<?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 Mikael Sundberg on Medium]]></title>
        <description><![CDATA[Stories by Mikael Sundberg on Medium]]></description>
        <link>https://medium.com/@msundb?source=rss-72ba2c10f9df------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*nDX0woYpvTHLDdMq.jpg</url>
            <title>Stories by Mikael Sundberg on Medium</title>
            <link>https://medium.com/@msundb?source=rss-72ba2c10f9df------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 16:14:21 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@msundb/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[Learnings from taming IE]]></title>
            <link>https://medium.com/@msundb/learnings-from-taming-ie-3c2f518f2d2f?source=rss-72ba2c10f9df------2</link>
            <guid isPermaLink="false">https://medium.com/p/3c2f518f2d2f</guid>
            <category><![CDATA[vuejs]]></category>
            <category><![CDATA[javascript]]></category>
            <category><![CDATA[internet-explorer]]></category>
            <category><![CDATA[material-design]]></category>
            <dc:creator><![CDATA[Mikael Sundberg]]></dc:creator>
            <pubDate>Wed, 22 Nov 2017 08:17:53 GMT</pubDate>
            <atom:updated>2017-11-22T08:17:53.953Z</atom:updated>
            <content:encoded><![CDATA[<p>We recently decided it was time to make our web app to work with Internet explorer 11 and Edge. When we started, no page loaded at all. And our knowledge of Babel, webpack etc was very limited.</p><p>Our frontend stack mainly uses <a href="https://vuejs.org/">Vue.Js</a> and <a href="https://vuematerial.io/">vue material</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZxheR_x-dHq0B28BR1FKHw.jpeg" /></figure><p>To verify things work, we used virtual machines from modern.ie. After spending way too long with the windows 10 version of the ie11 image, we realized windows has disabled changing localhost in the hostfile. So if you need to change localhost to point to your computer instead of the virtual machine, pick the windows 7 version of the image. To verify that edge works, you can either use another virtual machine or use Browserstack, that provides free testing with Edge.</p><p>For a lot of things internet explorer doesn’t support, there are plugins called polyfills you can add to webpack. These do magic so you dont have to think about it. we had to add two polyfills to most of our frontend projects.</p><blockquote>bundle: [‘url-search-params-polyfill’, ‘babel-polyfill’]</blockquote><p>We have a library that we fetch from git using npm, we had to add that to babels resolve path for it to be passed through babel in each app:</p><blockquote>include: [resolve(‘src’), resolve(‘test’), resolve(‘node_modules/ui-apps-common’)]</blockquote><p>I have a feeling this isn’t the best way to solve the problem, but it works. I assume you should babel your projects and publish them to a registry instead. But this was very little effort and we didn’t see any reasons to do a bigger change.<br>IE seem to be more aggressive in caching then other browsers, we dont want or need any caching in our api calls, so we decided to add all the headers:</p><blockquote>headers.put(tryFromString(“Access-Control-Max-Age”), “0”);<br>headers.put(tryFromString(“Cache-Control”), “max-age=0,no-cache,no . store,post-check=0,pre-check=0,must-revalidate”);<br> headers.put(tryFromString(“cacheSeconds”), “0”);<br> headers.put(tryFromString(“Expires”), “Mon, 23 Aug 1982 12:00:00 GMT”);<br> headers.put(tryFromString(“Pragma”), “no-cache”);<br> headers.put(tryFromString(“useCacheControlHeader”), “true”);<br> headers.put(tryFromString(“useCacheControlNoStore”), “true”);<br> headers.put(tryFromString(“useExpiresHeader”), “true”);</blockquote><p>We use flexboxgrid, it has support to Align content at the end of a row. This, however, doesnt work in ie, so you have to use text-align instead, see <a href="https://github.com/kristoferjoseph/flexboxgrid/issues/146">https://github.com/kristoferjoseph/flexboxgrid/issues/146</a>.<br>For change events in a medium type editor, we use, we used the value change.event.srcElement.innerHTML to get the HTML. That property doesn’t exist in ie 11, so we changed to change.event.target.innerHTML that seem to exist in all browsers.</p><p>In some cases, we used to have a margin set to align a menu with content. That had to be replaced with padding.<br>We have a bar fixed at the bottom of the screen in some views, we had to add position: fixed for it to stay fixed when scrolling. I feel the mystery here is more how it worked without it in chrome.</p><p>SVG’s as &lt;img&gt; doesn’t work, so we decided to use pngs instead. It is possible to use &lt;svg&gt; but it wasn’t worth the work in our case.</p><p>Overall, it was less trouble then we expected to make everything work. The biggest problem was probably our lack of knowledge of the builds we use. In the end, we at least have some more knowledge, but I still think we have quite a lot to learn there.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yTiAK-BurYI66qAOZjPiAw.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3c2f518f2d2f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Continuous Manual Deployment]]></title>
            <link>https://medium.com/@msundb/continuous-manual-deployment-74a3471b1a06?source=rss-72ba2c10f9df------2</link>
            <guid isPermaLink="false">https://medium.com/p/74a3471b1a06</guid>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[continuous-integration]]></category>
            <category><![CDATA[continuous-delivery]]></category>
            <category><![CDATA[agile]]></category>
            <dc:creator><![CDATA[Mikael Sundberg]]></dc:creator>
            <pubDate>Thu, 14 Sep 2017 15:01:26 GMT</pubDate>
            <atom:updated>2017-09-14T15:01:26.547Z</atom:updated>
            <content:encoded><![CDATA[<p>Continuous Manual Deployment</p><p>Continuous Deployment and Continuous Integration have been hot topics for a while now. And for me they have always been almost synonymous with using something like Jenkins. When you push code, something should pick up that commit, run al the tests and then deploy to production.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ynWeH1Ob5npYX7XzZfoTtg.jpeg" /></figure><p>But do we actually need Jenkins or any other form of pipelines? What problem does it solve? For me, it solves that deploying and running tests is hard, and should therefore be done often. So we let Jenkins do it so we don’t have to think about it. If we have branches we want to run tests on them so we know everything is green when we review the pull request.</p><p>But what if, deploying is extremely simple, takes 30s and is already done often? What if running all tests is always done anyway? What if you don’t have any branches?</p><p>The setup we have at work is very simple. We write small services, we write unit tests and we deploy to Kubernetes. We almost always do mob programming, and in the cases when we don’t mob we instead do review before commit. So there are no branches.</p><p>So what problem would Jenkins solve for us? Here are some i can think of:</p><ul><li>Get rid of “works on my machine” symptoms, we havent noticed any problems with that yet though.</li><li>A natural place to run end to end system tests. But we dont have any tests except unit tests. And i think canary releases and good monitoring is a better tool for the same job.</li><li>It forces a deploy for every push. And since we do that anyway, i dont think its worth introducing a tool for.</li><li>It forces you to always write deployable code. You cant change multiple services and then do a synchronised deploy. Im not sure if i think its worth to be able to skip complexity sometimes, at the cost of introducing coupling between versions of services.</li><li>You dont have to remember to run tests, deploy etc. But i think mob programming limits that problem alot.</li></ul><p>What we have to think about before adding Jenkins to the work flow, is that it(or any other similar system) comes with a cost. The problem it solves has to be worth the cost. Today, i don’t think it is. Someday it might be though, its always important to re-evaluate everything continuously.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=74a3471b1a06" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A year of mob programming]]></title>
            <link>https://engineering.klarna.com/a-year-of-mob-programming-e8cc7543ac2?source=rss-72ba2c10f9df------2</link>
            <guid isPermaLink="false">https://medium.com/p/e8cc7543ac2</guid>
            <category><![CDATA[mob-programming]]></category>
            <category><![CDATA[agile]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Mikael Sundberg]]></dc:creator>
            <pubDate>Tue, 23 May 2017 15:01:03 GMT</pubDate>
            <atom:updated>2017-05-23T15:01:03.559Z</atom:updated>
            <content:encoded><![CDATA[<p>About a year ago, after JFokus 2015. My team had a retro where we talked about our favorite sessions — what we could learn and if we could try to improve something in our daily work from each session. I remember we decided to try <a href="https://www.youtube.com/watch?v=9yG1c9Crnbk">mutation testing</a>, but we felt the investment wasn’t worth it. We did the same for a couple of other things but I honestly can’t remember what they were. The one thing that changed everything though was Woody Zuill’s <a href="https://www.youtube.com/watch?v=8cy64qkgTyI">talk on Mob programing</a>. We had done one or two 1–2 hour session of mob programming before, where we had simply booked a room and solved a hard problem together. Since we had liked that we thought “why not?”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RAXCUtKkZrtu22R7J8aa3Q.jpeg" /><figcaption>Example of a mob setup. Blue shirt not mandatory.</figcaption></figure><p>The next day we took a computer, connected it to a TV and started coding on a sofa. We didn’t do any preparations so we took what we had. We used a phone as a timer, decided on a rotation time and just went with it. Our communication was horrible, we discussed small things for way too long and didn’t produce at a very good pace. We all had a lot of fun that day though, and we felt we learned a lot. When I left the office that day I was totally exhausted.</p><p>The next day we decided to try it for at least a week before we drew any conclusions.</p><p>After the first week we had not improved our mob programing skills much, but we had aligned on the small things, didn’t need as much discussion, and started producing at a good pace again.</p><p>Fast-forward a year and our workdays now center around mob programming. The kanban board mostly has one or two stories in progress. We do have a stand up but it is less about sharing what we did and more about if our current story is going well, if we should slice it, etc. After the standup we decide if we should have something in addition to the mob and highlight if we have any meetings. Most days though, the mob goes on the entire day. It could be that everyone has meetings to attend, interviews to hold or personal errands to do. But the mob always goes on. So no stories are halted for long, and we produce at a constant pace.</p><p>We often get the question if we think it’s better to use mob programming than pairing or working alone. I must agree with what Woody says — for us it is way better, but i’m not claiming it’s better for everyone. We’ve found that we are much less likely to deviate from our agreed upon processes when there are more of us in the mob to point out that we’re straying from our working-agreement. Four people are more likely to remember and point out that we should verify the acceptance criteria before claiming something is done, e.g that we should have a test before writing production code. It is very unlikely the mob will take shortcuts and ignore the Sonar warnings.</p><p>Another question we often get is if we mob on small simple things. That I would say varies. But what we have noted is that we are more eager to automate things when we sit in the mob. That has led to not having as much small simple things.</p><p>Overall, doing mob programming for a year has been exciting. There has been times when it felt like a waste of time, there have been times when you learned more than ever. New members in the team this year has gone from scared and skeptical to true believers. But one things is for sure, for us mob programming was great. If it will be for you I have no idea, but why not try a week or two? What do you have to lose?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y_LcCQAzCL_blgtxhVmACQ.jpeg" /><figcaption>Everyone can do mob programming!</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e8cc7543ac2" width="1" height="1" alt=""><hr><p><a href="https://engineering.klarna.com/a-year-of-mob-programming-e8cc7543ac2">A year of mob programming</a> was originally published in <a href="https://engineering.klarna.com/">Klarna Engineering</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>