<?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 Tim Edwards on Medium]]></title>
        <description><![CDATA[Stories by Tim Edwards on Medium]]></description>
        <link>https://medium.com/@tim_shane?source=rss-2a6ecf601bae------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*h7tgklTYgcWNftF1F2BrBQ.jpeg</url>
            <title>Stories by Tim Edwards on Medium</title>
            <link>https://medium.com/@tim_shane?source=rss-2a6ecf601bae------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 01:19:25 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@tim_shane/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[I Don’t Miss Physical Media. I Miss Owning Things.]]></title>
            <link>https://medium.com/@tim_shane/i-dont-miss-physical-media-i-miss-owning-things-811960d11beb?source=rss-2a6ecf601bae------2</link>
            <guid isPermaLink="false">https://medium.com/p/811960d11beb</guid>
            <category><![CDATA[digital-media]]></category>
            <category><![CDATA[right-to-repair]]></category>
            <category><![CDATA[analog]]></category>
            <dc:creator><![CDATA[Tim Edwards]]></dc:creator>
            <pubDate>Tue, 07 Jul 2026 16:03:02 GMT</pubDate>
            <atom:updated>2026-07-07T16:03:02.819Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I5xJzVvV4sktNO7wW6cpfA.png" /></figure><p>There’s been a tremendous volume of content made over the past week after Sony announced the next Playstation will forego physical media. Good riddance. Even if discs survive unscathed, they still only have a 10–20 year lifespan due to the degradation of the imaging layer.</p><p>I grew up recoiling at skips on scratched CDs, rewinding VHS tapes, and games becoming unplayable from damage not even the Disc Doctor could buff out. I don’t miss it. Digital is better in <em>almost</em> every way: instant access, no clutter, nothing slowly dying on a shelf somewhere.</p><p>We’re arguing about the wrong thing.</p><p>We’re debating formats when the real issue is happening out in the open. Sony is removing 551 movies from Playstation libraries wether you purchased it or not. Terminator 2, Paddington, Total Recall, gone. Amazon deleted George Orwell’s 1984 from Kindles, which, <em>come on</em>. Microsoft and Google broke movie ‘ownership’ when they killed divisions and partnerships. Ubisoft shut down the video game “The Crew” despite people having paid up to $60 for it. The legal argument underpinning all of this? You never owned them in the first place. This isn’t about formats. Even if you purchased EA’s Anthem, it went offline this year and the game is dead even if you had the disc.</p><p>It’s not just media. BMW was testing a subscription for heated seats that were already physically installed in the car. Tesla turned “full self-driving” (this is a topic for another day) from a $15,000 option to a $99 month subscription. Want to own Photoshop? Sorry buddy — go back to 2013 and buy CS6. Samsung now requires an account to use basic TV features. John Deere locked their tractors so farmers couldn’t repair their own equipment (and a lawsuit is forcing it to pay $99M to correct their mistake).</p><p>The pattern is emerging everywhere. You think you bought something. What you actually bought was permission. And they can take it back whenever they want.</p><p>What I want people to realize is that digital ownership isn’t some impossible technical problem. It’s a business decision. You don’t rent your shoes solely (pun) because shoes are harder to take away. <a href="http://gog.com/">GOG.com</a> has sold DRM-free games for years. The EU’s Right to Repair passed in 2024 restricting electronics manufacturers from blocking repair and requiring OEM parts to be available. Microsoft’s Surface Pro went from a 0/10 repairability to 8/10 in response. Stop Killing Games hit 1.29 million signatures in the EU despite not being picked up. It’s still continuing to gain traction worldwide.</p><p>The disc is dead — and I’m find with that. I’m not fine with companies, in the hunt for aggregating the market for more profit, whittling away our ownership in the things we purchased. Our right to own and repair is worth fighting for. Check out <a href="http://repair.org/">repair.org</a> if you’d like to do something about it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=811960d11beb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Breaking Up with the iPhone]]></title>
            <link>https://medium.com/techburst/breaking-up-with-the-iphone-4c5e837c22c7?source=rss-2a6ecf601bae------2</link>
            <guid isPermaLink="false">https://medium.com/p/4c5e837c22c7</guid>
            <category><![CDATA[smartphones]]></category>
            <category><![CDATA[android]]></category>
            <category><![CDATA[google]]></category>
            <category><![CDATA[apple]]></category>
            <category><![CDATA[tech]]></category>
            <dc:creator><![CDATA[Tim Edwards]]></dc:creator>
            <pubDate>Mon, 13 Nov 2017 19:07:14 GMT</pubDate>
            <atom:updated>2017-11-16T01:42:04.279Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DurKnxRImfNZSMHKc5S3hQ.png" /><figcaption>The Google Pixel 2</figcaption></figure><p>It isn’t easy for me to say this: After ten years together, I’m leaving the iPhone. We’ve been slowly drifting apart the past couple of years. And while checking out Google’s latest event, a flicker of thought began. I felt like something was missing from the iPhone. To make a long story short: it’s Google’s software.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/390/1*1pckwTNAer6xlIDlBKKOMA.gif" /></figure><p>I’ve been planning on buying the iPhone X since before the iPhone 6s release.<a href="https://twitter.com/tim_shane/status/811651918519037956"> I wanted the X when it was more likely known as the ‘iPhone Pro’, or the ‘8’</a>. I even bought an SE this year just to get me by when my 6+became capricious. All of that has changed in the last few months. Read on as I dive into why Google impressed me so.</p><h3>Hardware</h3><blockquote>We’re finally getting back to the glory days of iPhone design: shatterable on both sides.</blockquote><p>I like the iPhone X, a lot. All stainless steel and glass. It is <strong>the</strong> premium phone. The iPhone 4 <em>was </em>the nicest to hold. The iPhone X is like a Bentley Continental. A beautiful look and feel. Powerful. Metal buttons. Hand stitched leather. Yet, the technology doesn’t move the needle very much. Now I know what you’re saying: “Tim, literally everyone agrees this phone is the future of phones.” The iPhone X is just the incremental hardware upgrade that we should have had last year. It has an IR dot projector and a (very nice) new screen. The facial features are neat. But they solve problems I didn’t really have. And let’s be honest, people will care about Animoji as much as iMessage stickers in 6 months. It has already quieted down, but with the holiday season right around the corner, we are sure to see a swell of them.</p><p>Another large (personal) complaint: I hate rounded edges. The iPhone 4, 5, C, SE, and yes, the Pixel 2 have lovely flat edges that let you hold them with assuredness. I was never so clumsy as with the iPhone 6+.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TZPnmLenPZmW38vCySHSSA.png" /></figure><p>The iPhone X is undeniably sexier than the Pixel 2. The Pixel 2 is subdued. It comes in a soft white, black, and a rainy-day ‘kinda blue’. The X is something you behold. The Pixel 2 is a Tesla to the iPhone’s Bentley. <a href="https://www.theverge.com/2017/10/4/16405192/new-google-pixel-2-xl-phone-photos-video-hands-on">The Verge called it Scandinavian</a> in their first look. It shrugs off it’s bezels when every other phone is compromising <em>something </em>to kill them.</p><p>The simple industrial design of the Pixel 2 is almost a statement about the current problem in the industry. It’s an issue that Jony Ive is certainly grappling with: How can you make a small, thin rectangle look new? Jony went the materials route. Google ignored the question entirely. How do you make it more helpful? They’ve made a metal body, but covered it with a textured coating not because it looks better, but because it is better to <em>use. </em>The Pixel 2 is designed with usability in mind above all else. The Pixel 2’s hardware is confidently basic.</p><h3>Software</h3><p>The software is where the Pixel 2 makes up for its subdued appearance. It’s a combination of a dozen little things.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bwg3E2IB0IA75INKBk8WpQ.jpeg" /></figure><p>They have Apple’s sense of ecosystem but in an open manner. Take The Google Home, Chromecast, and Nest relationship shown in the picture. “Ok, Google, who’s at the door?” lets the Google Home tell the Nest doorbell camera to display the doorbell camera onto your Chromecast. Sure, Apple has integrations. The difference is each of these components works independently from the Google platform. Your Nest, Chromecast, and Google Home work just fine if you have an iPhone. Want to control your Apple TV from your Pixel? Nah. Play a Spotify playlist from Siri? Lol. Siri can’t even control my thermostat because Apple uses its own home connectivity platform.</p><blockquote>Apple’s walled garden of services was fine when they were the best.</blockquote><p>This hits me in the face every day.</p><ul><li>Links open in Safari, rather than Chrome. Would it be so awful to cede control here, Apple? Google’s browser is better.</li><li>Apple now allows for a third party keyboard, but I constantly have to wrestle away Apple’s keyboard.</li><li>Apple Maps is getting better, but WHY? Google Maps is better in pretty much every way. <a href="https://www.justinobeirne.com/a-year-of-google-maps-and-apple-maps">Here is an excellent detailed comparison of the rate of development between the two.</a></li><li><a href="https://www.macobserver.com/tmo/article/audible_pulls_in-app_purchases_from_iphone">I can’t buy an audiobook in the Audible app.</a></li><li>Siri is comically bad most of the time.</li><li>When I get in my car while listening to Audible or Pandora, the Bluetooth connection forces Apple Music to start playing. (This is hopefully considered a bug in iOS 11. But Apple could conceivably think it’s a feature…)</li><li>Apple could open iMessage to the world, but instead chooses to bifurcate it (unlike Google with Allo).</li><li>Google Photos is spectacular, free, and open across platforms. It’s a joy to use from any device.</li><li>Let’s not forget Apple forcing Bing on us in Siri…</li></ul><p>Other services will always live as second-class citizens on iOS. This didn’t seem as apparent over the past couple of years. Possibly because iOS was really neck-in-neck with Android. After watching the Google event, I started to feel that <em>better</em> services live as second-class citizens. In Android, Spotify can be your default music player? Lastpass can be set as a default password store? Here I was, happy for the ability to press ‘Send’, select Lastpass, authenticate Lastpass, and finally select the credential. A little dance that surely slows the adoption of good password practices.</p><blockquote>A bunch of little conveniences that let your phone work better wouldn’t have been enough for me to jump platforms.</blockquote><p>I get glimpses of Google’s AI on the iPhone. The first time it happened was two years ago. I had taken several small clips of my kids during Halloween festivities with the goal to stitch them together the next day in iMovie. That morning, Google notified me that it had made a movie. It had automatically done what I was planning, and the automatic results were better than I would have produced. I was absolutely floored. Today, Google categorizes my photos with impressive accuracy and are available through a website that makes icloud.com look every bit the afterthought that it absolutely is.</p><p>The software prowess of Google is also becoming clear in the Pixel 2’s camera, which currently sits atop the smart phone camera throne. Using software and hardware techniques, the Google rivals Apple’s dual-camera portrait settings, <a href="https://petapixel.com/2017/10/05/power-googles-new-fused-video-stabilization/">stabilizes video</a>, <a href="https://store.google.com/us/product/google_clips?hl=en-US">and determines capture-worthy moments</a>. Google’s AI-first software will <em>only get better</em>.</p><p>Google also has neat tricks like searching Google with pictures you take, telling you what song is playing on the always-on home screen, hiding notifications temporarily the way Inbox can schedule emails, automatic photo sharing suggestions, joining a WiFi network from a photo of the network details, and more.</p><blockquote>Warning: I finish up with arm-waiving corporate philosophy. I have opinions, but do not harbor truth of what ought to be.</blockquote><p>While Google embraces an AI-centric platform that it iterates and sharpens constantly, Apple seems to be hoping to stumble upon the next big thing. <a href="https://www.theverge.com/2017/10/31/16579748/apple-iphone-x-review">Verge’s iPhone X review proclaims: “Face the future”</a>. The thing is, I don’t believe Apple cares about <em>the future</em>. Apple care’s about Apple’s future. I don’t blame them! It’s what companies typically do. Google, though, <a href="https://futurism.com/internet-delivering-balloons-puerto-rico/">looks </a>to <a href="https://waymo.com/">truly </a>care about <a href="https://verily.com/"><em>the future</em></a><em>.</em></p><p>Apple’s main selling point is a bundle of services that work reasonably well together on top of great hardware. And cracks in hardware decisions are starting to show. Previously, Apple sold a low-cost iPhone, last year’s model, and the new iPhone. Now, Apple sells the small 4-year-old version with 2-year-old internals, the medium and large 2-year-old versions, the medium and large 1-year-old version, and a brand new medium and large version that is completely out-shadowed by the flagship medium version with a large screen.</p><p>A more sensible and simple lineup would be: iPhone SE &gt; iPhone 8/8+ &gt; iPhone X, with the 7 sold while supplies last.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/860/1*uDAZL__VFmZA9emrJbJJSg.jpeg" /></figure><p>The failure of the Mac Pro, the forced march to USB-C in MacBooks in tandem with the proprietary lightning port on the phone, and the $20,000 Apple Watch Edition show a company with $250bn but no direction (also evidence: calling the top tier watches ‘Edition’. What does it even mean?). They are pushing into civic ventures with ‘Town Squares’. Beautiful Apple locations with community-driven events. Apple’s solution to competing in hardware is getting people just to love Apple more.</p><p>Looking at company values, there isn’t any question why Google’s new phone is so appealing. Google has the experience and positioning to take phones beyond the limitations of the form factor they are in, while Apple seems stuck competing with its own success.</p><p>I don’t want a shinier phone; I want a smarter one.</p><p>Thanks for reading my ramble about consumer technology. You can find me on Twitter ‘liking’ many different things. The ‘Clap’ button goes to 50 in case you were wondering but were too shy to test the limits.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4c5e837c22c7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/techburst/breaking-up-with-the-iphone-4c5e837c22c7">Breaking Up with the iPhone</a> was originally published in <a href="https://medium.com/techburst">techburst</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Getting Started with Machine Learning— Fall ‘17]]></title>
            <link>https://medium.com/techburst/getting-started-with-machine-learning-fall-17-af5a71204c8b?source=rss-2a6ecf601bae------2</link>
            <guid isPermaLink="false">https://medium.com/p/af5a71204c8b</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Tim Edwards]]></dc:creator>
            <pubDate>Tue, 24 Oct 2017 18:11:29 GMT</pubDate>
            <atom:updated>2017-10-25T16:50:26.775Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/682/1*b-fiRiBduey-TE39z_QExA.png" /></figure><p><a href="https://www.linkedin.com/search/results/content/?anchorTopic=84386">Over on LinkedIn</a>, there is a discussion on the <a href="https://www.nytimes.com/2017/10/24/learning/teaching-activities-for-tech-giants-are-paying-huge-salaries-for-scarce-ai-talent.html">New York Times article</a>, which states that entry level PhD’s or people with a few years of experience with machine learning are making millions from a shortage of AI skilled labor. Dozens of people are asking for good resources to get started. This is by no means a <a href="https://unsupervisedmethods.com/my-curated-list-of-ai-and-machine-learning-resources-from-around-the-web-9a97823b8524">compendium</a>, but by the end of it, you should have plenty of experience to find your own. These are just my picks that have helped me get moving. Think of it as your first semester before a lifetime of learning.</p><h3>Online Courses</h3><p><a href="https://www.coursera.org/learn/neural-networks">Neural Networks for Machine Learning | Coursera</a></p><p>Geoffrey Hinton’s course on neural networks. He basically started this whole field. Fun fact from Wikipedia: He is the great-great grandson of <a href="https://en.wikipedia.org/wiki/George_Boole">George Boole!</a></p><p><a href="https://www.udemy.com/deeplearning/">Deep Learning A-Z™: Hands-On Artificial Neural Networks</a></p><p>This course on Udemy walks you through several types of neural networks and more using neat tools, like <a href="https://keras.io/">Keras</a>, with homework after each section to help you understand the structure and flow of data. It’s paced very well. I feel like courses move too slowly, but this struck the right balance.</p><p><a href="https://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-867-machine-learning-fall-2006/index.htm">Machine Learning</a></p><p>This is a snapshot of the course from 2006, so don’t expect to use the latest tools, but this is a great course to go deeper into what modern tools do today. This will give you a deeper grasp on models, classification, and the math that powers ML. <a href="https://www.smbc-comics.com/comic/why-i-couldn39t-be-a-math-teacher">Emphasis on the math.</a></p><p><a href="https://pythonprogramming.net/machine-learning-tutorials/">Python Programming Tutorials</a></p><p>Harrison, <a href="https://www.engadget.com/2017/05/19/watch-an-ai-teach-itself-to-drive-in-gta-v/">of recent Python plays GTA5 fame</a>, has hours of in-depth Python tutorials that follow a terrific pattern of building a basic version from scratch, and using the popular tool sets. Honestly, it was this story hitting Reddit earlier this year that got the gears turning in my head about machine learning.</p><h3>Web</h3><p>More of a collection of ‘getting started’ pages and tools than ‘tutorials or courses’. These are great to get up to speed, and if you are constantly referring back to them, your probably doing things right. Don’t stress.</p><ul><li><a href="https://www.tensorflow.org/get_started/">Tensorflow</a></li><li><a href="https://keras.io/">Keras</a></li><li><a href="http://scikit-learn.org/stable/">Scikit-Learn</a></li><li><a href="http://archive.ics.uci.edu/ml/datasets.html">DATA!</a></li><li><a href="https://unsupervisedmethods.com/cheat-sheet-of-machine-learning-and-python-and-math-cheat-sheets-a4afe4e791b6">ML Cheat Sheets</a></li></ul><h3>Book</h3><p>Ok, so there’s just one that I’ve read so far about ML (web resources are just so rich):</p><p><a href="https://www.amazon.com/Make-Your-Own-Neural-Network-ebook/dp/B01EER4Z4G/ref=sr_1_1?ie=UTF8&amp;qid=1508867471&amp;sr=8-1&amp;keywords=neural+network">Make Your Own Neural Network</a></p><p>It’s a brief book, but I appreciate how it covers the mathematical interplay between layers of a neural net. It helped me see neural nets in a very clear way and spawned a couple of projects I’ve been working on. It’s also $3.98 on Kindle. Neural network knowledge for less than a latte.</p><p><a href="https://www.amazon.com/Algorithms-Live-Computer-Science-Decisions/dp/1627790365">https://www.amazon.com/Algorithms-Live-Computer-Science-Decisions/dp/1627790365</a></p><p>Algorithms to Live By might help you spark an idea about applying ML to your life or generate questions to try to answer. This was one of my favorite recent reads.</p><h4>That should get you started</h4><p>I’m not an expert in machine learning (yet). I’m just really passionate about technology, and this is the <a href="https://www.amazon.com/New-Thing-Silicon-Valley-Story/dp/0393347818">new, new thing</a>. Take the time in the middle of all of these to pursue an idea that pops up. If you’re like me, the real learning comes through trial and error. Try to encode eBay listings for classification to predict the final auction price. Try to find the average value of a home in your city based on location. These were random thoughts I had while I worked through these resources. Try to ask interesting questions and fit them into data.</p><p>There are an ocean of things to learn, and if you have any recommendations, please comment below. Thanks for reading.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=af5a71204c8b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/techburst/getting-started-with-machine-learning-fall-17-af5a71204c8b">Getting Started with Machine Learning— Fall ‘17</a> was originally published in <a href="https://medium.com/techburst">techburst</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Misadventures in Python]]></title>
            <link>https://medium.com/@tim_shane/misadventures-in-python-a2aa21cdca51?source=rss-2a6ecf601bae------2</link>
            <guid isPermaLink="false">https://medium.com/p/a2aa21cdca51</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[development]]></category>
            <category><![CDATA[learning-to-code]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[web-development]]></category>
            <dc:creator><![CDATA[Tim Edwards]]></dc:creator>
            <pubDate>Sat, 24 Jun 2017 22:38:55 GMT</pubDate>
            <atom:updated>2017-06-24T22:40:57.723Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*Hmke9oEBFs1YO_Jo0nqFhg.jpeg" /><figcaption>Hope you like word clouds. There are a quite a few in this article.</figcaption></figure><p>(TL/DR? Check out <a href="http://wordcloud.fun/">wordcloud.fun</a>! <a href="https://github.com/tim-shane/WordCloudSearch">Based off of this Github repo</a>)</p><p>Hope you like word clouds. There are a quite a few in this article. To make a long story short: I want to be a full-time software developer. There are ideas that I want to breathe life into, and programming is an outlet I’m dying to become proficient at to create these things. Python became the starter language for me due to its ease and flexibility. A few weeks ago, I had just finished scraping winespectator.com’s daily picks and was a proud owner of a database with over 5000 wine reviews. After learning that there is indeed a correlation to price and rating (*shock*), I had an idea to make a word cloud of the reviews. This led me to make an image mask to shape the cloud, which led to a better understanding of how an image is just an array. I made a thing, and it was neat:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qSiptKy7ah3WfoEXZ-_jhA.png" /><figcaption>Neat.</figcaption></figure><p>Two hours of fiddling and compiling, and compiling, and compiling led me to a point where I sat back and admired my creation. Mere seconds into my satisfaction I wondered, “How easy would it be to 1) search for a term, 2) take the detail text from each search result, 3) take the first image search result, and 4) turn all of that into a word cloud automatically. That sounds sort of simple. This thought led to much more learning than I had expected.</p><h4>Self-discovery:</h4><ul><li>Making my first PyPI package (pip install bing_cloud_search! Magic!)</li><li>My first Flask web-app that serves up MatPlotLib images.</li><li>My first web-hosted application.</li><li>Git is life.</li><li>I just can’t do this on Windows.</li><li>I’m just scratching the surface.</li></ul><h4>Bing_cloud_search</h4><p><em>H/t to </em><a href="https://github.com/tristantao"><em>Tristan Tao</em></a><em> for making this part possible through py_ms_cognitive.</em></p><p>When I first put <a href="https://github.com/tim-shane/cloudy_search">‘cloudy_search’ on Github</a>, I started to feel a little self-conscious about how little it resembled the other packages that I was using. For one, my naming conventions were not helpful.</p><pre>def make_it_rain(search_term, search_modifier=””):<br># Return Bing search snippets<br>text = return_txt(search_term)</pre><p>I had put the functions in separate .py files. I had to import get_txt.py to run ‘return_txt’. It just felt messy. You also had to hard-code a Bing API key into both ‘get_txt’ and ‘get-image’. Sloppy. Look at how grown up my little code looks now:</p><pre><strong>class CloudySearch</strong>(object):</pre><pre> <strong>def </strong>__init__(self, api, search_term, search_modifier=””):<br> <em>“””<br> </em><strong><em>:param</em></strong><em> api: Bing v7.0 api key<br> </em><strong><em>:param</em></strong><em> search_term: This is the basis for the web and image search.<br> </em><strong><em>:param</em></strong><em> search_modifier: This allows you to alter the image returned without modifying the web search.<br> A good example would be adding ‘vector’ to your search to make a more pleasing word cloud image without making<br> the words all about vector art or stock photo sites.<br> </em><strong><em>:return</em></strong><em>: <br> “””<br> </em>self.api = api<br> self.search_term = search_term<br> self.search_modifier = search_modifier</pre><p>Now that I’m not hard-coding my API, I’m free to pass it on from Heroku’s config variables, or a private config file! That was pleasing. One of many lessons I’ve learned. Some of the changes came from self-consciousness of posting code to the world. Others came from just making it easier to use for myself and others. At the end of the day, when you called this app and passed a search term, you got a word cloud!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*oL7BLFUWjyakDL6fHPl6Qw.png" /></figure><p>The first time a real, image-masked word cloud popped up, a dose of adrenaline ran through me. I was <em>thrilled. </em>I had a thought, and hours later, I created something. One nagging thought bothered me: If someone wanted to try my silly little thing, they needed a Bing API key and run it in a Python IDE. I wanted the masses to make word clouds. I needed this on the web.</p><h4>Enter Flask</h4><p><em>H/t to </em><a href="https://blog.miguelgrinberg.com/"><em>Miguel Grinberg</em></a><em>. He literally wrote the book. Almost every Flask question I had was answered by his blog, or by him on StackOverflow.</em></p><p>I ran through Grinberg’s tutorial last year. Combining the examples from the tutorial with general web dev knowledge, it was pretty simple to get a basic website built. This was a somewhat trial and error process. I had several questions to get through that were not common knowledge to me:</p><ul><li>What’s the best way to load an image from MatPltLib in HTML? Savefig to an image folder, then pass the saved file name to the next page.</li><li>What’s the best way to pass form data to the next page? Session worked out (Very related to the last question).</li><li>What’s the best way to not have your API hanging out in your app files? A config file.</li></ul><p>I had a locally running Flask app running pretty quickly with one huge asterisk. At some point of editing CSS, it just stopped working. My form stopped being allowed to communicate. Method not allowed — Error 405. My form could PUT and GET. Everything could. What method held it back? After 30 minutes of wheel spinning, the fix was changing:</p><pre>&lt;form action=&quot;&quot;&gt;     to     &lt;form action=&quot;/search&quot;&gt;</pre><p>But why? It worked as “” before. If only there were a way to bookmark working builds and work on a separate version, so as to not break anything… Git! I am only scratching the surface here, but after creating a repository and reviewing each commit’s changes, it gave me an entirely new perspective. I started thinking about the changes and how the code could be a bit shorter or a bit smarter. I’d later find that it makes package creation and app deployment smarter as well.</p><p>Once I narrowed my app down and made it look a little better, I started wondering how to include the pivotal ‘bing_cloud_search’ with my code. Should I just copy the folder into the project directory? That didn’t feel right. One Github project would be included in another. Change management cringe. What if it was a package you could just download? Pip install bing_cloud_search. How do you get your package on the pip? A trip to <a href="https://pypi.python.org/pypi?%3Aaction=submit_form">p</a>ypi.python.org and then g<a href="https://github.com/pypa/twine">ithub.com/pypa/twine</a> when I still didn’t quite get it did the trick.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/856/1*raqqkyb0iZcbbAg00tw1dA.png" /><figcaption>Ok, maybe I flailed a bit more than I let on.</figcaption></figure><p>Installing my package through pip was another exhilarating moment. My little tool became a real thing just then. I was Gepetto with less creepy creations. Maybe one day, my little word cloud machine will find its way to another oddly specific use. Some Twitter or Reddit bot perhaps!</p><h4>Heroku to the Rescue</h4><p>Honestly, when I first went to Heroku and started trying to upload my app, it felt intimidating. I started thinking “maybe this is too complicated for me.” My first app’s push to Heroku failed miserably, and I couldn’t figure out why from the logs. I briefly looked at Google, which was about twice as complex. Amazon’s simple pricing informed me that the free tier would only cost thirty bucks. Azure wanted me to contact customer service when my account created failed with no explanation. Heroku it is. I’ll make it work.</p><p>The biggest issue was that it needed to run the app with GUnicorn, and GUnicorn didn’t run on Windows. Not being able to test my builds before pushing them into production wasn’t very easy. In fact: Many things with Python just aren’t easy on Windows. I’m always trying to find alternate ways of doing things. This process solidified something with me: I’m forcing myself to develop on my iMac. There will be growing pains, but I need to be much more comfortable with Unix environments than I am.</p><p>Once I got everything set up on the Mac, I set out to find why things were broken. Google. Error. Something, something, edit code. I stopped launching my site after firing up the app. I just went straight to the logs to watch my app flail about and crash upon initialization. Google. Error tkinter. Something, something, matplotlib.use(‘Agg’). After entering heroku logs — tail , something magic happened: Nothing. After the process started its first worker on the fourth line of the log, the hundreds of lines of log errors that preceded starting over never appeared. heroku open. And there it was: the third fantastic feeling of this project. My app was on the internet!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/872/1*k6iXfzesMjEGuHzWabP4JQ.png" /><figcaption>It’s also a bit… hard on the eyes.</figcaption></figure><p>I quickly found that I can pass variables to my app (like API keys), and it can deploy the master branch from Github automatically. I’m pretty excited about using Heroku in the future.</p><h4>The Future of Word Cloud Search</h4><p>I’m going to keep tinkering with WCS because it still sucks.</p><ul><li>If you enter a search term in which it can’t download the image (beer, for instance), it just stops communicating (which is a result of my efforts on proper error handling. It used to just crash).</li><li>I want to send the images to a CDN.</li><li>I want to find out if my images are just being destroyed after each worker ends.</li><li>I want to check for duplicate searches to improve speed and reduce API calls.</li><li>I might want a better landing page with recent word clouds.</li><li>I want it to be more mobile friendly.</li><li>I want to improve the handling of the image. Right now, the image starts as a request for the jpeg URL → BytesIO object → PIL Image object → Numpy array → PNG. Is there a way to speed the process? *shrug*</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*8wbypQGenImt59utKyHEVA.png" /></figure><p># Return Bing search snippets+ text = return_txt(search_term)</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a2aa21cdca51" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Alexa, tell Apple history is repeating itself.]]></title>
            <link>https://medium.com/@tim_shane/alexa-tell-apple-history-is-repeating-itself-a8cbd2b2cc40?source=rss-2a6ecf601bae------2</link>
            <guid isPermaLink="false">https://medium.com/p/a8cbd2b2cc40</guid>
            <category><![CDATA[tech]]></category>
            <category><![CDATA[amazon-echo]]></category>
            <category><![CDATA[amazon]]></category>
            <category><![CDATA[smart-home]]></category>
            <category><![CDATA[apple]]></category>
            <dc:creator><![CDATA[Tim Edwards]]></dc:creator>
            <pubDate>Fri, 09 Jun 2017 16:38:23 GMT</pubDate>
            <atom:updated>2018-01-09T16:04:46.534Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*421a5zQMiVdutd16Lm0aBg.jpeg" /></figure><p>We’re in the middle of a new industry battle. Its significance is somewhere between Mac vs. PC and Blu-Ray vs. HD-DVD. In the future, who will turn on your oven? Alexa, Siri, or Google? In the AI-driven future, the software you talk to will be a big deal.</p><p>I appreciate both Amazon and Google’s approach to a smart-home assistant. Google’s prowess in AI, search, and knowing everything about you will give it a serious edge in this race. Their rumored home assistant/WiFi combo device is an excellent idea. Google is also already thinking through current issues in home assistants: multiple users. However, this story concerns mainly Amazon and Apple. Their two product lines are going in very different directions.</p><h4>The Two Rivals</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jetP-UWqPzA3hPg2CABW3A.jpeg" /></figure><p>Apple’s ‘fashionably late’ strategy shows up again with the HomePod, a Siri-powered speaker with some serious audio chops. If you already live comfortably within Apple’s ecosystem of Apple Music, iDevices, and are happy with Siri, there is no doubt you’ll be pleased. Not only do early impressions agree that it’s speaker array and audio processing sound impressive, but it also doubles as a HomeKit base. HomeKit, to date, requires the presence of your iPhone to authenticate commands to your smart home devices. With a HomePod, you can use your Apple devices to control your home from the internet. ‘Hey Siri, play my favorite songs.’ will spring the HomePod to life, playing you music you listen to often. Siri’s home presence is available across the room, or on your wrist.</p><p>The Echo started the home assistant race with its black cylinder that plays music, keeps your shopping list, or ordering things from Amazon just by talking to it. Its ability to connect with the accounts of your current smart objects and having an API to create ‘skills’ makes the Echo’s abilities grow rapidly. Amazon has since expanded the Echo line to the low-cost Dot, the portable Tap, the Look which analyses your wardrobe, and the Show which has a screen to display information and enable video calling. The echo lineup recently introduced a communication system that allows Echo owners to call or message each other. Amazon allows any hardware manufacturer to incorporate Alexa into any product with a speaker and microphone. The Ecobee 4 smart thermostat has Alexa built-in. LG has Alexa in it’s latest smart refrigerators. Ford is using Alexa for the most recent generation of Sync in its cars. Alexa will be everywhere.</p><p>This echoes (*wink*) back to the Mac vs. PC war. Apple, shunning third-party hardware, wanting to keep their software on well-crafted Apple-made hardware, and Microsoft wanting their software running on everything that processes. Apple fought in the PC market without advancing it, and it nearly wrecked them. When they joined the MP3 player race, they did so with a groundbreaking way to buy music. Amazon is the innovator in the home space. My two six-year-olds use Alexa daily to check the weather, the date, hear ‘kid jokes,’ and send messages to their grandparents. They’ve listened to books, confirmed animal facts, and more. If I’m a parseltongue, they are the AI native. Alexa’s ubiquity, skill, and forgiving ease of use will create a growing divide as my children grow up.</p><h4>The Future Will Be Alexa</h4><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2F98BIu9dpwHU%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3D98BIu9dpwHU&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2F98BIu9dpwHU%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/57e0489b0629988254fcf171b5086024/href">https://medium.com/media/57e0489b0629988254fcf171b5086024/href</a></iframe><p>The divide between what Alexa and Siri can do will continue to grow with Amazon’s ambitions and openness. Right now, Alexa can order you pizza. Siri can look up and call a pizzeria. Alexa can reorder paper towels. Siri looks up places to buy towels. Soon, you’ll order groceries delivered within an hour. Conversely, anything Siri can do, Alexa can do better, except for communication on your phone. I don’t see anything to suggest that changing. What could Apple’s killer feature possibly be? My wishlist would hardly be groundbreaking: Hey Siri, play my home playlist from Spotify in my living room. Hey Siri, buy John Wick 2. Siri certainly could do these things, and likely will one day (I’m calling it now: WWDC ’18. “This is the best Siri ever.”). By the time Apple makes comfortable, incremental improvements, Alexa will be making lattes, balancing your portfolio, and ordering groceries automatically through the Echo Eat.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a8cbd2b2cc40" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Going Back to the Mainframe.]]></title>
            <link>https://medium.com/futurework/going-back-to-the-mainframe-8571111db8ea?source=rss-2a6ecf601bae------2</link>
            <guid isPermaLink="false">https://medium.com/p/8571111db8ea</guid>
            <category><![CDATA[microsoft]]></category>
            <category><![CDATA[futurism]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[technology]]></category>
            <dc:creator><![CDATA[Tim Edwards]]></dc:creator>
            <pubDate>Mon, 05 Jun 2017 16:51:52 GMT</pubDate>
            <atom:updated>2017-06-12T13:54:01.520Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/585/1*eoGdcVmGwI6UN3l7m61Egg.png" /></figure><p>Decades ago, a young Bill Gates fought for access to code on a time-shared mainframe. A personal computer didn’t yet exist. Not for a lack of vision. Computing was far too expensive, and had little utility to offer the average family. When Gates graduated from Lakeside School in 1973, ‘personal computers’ were boards you built yourself after buying an Intel 8008 chip, and ran simple programs that you wrote yourself. Only a small (but quickly growing) group of hobbyists were involved in personal computing. The academic and business world fed their calculation needs to consoles with time purchased or reserved on a powerful processing machine.</p><blockquote><strong>“There is no reason anyone would want a computer in their home.”</strong></blockquote><blockquote><em>Ken Olsen, founder of DEC, 1977</em></blockquote><p>Perhaps the mistake Ken Olson made was not being more specific. Home computing was still a hobbyist domain. When PC power rose, and modern operating systems began to take form, the rest was history, and the mainframes gathered dust. Today, remote and local servers serve our PCs with fresh data through services and hardware made to run continuously, but they more closely resemble PCs with specifically robust hardware than the mainframes that came before them. At least they did.</p><p>With the rise in virtualization, containerization, and powerful distributed server networks from Google, Amazon, and Microsoft, more and more of our computing life is moving to the cloud. The PC is still ubiquitous, but sales of tablets, laptops, and desktops peaked in 2014. Our computing is moving to the cloud.</p><h4>Virtualization of the Personal Computer</h4><p>The rise in core-count and GPU power in servers has lowered the bar for even small businesses to virtualize their workstations and move computing to a single server system. A single server can now power over a 100 virtual desktops. These businesses can use cheap hardware known as thin clients to automatically connect users to their virtual hardware. The flexibility allows you to use multiple devices to connect to your ‘personal computer’. Your phone, your personal laptop, a hotel computer: All can fire up your virtual PC.</p><p>When you combine the rise of server networks like AWS and Azure, with the rize of PC virtualization, I see an inevitable rise of ‘desktop as a service’. Microsoft is uniquely positioned in this market. Owning the most popular operating system, Windows, and a rapidly growing server network, Azure, there is little in the way to renting your computer. They can even create a beautiful Surface thin client. In the not distant future, $20 a month and a small stick similar to a Chromecast, will connect you to your PC offering limitless hard disk space and a basic level of computing power. Like to play games? It will seem silly to buy a $500 graphics card when you can just spend an extra $15 a month and have graphics power added to your PC with a mouse click. Working on a project that requires intense calculations? Upgrade your computer for only as long as you need it. Amazon lets you rent a 64 core, 256gb RAM, 8 GPU server by the hour. A system that easily costs over $100,000 can be had for a few dollars per hour.</p><h4>Renting time from a powerful central server? Sounds familiar.</h4><p>As networks grow and become more reliable, more and more computing moves off-site. We are moving back to the mainframe model, but now it’s a personal mainframe. It isn’t Bell calculating account statements, it’s Google finding people in your photos. It’s Apple playing you music. The Chromebook comes close. The downside is that many people aren’t ready to live entirely on the web. Many need applications installed locally and I don’t see this changing soon. iTunes, Photoshop, video games and more require having a local file system to use. Having a cloud desktop is the answer.</p><p>Some of the biggest innovations I see in the next 5 years will be the creation and retooling of protocols and infrastructure to accommodate this change. Hardware manufacturers and vendors will struggle and resist, but we’ve seen this before. You can’t argue with the market trends. Price, ease, and reliability will win. If you would have told me a few years ago that I would love renting Office versus buying perpetual licenses, I wouldn’t believe you. After seeing the ease of management, reliability, and growing features, I’m a believer. Now, I want to rent the rest of my computing as well.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8571111db8ea" width="1" height="1" alt=""><hr><p><a href="https://medium.com/futurework/going-back-to-the-mainframe-8571111db8ea">Going Back to the Mainframe.</a> was originally published in <a href="https://medium.com/futurework">FutureWork</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>