of Gordon Ramsey, “<a href="http://www.channel4.com/programmes/ramsays-kitchen-nightmares/articles/all/gordon-ramsays-top-tips-for-starting-a-restaurant">the more dishes, the lower the standard</a>” (the linked article has great information, even for tech companies).</p><p>If you are tempted to add features (and you will get tempted), do it one by one. It’s a lot easier to test if that feature works if you change one thing at a time rather than releasing several at the same time.</p><p>Use services like <a href="http://www.usertesting.com/">UserTesting</a>, <a href="https://www.optimizely.com/">Optimizely</a>, <a href="http://www.olark.com/">Olark</a>, <a href="https://www.intercom.io/">Intercom</a>, <a href="http://gosquared.com/">GoSquared</a>, <a href="https://mixpanel.com/">Mixpanel</a> and <a href="https://www.kissmetrics.com/">KissMetrics</a> to test, talk and analyse your customers. The more data you have, the more accurate you can be in your decision making.</p><h3>Sales and marketing are key.</h3><p>Unless you or a co-founder is well known in your industry, your launch will be a quiet one. The opposite of “build it and they will come”. The next step is to get your startup’s name out to your audience.</p><p>Read the great <a href="http://www.danpink.com/books/to-sell-is-human/"><em>To Sell is Human</em> by Dan Pink</a>. Look into <a href="http://en.wikipedia.org/wiki/Customer_experience">customer experience</a> and why it matters too. <a href="http://www.amazon.co.uk/The-Sense-Style-Thinking-Persons/dp/1846145503"><em>The Sense of Style</em> by Steven Pinker</a> is another book I recommend highly. It isn’t much to do with sales or marketing directly but it will help you write better content for sales and marketing.</p><p>Depending on your startup and its industry, different techniques for sales and marketing will and won’t work for you:</p><ul><li><strong>Partnerships</strong> with companies whose customers would benefit from your startup. For example, if you’re a e-commerce site for new parents, give discounts to bloggers who write about parenting.</li><li><strong>Internet advertising</strong> such as Google Adwords, Facebook ads and Twitter ads. It’s relatively cheap and easy to see what’s working and what isn’t.</li><li><strong>Old-media advertising</strong> such as newspapers and radio. This will only work if you want to focus on a particular town or city to begin with. It’s also great for brand awareness but may not convert into customers.</li><li><strong>PR campaigns</strong>. I’ve seen many startups make the mistake of using PR campaigns to promote the founders rather than the business itself. Watch that ego!</li><li><strong>Recommendations</strong> are the best form of advertising. If people love what you’re doing, they will tell other people. Focus heavily on customer experience and it will improve the likelihood of them promoting you. You have to make every customer buy in to what you do and be your biggest fan.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Qp1fFJuh6YB2x6d7ea1W6g.png" /></figure><h3>Startups are like ducks</h3><p>My mum works as a counsellor and she always says is that no matter how calm and relaxed a person might seem on the outside, underneath there’s a likelihood that they’re a panicked mess. The duck may seem like it’s sailing smoothly across the water but underneath its little feet are thrashing away wildly, trying to go in a particular direction.</p><p>You may think that you’re the only business in the world where everything is uncontrolled and falling apart. Welcome to startup life. These feelings are <strong>completely normal</strong>. Every other startup founder feels like this too. How you deal with it to the outside world is what matters.</p><p>The best companies make it look completely stress-free when they’re not. Try reading <a href="http://www.amazon.co.uk/Hatching-Twitter-Nick-Bilton/dp/1444767046"><em>Hatching Twitter</em> by Nick Bilton</a> for proof that the most successful startups can still be full of in-fighting and dysfunction.</p><p>A great podcast on starting a new business is <a href="http://hearstartup.com/"><em>Startup</em></a> by Alex Blumberg — it’s something that I look forward to each week and it’s recommended listening for anyone wanting to start a business.</p><blockquote>“Launching a startup is like taking the brakes off your car when setting off on a long journey”</blockquote><h3>The future.</h3><p>The launch is just the start. Keep going. Keep iterating. Try out features. Change the design. Talk to your customers. Enjoy the journey.</p><p>Any founder will tell you, startups aren’t meant to be fun. It’s hard, stressful work and you’ll hit a lot of obstacles along the way. One day you might want to pack it all in, the next you’ll totally love it. You might become rich and famous or you might crawl out of the rubble of a failed startup.</p><p>Don’t be scared of failure. It’s better to try and fail than never try at all. Many successful founders have had companies shut down before getting the formula right. Don’t think it as failure, <a href="http://thinkexist.com/quotation/i_haven-t_failed-i-ve_found-ways_that_don-t/346094.html">it’s just a way that didn’t work</a>.</p><p>This is your opportunity to make a dent in the world. It could be in a tiny way or it could be in a huge way, but with a startup, you have a potential to change the way that people live their lives.</p><p><em>Want to run your startup idea past me? Email me at rik@riklomas.com or tweet at </em><a href="http://www.twitter.com/riklomas"><em>@riklomas</em></a><em>.</em></p><p><em>My new startup is called SuperHi! Sign up for it at </em><a href="http://www.superhi.com/"><em>http://www.superhi.com</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=af45eec09406" width="1" height="1" alt=""><hr><p><a href="https://medium.com/superhi/how-to-start-a-startup-without-ruining-your-life-af45eec09406">How to start a startup without ruining your life</a> was originally published in <a href="https://medium.com/superhi">SuperHi</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to create a simple jobs board in Ruby on Rails — even if you’ve never coded before]]></title>
            <link>https://medium.com/@riklomas/how-to-create-a-simple-jobs-board-in-ruby-on-rails-even-if-youve-never-coded-before-9b296c4df483?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/9b296c4df483</guid>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Fri, 10 Oct 2014 09:55:13 GMT</pubDate>
            <atom:updated>2016-05-25T20:10:30.846Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*efCN9PTffpp4jh6uaCnRMg.png" /></figure><h4>This tutorial is for programming beginners to show how easy it can be to get an interactive site up and running.</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/939/1*ynKKlM32zrt_q55xmHqb9g.png" /></figure><p>Yesterday, I launched a new site that I’ve been thinking about creating for a while.</p><p>I keep retweeting job adverts from <a href="https://twitter.com/riklomas">friends of mine</a> and thought it would be great if I could keep them in one place for other people to browse without having to see the rest of my tweets. Not everyone is into cat videos as much as I am. Unfortunately.</p><p>I bought the domain <a href="http://www.techjobsin.london/">techjobsin.london</a> on Wednesday morning and started coding in the afternoon. It was a quick site to build — around <em>two hours</em> from starting to getting it online — which surprised a few people. Was the quick speed of completing the site because I’ve been doing coding for years? Yes and no.</p><p>Programming languages power all the internet yet they’re not understood by 99.9% of internet users. It’s hard to understand what is possible — and how long it takes — if you don’t understand the tools used to create the internet.</p><p>I’ve taught a few hundred people how to code and one trait I see again and again is that students suddenly realise what is possible and how long it takes once they have a little bit of knowledge.</p><p>But how do we know which one of the hundreds of programming languages to choose? PHP? Python? Node? It’s down to picking a good tool for the job of creating the web. Ruby on Rails is my particular favourite way to code web sites. It’s a beautiful, friendly and quick way to make dynamic websites.</p><p>This Ruby on Rails tutorial is for people who’ve never programmed before. I’ve tried to go as slow as possible. If you can install applications on your computer, you should be able to complete this tutorial.</p><p>Developers have tendencies to get pedantic (read: pissy) about some of the ways taught below. The following is done for the sake of human clarity rather than computer clarity.</p><h3>Installing Ruby on Rails</h3><p>If you’re on a Mac and have Mavericks (10.9) or Yosemite (10.10), you’re fine for now. If you’re on a Windows computer or a Mac with 10.6, 10.7 or 10.8, the easiest way to install Rails is to use <a href="http://railsinstaller.org/en">RailsInstaller</a>.</p><p>RailsInstaller will install Ruby 1.9.3, Rails 3.2 plus a ton of other little, useful tools. However, we want to be using the very latest version of Rails which as of October 2014 is Rails 4.1.</p><p>Rails 4.1 introduces lots of nice things that we can use when we’re building bigger sites. We won’t be using most of the new features but as we’re starting a brand new site, we might as well use the very latest version of the software. We’ll talk about how to upgrade to the latest version of Rails in a minute.</p><h3>The command line</h3><p>As with any programming, we’re making <strong>custom software</strong> that’s never been created before so we have to use tools that help us create custom software.</p><p>The first tool we use heavily in coding is the command line. This is a tool that lets you get under the hood of your operating system.</p><p>You’ve probably seen a command line in movies. When the nerdy one tries to break into the bank’s mainframe, they’ll be using a black screen with green text. That’s a command line. Imagine hacking bank’s mainframe using Excel or iTunes — it would be impossible. You need powerful tools to do powerful work. This is why we use the command line.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/972/1*Rhnbb7dDAPbV4UTlRldtXg.png" /><figcaption>Don’t worry if your command line doesn’t look exactly like this. As long as you can type into the box, it’s cool.</figcaption></figure><p>There are tools built into your operating system that let you play around with the command line. On Windows, it’s called <a href="https://www.google.co.uk/search?q=Command+Prompt&amp;safe=off&amp;espv=2&amp;biw=1280&amp;bih=925&amp;source=lnms&amp;tbm=isch&amp;sa=X&amp;ei=DV02VMrkDrG07Qaw4oGgCw&amp;ved=0CAYQ_AUoAQ">Command Prompt</a> which can be found under the Start menu, and on a Mac, it’s called <a href="https://www.google.co.uk/search?q=terminal+mac&amp;safe=off&amp;espv=2&amp;biw=1280&amp;bih=925&amp;source=lnms&amp;tbm=isch&amp;sa=X&amp;ei=Il02VL7XBYjd7Qb8yICYDQ&amp;ved=0CAYQ_AUoAQ">Terminal</a> which can be found in your Applications folder in the subfolder Utilities. Open it and you now have the same power as a movie’s hacker.</p><p>Most “commands” — the things we type into the command line — are pretty boring most of the time. Commands like change folder, make me a new file, list the files in a folder. It’s not always folder-related but 80% of the time, it will be basic.</p><p>If you’re on a Mac, it’s quite easy to try out the command line. Open Terminal, type into your command line the following and press enter:</p><pre>say &quot;I am a hacker&quot;</pre><p>You should hear the operating system saying those words. “Say” is a command and we’re telling it to say a particular phrase.</p><p>In coding, we call phrases like this <strong><em>strings</em></strong>. We’re sending a string to the command “say” to make the computer do something — in this case talk to us.</p><h3>Updating the latest version</h3><p>We’ll be using the command line quite heavily from here so open it ready to go.</p><p>You may or may not already have Rails installed. To see if you do, you can type in and press enter:</p><pre>rails --version</pre><p>If you do, it will say something like “Rails 3.2.15&quot; or “Rails 4.1.6&quot;. If not, you will get an error.</p><p>If you don’t see an error, to update to the latest version of Rails, type in and press enter.</p><pre>gem update rails</pre><p>If you do see an error, you will need to install Rails. Type the following and press enter.</p><pre>gem install rails</pre><p>You should see things happening. If you get a permissions error, you may need to add the word “sudo” to the start of either command and it will ask for your computer’s password.</p><pre>sudo gem install rails</pre><p>You should have the latest version of Rails set up on your computer now.</p><h3>Where to put our project</h3><p>The next step is we need to decide where to put our project. By default, the command line will drop us into what’s called the “home directory”. This is where your user account on the computer is.</p><p>Personally, I like to make a folder to keep all my projects in. I usually use the folder name “Sites”, but it’s up to you. Let’s create a folder to store any project that we make.</p><pre>mkdir Sites</pre><p>What is this ugly “mkdir”? We can break it down into two parts — “mk” and “dir”. First part is “make” and the second is “directory” (another way to say folder). So we’re telling the computer to make a directory called Sites.</p><p>Typing that command and pressing enter is the hacker way to create folders. We could have just found a place to right-click in Finder and pressed “New Folder” but this way makes you feel more like you’re in a movie, right?</p><p>We only ever need to do this once. Once we have the folder, it’s there until we delete it. We don’t need to do this again when we make a new project.</p><p>Next, we’ve made a folder but we’re not inside that folder. We’re still in our “home directory”. Let’s go inside the folder by “changing directory” (or cd for short).</p><pre>cd Sites</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/217/1*aApMiHDO2r0cTy2d9fnvHA.png" /></figure><p>You’ll see the bit next to the flashing cursor change to say something like Sites near the end. That means that we’re now inside the Sites folder. Once we’re in here, we can create our first project.</p><h3>Creating our project</h3><blockquote>There are only two hard things in Computer Science: cache invalidation and naming things.</blockquote><p>If I had a dollar for every second I’ve spent thinking what names to give to projects, I’d be able to afford some decent domain names.</p><p>I was lucky to get the domain <a href="http://techjobsin.london/">techjobsin.london</a>, it was still available, so it was obvious what I should call the project. I’ve called that project on my computer the very exciting name of “techjobs”.</p><p>You can call your Rails project whatever you like, just as long as it doesn’t have any spaces in the name and it’s all lower case, e.g. no to “Jobs Board” but yes to “jobsboard”. Your users won’t see this at all, you can brand your site however you like, this is just for your reference. Let’s stick with jobsboard for now. To create your project, type in the following and press enter.</p><pre>rails new jobsboard</pre><p>A lot of things will happen now. Rails will create all the files and folders you need for your project, plus download some gems. Gems are plug-ins for making your life easier and we’ll talk about them later. This process may take a minute or two depending on your internet connection.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*rCMA8wotziWxD3Ew21XomA.png" /></figure><p>One last command before we start. We need to go into the project itself, otherwise we’ll just say in our parent projects folder.</p><pre>cd jobsboard</pre><h3>One more thing before we start coding</h3><p>Rails has just generated a lot of files and folders for us. This is the skeleton for our jobs board but we need something to write our code in.</p><p>Code editors sound scary but just think of them like Microsoft Word. Instead of writing an essay for a human, you’re writing an essay for a computer to understand. The biggest the story, the more you need to write.</p><p>There are lots of good code editors out there. My personal choice is <a href="https://atom.io/">Atom</a>, but a great, free alternative is <a href="http://www.sublimetext.com/3">Sublime Text 3</a>. Download one of them and install it to your computer.</p><p>Open it once installed and you may get something that looks like this.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kjZGl4H7HX48I8HzaQ-12Q.png" /><figcaption>Oh hi, Atom.</figcaption></figure><p>You may get a welcome screen or you may get a blank screen. For now, we just want to work on our jobs board project. Go to File &gt; Open and find your project.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/578/1*0zRsg_GvHrE_S4ySRAeQkw.png" /></figure><p>If you can’t find your project, you can go back to the command line and type in:</p><p>Mac:</p><pre>open .</pre><p>Win:</p><pre>start .</pre><p>This will open the folder in your Finder/Explorer. You can also drag the folder to your dock icon which will open the folder. Once you’ve opened your project it should look something like the following.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/998/1*rYH_z6ee-cHTNXzKS27PxA.png" /><figcaption>Mmmm, folder sidebar time.</figcaption></figure><p>We’re now ready to do some coding. Or are we?</p><h3><em>Proper Planning and Preparation Prevents Piss Poor Performance</em></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*t6uWu33Yl-Ij5VSv5grc2A.jpeg" /></figure><p>The most important part of doing any project is the planning. Let’s stop and think for a minute what we want to do on this site.</p><p>There will be a home page which will be a list of jobs, and there will be an area to add jobs (let’s leave it open to the public at the moment — we could add in admin only as a later feature).</p><p>There’s some more questions we need to answer here though. What if there’s 300 jobs… should there be pagination? What information should we take for people submitting jobs?</p><p>The best way to plan anything out is to sketch, sketch and more sketch. Invest in some good paper and pens (the hipster in me prefers Moleskine notebooks and Paper Mate Flair pens).</p><p>Here’s a sketch of how would do the project. Again, two pages needed, not that detailed, but I can see what I need before I start designing or coding the web site.</p><h3>Gems</h3><p>Once, I’ve sketched every page out, I know what I need for my project. I can see that I will need some pagination due to the left and right arrows at the bottom of my home page.</p><p>I could code the pagination myself but pagination is a fairly common thing on web sites. There may be another Rails developer in the world who has written some pagination code that’s better than I could write.</p><p>In Rails, you can download little snippets of code that you can use in your project — these are called <strong>gems</strong>. There are large and small gems for lots of different purposes. From making forms easier to whole CMS platforms.</p><p>I’ve previously <a href="https://medium.com/@riklomas/my-favourite-ruby-gems-services-89fb47341c05">written a Medium article on the best gems</a> covers a lot of the common functionality that you get on websites. We’re going to use two gems in our project. <a href="https://github.com/amatsuda/kaminari">Kaminari</a> (a pagination gem) and <a href="https://github.com/plataformatec/simple_form">Simple Form</a> (a gem to make coding our new job form easier).</p><p>To add the gems to our project, we need to go to our code editor and edit the file called Gemfile.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/999/1*b_Rrb090ZJtm_Lyc1sQjNQ.png" /></figure><p>Open the file, go right to the bottom and add in the following:</p><pre>gem &#39;simple_form&#39;<br>gem &#39;kaminari&#39;</pre><p>Remember to save your changes!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R9zdg251bdC-E7WgX9wa-A.png" /></figure><p>Your gems are ready to be installed. To download them to your project, go back to your command line and type in:</p><pre>bundle install</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/737/1*1BHP0_hXqPH0mbitjnyHVg.png" /><figcaption>Your bundle is complete!</figcaption></figure><p>If we want to add any other gems to our project later, just edit your Gemfile again and run the <em>bundle install </em>command.</p><h3>Models, views and controllers</h3><p>Okay, we’re now going to stop for a minute to talk about Rails terminology. We’re going to separate our site into three different parts; models, views and controllers.</p><p>The first is called the <strong>views</strong>. For our project, this means the HTML for the site. For instance, we know that we’ll have some paragraphs so we code them in &lt;p&gt; tags. We might want to link somewhere, so we have &lt;a&gt; tags. If you don’t know much about HTML, <a href="http://htmlandcssbook.com/">Jon Duckett’s HTML and CSS Book</a> is truly amazing. Buy it.</p><p>The second is called the <strong>models</strong>. Think of models as if they were sheets in an Excel (or Numbers) spreadsheet.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-QmSVsRwYpxQDfxQCltOYw.png" /><figcaption>Models as if they were spreadsheets. Notice the tabs at the top signifying three different models; Jobs, Users and Bookmarks</figcaption></figure><p>In my spreadsheet, I have a jobs sheet, a users sheet and a bookmarks sheet — I have three different types of data <em>modelled</em> in my spreadsheet.</p><p>In my jobs model, I have four columns; a job number, a title, a company and a URL, and there’s three job entries in my sheet.</p><p>The third part is called the <strong>controllers</strong>. This is Rails’s way for everything to talk together. For example, when we’re on the homepage, we want to get all the jobs from our spreadsheet and put into our index.html file in a particular way. When we’re on our add-a-job page, we want to make a new row in our spreadsheet and we want to show the user a form to complete.</p><h3>Feature process</h3><p>What counts as a feature? On our site, it would be the ability to add and show jobs to the site. Another later feature could be the ability to submit your CV to a particular job. Another could be register as a user, or bookmarking jobs.</p><p>There’s a process I use whenever I add a new feature:</p><ol><li>Generate any models needed for the feature</li><li>Update the database</li><li>Generate any controllers needed</li><li>Add in your URLs</li><li>Fill in your models</li><li>Fill in your controllers and views</li><li>Finish feature, think of new feature and start from step 1 again</li></ol><h3>Lets start!</h3><p>First thing in my feature process list is generate any models needed. Earlier we saw that in our spreadsheet for jobs we had four columns in there; job number, title, company and URL. We can ignore the job number as Rails handles this for us.</p><p>Rails has a few conventions — one of them is that the models are always title case and singular. For us, we’re going to call our model “Job”. Other models we could have include “User”, “Bookmark”, “Comment” or “Company”, but not “jobs”, “users”, etc.</p><p>Our first model is a Job model and it has three columns (ignoring the job number column) which are title, company and URL. We’re going to store them all as phrases or as we call them in computing, <em>strings</em>.</p><p>We can store data in lots of different formats such as numbers, true or false (called “booleans”), and dates. For what we need, we’re keeping with strings.</p><p>In our command line, lets type in:</p><pre>rails generate model Job title:string company:string url:string</pre><p>Lets break this down. We are generating a Rails model called Job with three columns which are strings. Make sure you get all the spaces correct.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/716/1*TG4mTf-_AF6dgtB6NoMSbQ.png" /></figure><p>This will create a few files that we need for our site which we’ll cover in the next step and in step 5.</p><h3>Step 2 — your database is not your site</h3><p>Unfortunately, when you make a model, it doesn’t automatically go into your database.</p><p>The database is the place you store data. This doesn’t live within Ruby or Rails but in a storage unit. We’ll use the default storage which is called SQLite3 and part of Rails by default, but we could use other ones such as MySQL or PostgreSQL. Don’t worry about this too much as Rails handles 95% of data handling for you. The only bit is doesn’t handle is updates to our database such as adding a new model or updating a model.</p><p>So let’s update our database with our brand spanking new Job model. In our command line, type in:</p><pre>rake db:migrate</pre><p>Watch out for spaces and colons here.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/715/1*Y7t_X5DCAMXeTWvHM3lVbg.png" /></figure><p>You should see something like “CreateJobs: migrated” towards the end when you press enter.</p><p>Now you might have noticed that this starts with “rake” not “rails”. There’s a technical reason for this but to keep it simple, rake is a way to run particular ruby commands on your command line. This is updating (or migrating) our database to the very latest version that it needs to be.</p><h3>Step 3 — add your controller and views</h3><p>Even though we have just one job model, we have two different pages on the site. We have a home page (or <em>index</em> page) and an add page (or <em>new</em> page).</p><p>Rails wants to be able to control which jobs you’re getting on which page. Our home page will behave differently to the new job page. This is where our controller comes in.</p><p>For this section of the site, we have a jobs area. On bigger sites, we will have multiple areas of the site with multiple controllers. For instance, we might want a résumé submission area of the site in the future — this would come under a different controller, one that deals solely with résumés. Other controllers could include a payments controller (if we decide to charge people) and a comments controller (if we decide to let people add comments to jobs).</p><p>We only need one controller at this point — solely dealing with the listing and addition of jobs. Just like with models, Rails makes controllers conform to conventions. Controllers are lower case and plural — so for this it would be “jobs”, “users”, “bookmarks” or “photos”.</p><p>We also want to make two pages to go along with this controller, the home page (in Rails, we call this the <em>index</em> page) and the add page (in Rails, we call this the <em>new</em> page).</p><p>Lets go back to our command line and type in:</p><pre>rails generate controller jobs index new</pre><p>Breaking that down, Rails is generating a controller called jobs and we’re also creating index and new pages with that controller.</p><p>Awesome. That’s all we need to generate for now.</p><h3>Step 4— Adding in some URLs</h3><p>URLs — or web addresses — are the biggest part of the web. We know that if we type in <a href="http://www.twitter.com/">twitter.com</a> into Chrome’s address bar, we’ll go to Twitter’s home page. If we go to <a href="http://www.twitter.com/riklomas">twitter.com/riklomas</a>, we’ll still go to Twitter but a particular page on Twitter — my profile page. The difference between the two links, and all links in fact, are key to how the web works.</p><p>For our site, we want to add in some URLs for going to different page. We have two particular pages that we want to go to. The home page and the add a job page.</p><p>To add these URLs we need to go back to our code editor. We need to find the file inside the <strong>config</strong> folder called <strong>routes.rb</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1002/1*96_pyHCbtuWAMKHkLbKQvQ.png" /></figure><p>This is where we can add our URLs. Keep the first line in tact and on the second line add the following.</p><pre>resources :jobs<br>root &#39;jobs#index&#39;</pre><p>You can even delete all the lines with “#” and with “get” in front of them. Keeping it super simple.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/683/1*A8S5mdS03ls-yuEaLSL3Dw.png" /></figure><p>Save your file! There’s been a million times when I’ve forgotten to save files and code hasn’t worked properly.</p><p>We have some URLs we can go but our project’s code isn’t running. Lets go back to the command line and run our code. Type in:</p><pre>rails server</pre><p>You should get something like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/881/1*EnL1tQp3NJtek9UqDhaP4g.png" /></figure><p>We have a website! If we go to <a href="http://0.0.0.0:3000/">http://0.0.0.0:3000</a> or <a href="http://localhost:3000/">http://localhost:3000</a> in our browser, our code will run. It’s not much, but it’s something!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/704/1*u3jzYk6tt_Gfa5Vy9uNHpw.png" /></figure><p>Don’t worry about the content just yet. We’ll start filling that out in step 6.</p><h3>Step 5 — Feeding our models</h3><p>We want to make sure the data that comes through from our form. At the moment we could leave everything completely blank in the form in our sketches and that would be A-OK.</p><p>We want to make sure that the job <em>definitely</em> has a title, a company and a URL, otherwise the site would be full of blank jobs.</p><p>In our code editor, lets go to our model folder. You can find this within the app folder.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/329/1*Rh5Bk6AFunMLOROgDYLxLQ.png" /></figure><p>Notice in here we have a file called <em>job.rb.</em> This is the model that we generated earlier, let’s open this file and change:</p><pre>class Job &lt; ActiveRecord::Base<br>end</pre><p>to:</p><pre>class Job &lt; ActiveRecord::Base<br>  validates :title, presence: true<br>  validates :company, presence: true<br>  validates :url, presence: true, uniqueness: true<br>end</pre><p>What we’re doing here is making sure that our job’s title has some presence. In other words, making sure it isn’t blank. Same thing with our job’s company row. For our job’s URL, we’re making sure that it isn’t blank <em>and </em>it is unique as we don’t want any duplicate job posts.</p><p>These are called Rails validations and <a href="http://guides.rubyonrails.org/active_record_validations.html">there’s a great list of them here</a>.</p><p>Save the file and let’s go on to step 6 — the biggest step.</p><h3>Step 6 — filling out our site</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/467/1*qxAWgWSbvVzILkjkanJ4Aw.png" /></figure><p>At the moment, there aren’t any jobs on our site. Our spreadsheet is empty. So let’s start with our new job page first. Go to <a href="http://0.0.0.0:3000/jobs/new">http://0.0.0.0:3000/jobs/new</a> or <a href="http://localhost:3000/jobs/new">http://localhost:3000/jobs/new</a> and you should see the HTML page on the left.</p><p>First of all, we need to create a new line in our database. To do this lets go to our jobs controller. Back to our code editor, go to the <strong>app</strong> folder, then the <strong>controllers</strong> folder and open <strong>jobs_controller.rb.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/829/1*9oCj0s-A2b-SAaFBzyYwyw.png" /></figure><p>We want to change the new page, so the code that we’ll add will be in between lines 5 and 6 — the “def new” and the “end”.</p><p>“def” is just short for define — so when we define the new page, we want to say something about it. We want to create a new row in our job model.</p><p>Change</p><pre>def new<br>end</pre><p>to</p><pre>def new<br>  @job = Job.new<br>end</pre><p>Job.new means “hey, Job model, can we make a new one of you?”. But what about the first bit “@job”? This means send the new job we’ve created to our HTML so we can use it.</p><p>Think of the controller like a boxing match referee. In the red corner, we have the models and in the blue corner, we have the views. The controller is making sure both play by the rules that we decide.</p><p>Next thing to do is edit our HTML on our add a job page. Let’s go to the <strong>app</strong> folder, then <strong>views</strong> folder, then <strong>jobs</strong> folder and find <strong>new.html.erb.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/848/1*-EOc9RStruwF2ROae8U6Vw.png" /></figure><p>We want to put a form in place of the dummy text. Replace all the text in this file with:</p><pre>&lt;%= simple_form_for @job do |form| %&gt;<br>  &lt;%= form.input :title, label: &quot;Job title&quot; %&gt;<br>  &lt;%= form.input :company, label: &quot;Your company&quot; %&gt;<br>  &lt;%= form.input :url, label: &quot;Link to job&quot; %&gt;<br>  &lt;%= form.button :submit %&gt;<br>&lt;% end %&gt;</pre><p>Remember that Simple Form gem we added earlier? This is where it comes in handy! Rails’s default form tags are a pain in the arse and Simple Form lives up to its name by making it a ton easier.</p><p>What this is doing is making a form for the new job we made in our controller with the columns in a particular order that we want. Save the file and refresh the page and you’ll get something like:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/568/1*AbEVK5rGEF3NnAfJFfn4cA.png" /><figcaption>Simple Form even puts in asterisks for required fields. It knows this due to the presence: true we added to our models</figcaption></figure><p>Try the form out. Fill it in with some details and submit the form. We should get an error. We’ve filled in the form but we can’t do anything with that form data yet.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/666/1*36W6JrWzu_s2ZDylPtoyUQ.png" /></figure><p>So what’s happened? We’ve submitted the form but the form has nowhere to go.</p><p>Again, Rails has conventions. Whenever we have a <em>new</em> page, we submit the form and it goes to a <em>create</em> page. The create page is where the form data gets checked against the validations and if it passes, saves into the database. If it doesn’t pass, we want to show the form again with errors.</p><p>We actually have three pages on our site: <strong>index</strong> (home page), <strong>new</strong> (form page) and <strong>create</strong> (actually add the data from the form page).</p><p>Let’s go back to our jobs controller and add this create page to it. I’ve added in the following to our controller:</p><pre>def create<br>end</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/815/1*ypkUnSKjkYf9S0oxaLf1pA.png" /></figure><p>Next, we want to make a new job. Let’s change the create page to do that.</p><pre>def create<br>  @job = Job.new<br>  if @job.save<br>    redirect_to root_path<br>  else<br>    render &quot;new&quot;<br>  end<br>end</pre><p>Phew, that’s a big change. So we’re creating a new job from our model, then if it saves, go to the home page (Rails likes to call this the “root path”), if it doesn’t save, render (or plainly, show the user) the HTML view called new.</p><p>If we try this now, it still won’t work because we haven’t put in our form data. We want to make sure that this data we send from the form is nice and secure. Remember the movie hackers from earlier? They really exist and we want to make sure that they can’t do bad things to our site such as making themselves administrators or stealing credit card details.</p><p>Let’s change one line in our create method to add this form data in. Change line 10 from:</p><pre>@job = Job.new</pre><p>to</p><pre>@job = Job.new(params.require(:job).permit(:title, :company, :url))</pre><p>Phew, that’s another big one. What we’re saying here is make a new job from the form data <strong>but </strong>only allow jobs to be submitted with title, company and url data. Our controller should look like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*c_fZqYRKmKpOa44uABDOCg.png" /></figure><p>Now when we add in a new job, it should go back to the home page. That’s great but the last problem is that we can’t see any jobs on the home page!</p><h3>The home page</h3><p>Remember that Rails calls our home page the <em>index</em> page. Think of an index as being a list and in this case, our home page is a list of jobs. In our controller we want to get all our jobs in our database to show our users.</p><p>Let’s change our index page in our jobs controller from:</p><pre>def index<br>end</pre><p>to:</p><pre>def index<br>  @jobs = Job.all<br>end</pre><p>Here we have the home page getting all the jobs from the Job model and sending it to the HTML view as @jobs (we could name @jobs anything we want by the way, @all_jobs for example).</p><p>Our jobs controller is now finished:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HOKcURzsaXHFZwb9bycBJw.png" /></figure><p>Last thing we need to do is add those jobs to our home page’s HTML. Let’s go to our view for the jobs index page. This is in the <strong>app</strong> folder, in the <strong>views</strong> folder, in the <strong>jobs</strong> folder then open <strong>index.html.erb</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/838/1*_DMP6BShSuHIw5qXmBIHHQ.png" /></figure><p>We added a job earlier but there will be multiple jobs in here so we want to loop over each job and do some HTML with it. Let’s change all the text in index.html.erb to:</p><pre>&lt;h1&gt;Jobs&lt;/h1&gt;</pre><pre>&lt;% @jobs.each do |job| %&gt;<br>  &lt;div class=&quot;job&quot;&gt;<br>    &lt;h2&gt;&lt;%= link_to job.title, job.url %&gt;&lt;/h2&gt;<br>    &lt;p&gt;&lt;%= job.company %&gt;&lt;/p&gt;<br>  &lt;/div&gt;<br>&lt;% end %&gt;</pre><pre>&lt;p&gt;&lt;%= link_to &quot;Add a job&quot;, new_job_path %&gt;&lt;/p&gt;</pre><p>Firstly, we have a title at the top called “Jobs” — that’s not anything special, you can change “Jobs” to “London jobs”.</p><p>The mid part is a loop. It takes each individual jobs from the controller and adds some particular HTML for each one. If there are 5 jobs, there will be 5 &lt;div&gt; tags with the class “job”, each with a &lt;h2&gt; tag and a &lt;p&gt; tag inside.</p><p>The last part has a tag called link_to. This is part of Rails and will make a link that goes to the new job page. There’s some magic underneath it all but it saves you from having to write &lt;a&gt; tags yourself.</p><p>If you save and refresh your homepage, you should see something like:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/446/1*akhxnE2_svzmV4EQhQuNvA.png" /></figure><p>Click “add a job” and fill in the form again.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/424/1*u2o3RJwqR-sjEcv658AsiA.png" /></figure><p>You can do this as many times as you like. The more jobs you add, the more there will be on the home page!</p><h3>Design</h3><p>Our site looks <em>so</em> 1995 right now. 90s fashion seems to be having a comeback but for now, let’s give it a little bit of 2010s style.</p><p>Let’s go to our assets — these would be familiar to anyone who has done any front-end web development. This is where your CSS, your Javascript and your images live.</p><p>For styling, we need to edit our stylesheets. For this, we’re going to go our <strong>app</strong> folder, then our <strong>assets</strong> folder, then <strong>stylesheets</strong> folder and open <strong>jobs.css.scss.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1005/1*rdpg4EG1PwUWrCeCTHXzKQ.png" /></figure><p>Replace all text with:</p><pre>body {<br>  background: #a71b3e;<br>  color: white;<br>  font: 14px/1.5 Helvetica, Arial, sans-serif;<br>  margin: 20px auto;<br>  width: 500px;<br>}</pre><pre>a {<br>  color: #efaaca;<br>  text-decoration: none;<br>}</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/868/1*ohkV3kuxj8DaEuQJRQH8dQ.png" /><figcaption>Before and after</figcaption></figure><p>It’s not a lot of CSS and you can go crazy on how your design looks. I’m keeping it nice and simple for now.</p><h3>Extras!</h3><p>Remember back to the gems section? We said that we want pagination involved. We currently don’t have any pagination if we go over a particular number. All the jobs we ever post will be on the home page.</p><p>So how do we add in pagination? Remember we added in <a href="https://github.com/amatsuda/kaminari">the Kaminari gem</a> which does a lot of the hard work for us.</p><p>When adding a gem’s functionality, it’s always worth following the instructions, so for Kaminari, we need to add it into two places.</p><p>Firstly, back to our posts controller, we need to change:</p><pre>@jobs = Job.all</pre><p>to</p><pre>@jobs = Job.page(params[:page]).per(25)</pre><p>This means for every 25 jobs, paginate using the parameters sent through from the browser — I won’t go into this but look at the URL once it works.</p><p>The last part is to add some links on our home page to go back and forth. Go back to index.html.erb in app/views/jobs and add near the bottom (but not in the loop):</p><pre>&lt;p&gt;&lt;%= paginate @jobs %&gt;&lt;/p&gt;</pre><p>This means that there will be pagination for the @jobs in our controller near the bottom of our home page in a paragraph tag. That’s it!</p><h3>Getting help</h3><p>Stuck with any part of Rails? The best places to look are Google and <a href="http://stackoverflow.com/">Stack Overflow</a>. If you get an error, copy and paste it into Google and see if anyone else has had the same problem before. If not, search on Stack Overflow and if nothing good, sign up and post a new question.</p><p>Commons errors are:</p><ul><li>Missing or too many “end”’s in your model or controller</li><li>Check the difference between &lt;%= and &lt;% . I didn’t go into this but it’s always worth re-checking</li><li>Spaces and non-spaces between things like colons</li><li>Typos — they happen a lot, watch out for @jbos instead of @jobs for instance</li><li>Not saving files — it’s a common mistake that professionals make</li></ul><h3>Conclusion</h3><p>The average web developer salary in London is around £40,000 and developers are highly sought-after. I’ve been in the position of hiring developers and being a freelance developer, and what I’ve learnt is that a lot of development work just isn’t that hard.</p><p>A lot of developers <em>hate</em> me saying that because it “devalues their craft” but it’s a lot easier than it looks.</p><p>For the jobs board that we’ve made, <strong>I’ve edited 7 files and added 55 lines of code</strong>. Several of those lines were blank too. And of course I can make it more fully featured but it shows how quick it is to build sites like a jobs board.</p><p>Other things you could build using exactly the same tutorial:</p><ul><li>A blog (replace job with post)</li><li>A link board like Delicious (replace job with link)</li><li>A real estate listing site (replace job with house)</li><li>A review site (replace job with album/movie/gig)</li><li>A news site like Hacker News or Designer New (replace job with story)</li></ul><p><a href="http://www.techjobsin.london/">techjobsin.london</a> took about two hours to make and most of the time wasn’t coding, it was fiddling around with the design and stylesheets to make it look better.</p><p>Hopefully you can see that coding is a fun, creative outlet for lots of people that’s nowhere near as hard as it look at first glance.</p><h3>Further resources</h3><ul><li>The official Rails guides are great: <a href="http://guides.rubyonrails.org/index.html">http://guides.rubyonrails.org/index.html</a></li><li><em>RailsCasts</em> are amazing, they’ve not been updated in a while as Ryan the owner is AWOL (come back soon Ryan) but they’re still very useful: <a href="http://railscasts.com/">http://railscasts.com/</a></li><li><em>The Rails 4 Way</em> book is meant to be very good: <a href="http://www.amazon.com/Rails-Way-Addison-Wesley-Professional-Ruby/dp/0321944275/">http://www.amazon.com/Rails-Way-Addison-Wesley-Professional-Ruby/dp/0321944275/</a></li><li>Recommendation by <a href="https://twitter.com/gazaston/status/520518708029313024">Gaz Aston</a> – <em>The Ruby on Rails Guide eBook</em>: <a href="http://www.amazon.co.uk/Ruby-Rails-Guide-Stefan-Wintermeyer-ebook/dp/B00E25KVLW">http://www.amazon.co.uk/Ruby-Rails-Guide-Stefan-Wintermeyer-ebook/dp/B00E25KVLW</a></li><li>For beginner Ruby, check out <em>Why’s (Poignant) Guide to Ruby. </em>Again, the author “Why” is AWOL (come back soon Why) but the book is a hilarious and fascinating way to learn Ruby. <a href="http://mislav.uniqpath.com/poignant-guide/book/">http://mislav.uniqpath.com/poignant-guide/book/</a></li><li>For the best gems and services worth checking out, try my previous article. <a href="https://medium.com/@riklomas/my-favourite-ruby-gems-services-89fb47341c05">https://medium.com/@riklomas/my-favourite-ruby-gems-services-89fb47341c05</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9b296c4df483" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My favourite Ruby gems & services]]></title>
            <link>https://medium.com/@riklomas/my-favourite-ruby-gems-services-89fb47341c05?source=rss-9ee66467202a------2</link>
            <guid isPermaLink="false">https://medium.com/p/89fb47341c05</guid>
            <dc:creator><![CDATA[Rik Lomas]]></dc:creator>
            <pubDate>Tue, 13 May 2014 09:04:44 GMT</pubDate>
            <atom:updated>2014-05-13T09:41:58.466Z</atom:updated>
            <content:encoded><![CDATA[<h4>With Ruby and Rails, you have a great eco-system in which to build fully functioning webapps quicker.</h4><p>I love Ruby and I love Rails. It’s not unconditional love — there’s still lots of problems — but for me, it’s by far the best language <em>and</em> framework out there due to their readability, ease of use and the sheer speed in which you can make complex sites.</p><p>There’s a huge number of gems and services out there and for beginners, the good ones are hard to find. There’s also the temptation to write your own code to do something that others have already created and made open source because you didn’t know about it.</p><p>If you know of any gems or services that are great and I’ve missed them out, please add them into the comments.</p><h4>Helpers</h4><p>One of the first things I always add to any Rails project is <a href="https://github.com/plataformatec/simple_form"><strong>simple_form</strong></a> as it’s far better than the default Rails form helpers which in my opinion are fairly useless. I reckon simple_form should be integrated into a future version of Rails as standard.</p><p>For pagination I use <a href="https://github.com/amatsuda/kaminari"><strong>Kaminari</strong></a> as it’s super easy to install and use. For adding Markdown to longer text, <a href="https://github.com/davidfstr/rdiscount"><strong>RDiscount</strong></a> is pretty good. For pretty URLs I use <a href="https://github.com/rsl/stringex"><strong>Stringex</strong></a> as it’s better than the default “parameterize” in Rails and also handles non-roman characters well.</p><p>For tagging, there’s no need to do it from scratch as <a href="https://github.com/mbleigh/acts-as-taggable-on"><strong>acts-as-taggable-on</strong></a> has loads of in-built features such as related tags and search. For date parsing, <a href="https://github.com/mojombo/chronic"><strong>Chronic</strong></a> is pretty amazing and fully featured. In case your users are likely to accidentally delete things, <a href="https://github.com/radar/paranoia"><strong>Paranoia</strong></a> can let your users undo.</p><h4><strong>Geolocation</strong></h4><p>Another gem is I think is incredible is <a href="http://www.rubygeocoder.com/"><strong>Geocoder</strong></a>. As someone who worked at a business review site years ago, this gem would have saved me <em>three months</em> of coding. The following code example shows how powerful it is.</p><pre>Business.near(“Islington, London”)</pre><p>How hot is that code? If code could turn me on, that would definitely do it. On the front-end, I like to use <a href="http://leafletjs.com/"><strong>leaflet.js</strong></a> with <a href="https://www.mapbox.com/"><strong>Mapbox</strong></a> to have more control over how the maps look.</p><h4><strong>Administration</strong></h4><p>For any kind of content management, I use <a href="http://activeadmin.info/"><strong>Active Admin</strong></a><strong> </strong>— again, another gem that’s ridiculously easy to install considering the power that you get with it. The down side of Active Admin is that it’s still in the works for Rails 4 so you have to get the development version which occasionally has bugs.</p><p>For customer management (e.g. queries, newsletters, auto-emails, etc.), I usually use <a href="http://www.intercom.io/"><strong>Intercom</strong></a>. It is a paid service but it is definitely worth the $49/month that you’d need to spend. There’s a few features that they could add to make it killer but it’s very good as it is.</p><h4>Uploads</h4><p>For image and file uploads, I always use <a href="https://github.com/thoughtbot/paperclip"><strong>Paperclip</strong></a>. Again, super easy to install and it even comes with image resizing and cropping as standard (as long as you have Imagemagick installed — <a href="http://cactuslab.com/imagemagick/">Mac installer here</a>, <a href="http://imagemagick.com/script/binary-releases.php#windows">Windows installer here</a>). As I use Heroku (more later on that), I add my file uploads to <a href="http://aws.amazon.com/"><strong>Amazon S3</strong></a><strong> </strong>with the <a href="http://aws.amazon.com/sdkforruby/"><strong>aws-sdk</strong></a> gem. There’s a <a href="https://devcenter.heroku.com/articles/paperclip-s3">good guide to Heroku, Paperclip and S3 here</a>.</p><h4><strong>Search</strong></h4><p>Adding search to a site is hard work and isn’t an easy job so it’s always worth relying on a service to do it. I’ve been using <a href="http://www.algolia.com/"><strong>Algolia</strong></a>, it’s very impressive and incredibly quick to get up and running. Again, something that could take <em>months</em> to build from scratch can be added within 30 minutes. Check out <a href="http://www.algolia.com/demo">their demos</a> to see how easy it is to get something incredibly powerful. Definitely worth looking in to.</p><h4><strong>User systems</strong></h4><p>Generally, I write my user authentication from scratch as I like having the full control over it, but there are good gems out there such as <a href="https://github.com/plataformatec/devise"><strong>Devise</strong></a> if you want to add a quick user system. With my authentication, I like to use <a href="https://github.com/tenderlove/namecase"><strong>Namecase</strong></a> to properly case customers names and <a href="https://github.com/ryanb/cancan"><strong>Cancan</strong></a> to add in user roles quickly.</p><p>For any kind of social login (e.g. “Log in with Facebook”), I always use <a href="https://github.com/intridea/omniauth"><strong>Omniauth</strong></a> which is great but for my tastes it’s a little bit overcomplicated. If anyone knows anything simpler, just let me know!</p><h4><strong>Accepting payments</strong></h4><p>I’m a big, big <a href="https://stripe.com/"><strong>Stripe</strong></a> fan — if you need to accept credit cards for one-off payments or subscriptions, look into them. There’s <a href="https://stripe.com/docs/tutorials/forms">a little bit of Javascript to add</a> as well as <a href="https://stripe.com/docs/api/ruby">Ruby</a>, but it’s very quick to get going. You should definitely use Stripe over Paypal, it’ll be quicker, cheaper and they’re a really nice company too.</p><p>If you prefer to accept your payments via Direct Debit, there’s also the great <a href="https://gocardless.com/"><strong>GoCardless</strong></a> which again has <a href="https://github.com/gocardless/gocardless-ruby">some great documentation</a>.</p><h4><strong>User testing</strong></h4><p>If you’re a stats nerd (like I am) and you want to add <a href="http://en.wikipedia.org/wiki/A/B_testing">A/B testing</a> on to your site, look into <a href="https://github.com/andrew/split"><strong>Split</strong></a> which is very easy to implement. The only thing it relies on is Redis — I use the <a href="https://openredis.com/">openredis</a> Heroku add-on to use it.</p><h4><strong>APIs</strong></h4><p>To do any kind of scraping, such as what I do with <a href="http://twitter.com/newsycombinator">the @newsycombinator Twitter bot</a>, I generally use <a href="https://github.com/jnunemaker/httparty"><strong>HTTParty</strong></a> and <a href="http://nokogiri.org/"><strong>Nokogiri</strong></a> together. I use HTTParty to grab the content from the 3rd party site, then use Nokogiri to turn it into something I can parse.</p><p>There are some specific gems out there for particular social networks, for instance, <a href="https://github.com/sferik/twitter"><strong>the Twitter gem</strong></a> and <a href="https://github.com/arsduo/koala"><strong>Koala for Facebook</strong></a> are good.</p><p>For any kind of real time sites, for instance if you’re building a chat site or an online game, look into <a href="http://pusher.com/"><strong>Pusher</strong></a>.</p><h4><strong>Notifications</strong></h4><p>As I like being completely up-to-date with everything, I like getting notifications <em>everywhere</em>. If I need to be alerted to something super important, I like to get notified by SMS so I integrate <a href="https://www.twilio.com/"><strong>Twilio</strong></a> with <a href="https://twitter.com/stevegraham">Steve Graham’s</a> unofficial-but-better <a href="https://github.com/stevegraham/twilio-rb">twilio-ruby</a>.</p><p>For other notifications, I like to be notified in team chat rooms. I prefer <a href="http://www.slack.com/"><strong>Slack</strong></a> over <a href="https://www.hipchat.com/"><strong>Hipchat</strong></a> as the design is better, but they’re both great. Use the <a href="https://github.com/sosedoff/slack-notify">slack-notify</a> or <a href="https://github.com/hipchat/hipchat-rb">hipchat-rb</a> gem to hook your site up!</p><h4><strong>Emails</strong></h4><p>When I’m sending out emails, I like to use <a href="http://sendgrid.com/"><strong>Sendgrid</strong></a><strong> </strong>as it’s super easy to set up and get running. HTML emails are a pain in the arse to build so using the <a href="https://github.com/Mange/roadie"><strong>Roadie</strong></a> gem is a great way to solve some of that pain by inlining your stylesheets.</p><h4>Errors</h4><p>Every site has errors, there isn’t a developer out there who doesn’t. Even <a href="https://twitter.com/dhh">DHH</a> gets bugs. Even if you test your site to death, there will be errors popping up that you didn’t realise you could get. Don’t feel bad about getting them, instead just be notified properly and learn to debug better.</p><p>I use <a href="https://getsentry.com/welcome/"><strong>Sentry</strong></a> to let me know via email and Slack if there’s errors popping up. <a href="https://papertrailapp.com/"><strong>Papertrail</strong></a> is also great if you want to search through your logs to find certain events happening.</p><p>For development debugging, <a href="https://github.com/charliesome/better_errors"><strong>better_errors</strong></a> is ridiculously good (install it with <a href="https://github.com/banister/binding_of_caller">binding_of_caller</a>). It replaces the usual Rails error page with much more information and also live in-browser debugging too. Again another gem that should be part of the default Rails platform.</p><h4>Testing</h4><p>Now of course, we don’t want to have to use the error services and gems above. We still want to make sure our code still works whenever we change things.</p><p>I’m a fan of <a href="https://github.com/seattlerb/minitest"><strong>Minitest</strong></a><strong> </strong>over <a href="https://github.com/rspec/rspec">RSpec</a> but they’re both good for general automated testing. For writing integration tests (e.g. automating users clicking around your site), look into <a href="https://github.com/jnicklas/capybara"><strong>Capybara</strong></a> — it’s pretty awesome and very useful.</p><p>For making test data, I prefer to use <a href="https://github.com/thoughtbot/factory_girl_rails"><strong>Factory Girl</strong></a> over fixtures for its flexibility. To make testing something that you just leave running in the background, look at <a href="https://github.com/guard/guard"><strong>Guard</strong></a>.</p><p>Other useful testing gems include <a href="https://github.com/vcr/vcr"><strong>VCR</strong></a> (a way to test with APIs) and <a href="https://github.com/travisjeffery/timecop"><strong>Timecop</strong></a> (a way to freeze and fast forward time).</p><p>One service I’ve used which I really like is <a href="https://codeclimate.com/"><strong>Code Climate</strong></a> which helps you keep the quality of your code very high — the only downside, which is why I stopped using it, is that the price is just far too high for most startups.</p><h4>Deployment</h4><p>I admit that I’m a lazy developer. The reason I code is to do things that I can’t be bothered to do manually. This is why I use the amazing <a href="https://www.heroku.com/"><strong>Heroku</strong></a>. I don’t want to worry about setting up Linux and running patches in the future, I’d rather pay a little bit more to have other people fixing things behind the scenes. Heroku is by far the easiest platform to get set up on.</p><p>I like to set up three different servers to have different environments to test with. They are development (just to try out things), production (the live site) and staging (the same database as production but to try out code before it goes live). To do this I use the <a href="https://github.com/fastestforward/heroku_san"><strong>heroku_san</strong></a> gem to help me out. All I need to do to deploy to staging is:</p><pre>rake staging deploy</pre><p>Easy peasy.</p><p>To make sure all the code is super quick, I use <a href="http://newrelic.com/"><strong>New Relic</strong></a> to monitor speeds and spot any slow areas of sites. I also use caching on production using the <a href="https://github.com/mperham/dalli"><strong>dalli</strong></a><strong> </strong>gem with<strong> </strong><a href="https://www.memcachier.com/"><strong>Memcachier</strong></a>. The <a href="https://addons.heroku.com/scheduler"><strong>Heroku Scheduler</strong></a><strong> </strong>add-on is great.</p><p>If you’ve never tried Ruby or Rails before, I would definitely recommend looking into it. I used to write PHP, then I moved on to Python with Django but I’ve found Ruby on Rails to be by far the best framework to create sites. This is why I recommend it to clients and students of mine.</p><p>The best way to learn Rails is to <a href="http://guides.rubyonrails.org/">read the official guides</a> and ignore most of the online learning platforms like Codecademy as they’re not very practical. Have a project that you want to build, have some grit and be determined!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JVbhXXixYlprjo9ib6lpKw.png" /><figcaption>New side project</figcaption></figure><p>I have a new project that I’m launching the next few weeks that will let people find and discover new gems called <a href="http://www.bundleupdate.com/"><strong>Bundle Update</strong></a>. I’m very excited to get it live and it’s looking great. I want to make it easier to find out what other people use and make it easier for beginners to get started. Sign up and I’ll let you know when it launches.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=89fb47341c05" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>