<?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 Georgy Grigoryev on Medium]]></title>
        <description><![CDATA[Stories by Georgy Grigoryev on Medium]]></description>
        <link>https://medium.com/@iamkarlson?source=rss-453d18cfad74------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*tY7gKvVt1TTx9P3V.</url>
            <title>Stories by Georgy Grigoryev on Medium</title>
            <link>https://medium.com/@iamkarlson?source=rss-453d18cfad74------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 19:22:08 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@iamkarlson/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[How is it to lose luggage in the Russian airport ]]></title>
            <link>https://medium.com/@iamkarlson/how-is-it-to-lose-luggage-in-the-russian-airport-5efd17958b75?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/5efd17958b75</guid>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Sat, 24 Jun 2017 22:07:48 GMT</pubDate>
            <atom:updated>2017-08-30T20:26:47.678Z</atom:updated>
            <content:encoded><![CDATA[<h3>How is it to lose luggage in the Russian airport</h3><p><em>Disclaimer: this post is a bit funny and, please, do not count it very serious.</em></p><p>Just to be clear — I’m not a proud citizen of Russia and I complain a lot. Usually, I complain about lots of Russian specific customs and features. Alcoholism, rednecks, and stupid laws and rules are my <em>favourites</em>.</p><p>So my wife and I lost a luggage. Maybe it more likely that I am alone responsible for the lost. We were in the duty-free shop, buying some stuff for kids. I mention the duty-free shop to let you understand that we already crossed the border. Those kids started creating a mess around. Like that mess when you prefer to run away from the place where you are. Just now. Run away. So did we. Then we had some coffee and snacks, bought a sunscreen lotion, and were waiting for the flight. And apparently discovered that our bag is lost. We came back to the same shop. They told us that they had called security already. Ok, no problem. We went to information table for information about security and our bag. We got this information.</p><ol><li>They came with sniffing dogs</li><li>Took our bag away</li><li>Crossed the border</li><li>Put in last and found</li></ol><p>Sorry, waaat??</p><p>Yes, that’s right. <strong>Crossed</strong> the border. And yes you can’t get it back without crossing the border again. Just did information, your flight is after 30 minutes. Cross the border. I asked them is it a joke. No, it’s not. Only one option which we have is to cross the border and take a boarding pass again. But registration was already closed and it’s impossible to take the pass again.</p><p>I know that we are in charge of missing the bag. However, I think there are a lot of things to improve in Russia. That procedure is one of them.</p><p>Future advice: tie your luggage to you. Even you are a sane person and think that something can happen.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5efd17958b75" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to add Let’s Encrypt to your OpenShift Wordpress deployment.]]></title>
            <link>https://medium.com/@iamkarlson/how-to-add-lets-encrypt-to-your-openshift-wordpress-deployment-899c945f7548?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/899c945f7548</guid>
            <category><![CDATA[wordpress]]></category>
            <category><![CDATA[lets-encrypt]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[certificate]]></category>
            <category><![CDATA[openshift]]></category>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Sat, 03 Jun 2017 18:17:19 GMT</pubDate>
            <atom:updated>2017-08-30T20:26:49.410Z</atom:updated>
            <content:encoded><![CDATA[<p>I haven’t been on vacation for one year. Year before we spent two weeks in a trip over the ocean and that’s all that I have for two years. Since my last vacation, I’ve tired mentally and decided to spend whole days behaving like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/169/0*EPGHDqSrDY-P8DVj.jpg" /></figure><p>So, while I was on my digital vacation without laptops and PCs, SSL certificate on my site got expired. Almost. I made a reminder in my calendar to renew but didn’t leave any details how to do that. After terrible night without sleep, I started remembering how to renew certificate on <a href="https://letsencrypt.org/">Let’s Encrypt</a>. When I created SSL on my site, I figured out how to create required files on the openshift’s node from a third or fourth attempt. These steps are pretty simple if you know the structure of the deployment on openshift. &gt;_&lt;</p><p>But if <strong>you don’t</strong> this is a stupid stepping one by one over and over again. For example, wordpress deployment is located at following path:<br><br>/var/lib/openshift/%_YOUR_APP_ID_%/app-root/repo/php</p><p>Therefore, when you stepped down to this directory you can do all that LetsEncrypt’s <a href="https://certbot.eff.org/">certbot</a> asks you to do.</p><p>By the way, don’t make anything that you can’t debug tomorrow.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*k43OH536J-fpG9yh.jpg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=899c945f7548" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Get-Shares.ps1]]></title>
            <link>https://medium.com/@iamkarlson/get-shares-ps1-8a5b9baeedff?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/8a5b9baeedff</guid>
            <category><![CDATA[powershell]]></category>
            <category><![CDATA[syncing]]></category>
            <category><![CDATA[share]]></category>
            <category><![CDATA[unc]]></category>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Thu, 25 May 2017 14:04:45 GMT</pubDate>
            <atom:updated>2017-08-30T20:27:04.407Z</atom:updated>
            <content:encoded><![CDATA[<p>So, let’s continue blogging about my PowerShell tools. I’m browsing throughout my corporate network using PowerShell mostly. The biggest problem with this approach is that when you go to the network share (by <strong>UNC</strong> path) you can’t step on <em>\\server_name</em> without a particular share. For example, you can set location to <em>\\my_server\my_share</em>, but you can’t to <em>\\my_server</em>. To find shares what I used to do was:</p><ul><li>open Explorer</li><li>copy required share name</li><li>copy share name</li><li>go to terminal</li><li>paste my share name</li></ul><p>As you can see it’s a bit annoying. I struggled a lot with that and tried to find out a native workaround, but with no success. Finally, I made coffee, sat in front of my laptop and started thinking how Explorer acquires network shares, it’s impossible if it uses some internal APIs to which you don’t have access. <strong>net view</strong>. Of course!! Oh my god! The answer was right under my nose. Therefore, I wrote the simplest script. It’s not a sophisticated solution, but it works! Next script is quite a useful shortcut to acquire server shares through <strong>net view</strong>.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/b7573caba8e445949d1072ff7af92a84/href">https://medium.com/media/b7573caba8e445949d1072ff7af92a84/href</a></iframe><p>When you sourced this script, to get shares run the following code:</p><p>Get-Shares \\my_server</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8a5b9baeedff" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Group-ByMonth.ps1]]></title>
            <link>https://medium.com/@iamkarlson/group-bymonth-ps1-93679a79581e?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/93679a79581e</guid>
            <category><![CDATA[powershell]]></category>
            <category><![CDATA[logs]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[remote-server]]></category>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Sun, 21 May 2017 14:05:53 GMT</pubDate>
            <atom:updated>2017-08-30T20:27:03.363Z</atom:updated>
            <content:encoded><![CDATA[<p>Bare PowerShell is quite odd if you’re using it on daily basis. There are no out-of-box such functions as add to environment path, get server shares, or convert to YAML. Eventually, you create a batch of scripts to regular usage. My toolset for PowerShell contains around 30 small scripts and modules. However, I remember each line of code, as my favourite post about vimrc (<a href="https://dougblack.io/words/a-good-vimrc.html">Good vimrc — Doug Black</a>) is telling us:</p><blockquote>Don’t put any lines in your vimrc that you don’t understand.</blockquote><p>So, when you are working with plenty of log files on a remote server, it takes some time to receive file list. There are two options: set a logger to put files in separated directories or to arrange these files by a script periodically when needed. I am not keen on any side, but when I need to arrange files (not only logs) I’m using following script.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/48bca4f517851bf4f9b549cb93c076ca/href">https://medium.com/media/48bca4f517851bf4f9b549cb93c076ca/href</a></iframe><p>The script creates folders for each found file group and moves them there. It works only in a passed folder without recursive.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=93679a79581e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DDL and SQL Server Transactional Replication]]></title>
            <link>https://medium.com/@iamkarlson/ddl-and-sql-server-transactional-replication-99e3517ee32b?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/99e3517ee32b</guid>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[sql]]></category>
            <category><![CDATA[olap]]></category>
            <category><![CDATA[emotionally]]></category>
            <category><![CDATA[syncing]]></category>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Thu, 11 May 2017 07:52:38 GMT</pubDate>
            <atom:updated>2017-08-30T20:27:07.389Z</atom:updated>
            <content:encoded><![CDATA[<p>Disclaimer: This post is not about detailed how-to and not about replication at Microsoft SQL Server at all, but is about my experience and faults.</p><p>Chapter 1. Never run <em>alter column</em> on replicated table (almost never).</p><p>Of course, there are many cases when you just do not have any other option. For example when you definitely know that changing from <strong>bit</strong> to <strong>smallint</strong> will take only a few minutes (<em>it’s schema change only, just one table!</em>). What is exactly happening in this case? You have OLTP server and store only monthly data there. Meanwhile, OLAP server stores all your data from the beginning. You thought about that a lot. Change bit to <em>smallint</em>. What can happen? I’m pretty sure that we’ll do it right before the end of the day and then go home. When I sat opposite my laptop in the middle of the night I had a lot of time to think about that twice.</p><p>Explanation: Yes, it took no more than 5 minutes on OLTP server, but when this transaction was sent to the OLAP, SQL Server started generating default values and updating transaction log. It took around a day to understand that something was going wrong and this just didn’t work.</p><p>Chapter 2 Choose correct type of the data</p><p>Yeah, I know that you do it all the time. I do it too. But I made a mistake. Everybody makes mistakes. In my particular example, I chose a bit type for a state property. I have records which I am checking by another source. If this another source has this record I raise a flag approved, otherwise set it to zero. Pretty simple, isn’t it? By default it is <em>null</em>. My assumption was that I have only two states of the flag: approved and not approved. And I missed <em>null</em> state. Then I added a default constraint with <em>not approved</em> value and I left only two states (one or zero). Then I was asked to add this third value. Hence I started this journey.</p><p>Advice: if you can use anything except <em>null</em> as value — use it.</p><p>Chapter 3 Re-initializing subscription is the worst option if you have OLTP-OLAP replication</p><p>The best solution is to wait until it is complete. You think about that first 24 hours. Then you start looking for other options. There are all options below which I found out:</p><ol><li>Wait until it is complete</li><li>Re-initialize subscription</li><li>Stop distribution and run the same query on the subscription database</li><li>Rename column on the subscriber and create brand new with required settings</li></ol><p>I put these options in an order which I thought how to choose them.</p><ol><li>I picked up first as the simplest one. I was waiting a lot of time when I started noticing that PID of the process was being changed each 5–6 hours. Finally, I found out what was going on. Your transaction can be interrupted and aborted and then replication tries to apply it again. And again and again. The reason can be any: timeout, server broke the session, the server was restarted (our sysadmin had a planned maintenance), anything.</li><li>I walked around, investigate a bit and understood that if you are re-initializing replication whole data on the target machine will be replaced with data from the source. I suppose there are a couple of methods how to workaround this issue. However, I decided to go on.</li><li>I ran the same query, and nothing changed. Still same enormous amount of time</li><li>I was afraid that my replication will be broken</li></ol><p>But when I tried all options I started researching.<br>Summary: try to find out as many applicable solutions as you can. Eventually, you will find out something.</p><p>Chapter 4 Rename column in the subscriber database can break your replication.<br>My investigation led me to only one promising idea. What will be if I rename a column to something like “myColumn_old” and create new but with old name? How does it work in general? I stuck to this idea.<br>When you are initializing, SQL Server creates for you following procedures for each table (an article) in the source:</p><ol><li><em>sp_MSdel_dbotable</em></li><li><em>sp_MSins_dbotable</em></li><li><em>sp_MSupd_dbotable</em></li></ol><p>When I found out that, I almost screamed. That is what I looked for! Want to go away from any delete on the OLAP? Here you are! Want to rename columns? It’s here too!<br>Replication with no-delete-on-subscriber policy was configured by my manager, and until now I haven’t known how it is configured! These procedures opened my eyes how replication does really work. It takes data from one server and runs procedures on the another. No black magic. No artificial intelligence. No undocumented APIs. Just simple as possible. And of course, after I renamed columns as I mentioned above and my transactions had flowed from publisher to subscriber immediately.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=99e3517ee32b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Programmers Assessment]]></title>
            <link>https://medium.com/@iamkarlson/programmers-assessment-ac2c4c972d89?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/ac2c4c972d89</guid>
            <category><![CDATA[cheatsheet]]></category>
            <category><![CDATA[assessment-matrix]]></category>
            <category><![CDATA[backend]]></category>
            <category><![CDATA[developing]]></category>
            <category><![CDATA[management]]></category>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Thu, 30 Mar 2017 20:37:08 GMT</pubDate>
            <atom:updated>2017-08-30T20:26:52.694Z</atom:updated>
            <content:encoded><![CDATA[<p>Programmers assessment one of the most difficult procedure in software development. How to decide if interview was good or bad? Do you ready to promotion? Do you deserve bigger salary? These questions can be answered by assessment, but the assessment is the question which is very difficult to describe only from one side, technical or social (soft skills).</p><p>At my previous job we used to have annual assessment with public available competency matrix. I was wondered when I saw it first time. Most of skill were not about programming. Usually such skills are called “soft skills”. These skills are “Communicating with teammates” or “Providing feedback”, “Leading”,etc.. Moreover, there was “an English skill coefficient”. For example, if you have Upper-intermediate or Advanced, you’re automatically get 1.2 multiplied score at the assessment despite you coding attitude. It can be due to specific company needs, I’m not sure. And yes, my employer provided all required services to increase your skills and to develop you as a better team-player including English courses.</p><p>I think each of them is very important but cannot make your the programming <strong>super-star</strong>. You must do code and do it well.</p><p>However, there were skills like “Developing code” or “Using frameworks”, but not a lot. I found out a bigger technical matrix couple days ago. You can look at it here: <a href="http://sijinjoseph.com/programmer-competency-matrix/">Programmer Competency Matrix — Sijin Joseph</a>. As far as I understand it is quite popular among developers and team leads. The matrix has skewness to technical skills, but is rather better than my previous one. It covers a lot of technical sides. However, I think communication skills is less important if we’re talking about <em>domain expert with 10 years’ experience and author of framework in Erlang</em>, in my humble opinion.</p><p>I have been arguing a lot with mates about the subject. Usualy this discussions ends with an idea to make more particular scores. What I expect from programmers assessment is to differentiate matrix to multiple types of programmers. I mean it is better to have different matrixes for</p><ul><li>Application Developer</li><li>System Developer</li><li>Kernel Developer</li><li>Backend Developer</li><li>Frontend developer</li><li>Data scients</li><li>Database developer</li><li>etc.</li></ul><p>Most of them are <strong>very</strong> different. Past years I noticed that more people do not like to take a wider range of responsibilities. They just want to do only one thing whatever it is (front,back, db). I suppose it will continue for a while until developers become blue collar workers completely. Then we will definitely have a government standard for programmers assessment.</p><p>P.S. The matrix from my previous job was the same for developers (doesn’t matter specialisation), testers, and analytics. All of the differences were managed at interviewing step with PM.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ac2c4c972d89" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[IIS Web Deploy by VSTS]]></title>
            <link>https://medium.com/@iamkarlson/iis-web-deploy-by-vsts-ca6c9c0f7623?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/ca6c9c0f7623</guid>
            <category><![CDATA[powershell]]></category>
            <category><![CDATA[devops]]></category>
            <category><![CDATA[isis]]></category>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Mon, 27 Feb 2017 06:27:41 GMT</pubDate>
            <atom:updated>2017-08-30T20:27:00.219Z</atom:updated>
            <content:encoded><![CDATA[<p>A task is simple: release my site to IIS. Web Applications can be released through several ways. Usually it is better to choose <a href="https://www.iis.net/downloads/microsoft/web-deploy">web deploy</a>. Web deploy is a good technology, but in my opinion it has a lack of simple tutorials as well as VSTS. I spent plenty time trying to set up web deploy in VSTS without any results. After couple attempts I just forgot about that and continue to deploy my apps from Visual Studio. However, yesterday I had to deploy a new app, and googled it again. What is good that I found out that the Microsoft is about to release <a href="https://www.visualstudio.com/en-us/docs/build/apps/aspnet/cd/deploy-webdeploy-iis-deploygroups">new features of VSTS to easier IIS App Deployment</a>. I got crazy and started new setup for my project. Only after few steps over I read this:</p><blockquote>This is a preview feature that is only available to some accounts in Team Services.</blockquote><p>Damn. But half of those features is already published in preview and you can use it.</p><p>Publication process in VSTS has two parts: build and release.</p><p><strong>Build</strong></p><p>There are should be only compiling and preparing steps in the build part. There are no any copying to target machine or modification actions. In other words it must be a batch of jobs which can be run as many as possible times and in the you environment nothing is changed. Web Application building to zip package when using web deploy. What is released by the Microsoft is ASP.NET build template.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*kSW_oud9U34hi0xh.png" /></figure><p>This template contains regular C# build steps but with required parameters to build deploy zip package. It should be <strong>Build Solution</strong> step with following MSBuild arguments:</p><p>/p:DeployOnBuild=true /p:WebPublishMethod=Package /p:PackageAsSingleFile=true /p:SkipInvalidConfigurations=true /p:PackageLocation=&quot;$(build.artifactstagingdirectory)\\&quot;</p><p>For builds I use a local agent and a local file share. You can set up <strong>Copy and Publish Build Artifacts</strong> to copy zip package to share. My share address is $(sharename)\$(Build.DefinitionName)<br>where $(sharename) = &quot;\\my\share&quot; and $(Build.DefinitionName) is built-in variable. All variables list you can find <a href="https://www.visualstudio.com/en-us/docs/build/define/variables">here</a><br>If you have only one artifact, you can trick agent and put in <strong>Artifact Name</strong> <em>$(Build.BuildNumber)</em>. In that case your builds will be in the next path:<br>\\my\share\MyBuildName\20170227.01\</p><p>That’s it. Your build is ready to deploy.</p><p><strong>Release</strong></p><p>Since our package is ready we’re behind two steps to deployed app:</p><ol><li>create and run site and app pool — <em>WinRM — IIS Web App Management</em></li><li>publish our app — <em>WinRM — IIS Web App Deployment</em></li></ol><p><strong>Step 1 — setup app pool</strong><br>Machines: our target server name<br>Admin login and pass — your credentials.<br>Http or Https switch — ssl connection can be chosen to use to publish sites in public networks. But in local network http protocol is fine. <br>Website settings group: setup your site properties.<br>Application Pool: your app pool properties.</p><p>When this properties are configured, when you release your app, you should have configured and started application app pool with application which is using this pool.</p><p><strong>Step 2 — deploy our app</strong><br>Machine — our target server. <br>Admin login and pass — your credentials.<br><em>Actually, I use variable to put it in both places.</em></p><p><strong>Deployment</strong> settings group<br>Web Deploy Package — full path to your zip package. If your are using VSTS share $(System.ArtifactsDirectory). When you are using local share, you should specify particular path to you zip package. For instance, my setting is:</p><p>$(sharename)\$(Release.Artifacts.mysite.DefinitionName)\$(Release.Artifacts.mysite.BuildNumber)\MySite.zip</p><p>You can see that I use artifact variable — <em>$(Release.Artifacts.mysite.DefinitionName)</em>. VSTS allows to use multiple artifacts as release source. Moreover, there are a lot of variables for release, list of them can be found <a href="https://www.visualstudio.com/en-us/docs/build/concepts/definitions/release/variables">here</a><br>Same settings to Parameter file:<br>Web Deploy Parameter File: <br>$(sharename)\$(Release.Artifacts.mysite.DefinitionName)\$(Release.Artifacts.mysite.BuildNumber)\MySite.SetParameters.xml</p><p>That’s all. Just push and you will get you site published.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ca6c9c0f7623" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[SIlent letters]]></title>
            <link>https://medium.com/@iamkarlson/silent-letters-10eb1d1878db?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/10eb1d1878db</guid>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Mon, 13 Feb 2017 20:34:48 GMT</pubDate>
            <atom:updated>2017-08-30T20:26:51.292Z</atom:updated>
            <content:encoded><![CDATA[<p>There are a lot of silent letters in English. I was close to say too many but list of those is not very long. Whole list you can find at many places on the Internet. I found <a href="https://www.easypacelearning.com/all-lessons/english-lessons-level-3/544-silent-letters-from-a-to-z-list-and-examples">this one</a>.</p><p>I have noticed that Russian speaking people have a specific accent with strong and loud pronunciations of some sounds. Like <em>h</em> in <em>how</em>. Russian natives usually pronounce it like <strong>kh</strong>. But what is astonishing for me is a pronunciation of <strong>calm</strong> or <strong>climb</strong>. There’s not <em>l</em> in calm and is not <em>b</em> in climb. One more word which I use to hear everyday is receipt. There is not p! OMG!</p><p>Russian language is bit simpler than English for that case. It has rule which defines how to check word spelling. In oral language you possibly won’t notice that there’s one more letter. It’s really difficult I swear. I tried with <em>счастье</em> and <em>солнце</em>. That becomes important only in writing. But receipt! The sound <strong>p</strong> changes word completely!!</p><p>That’s my third list of English words which should be just remembered.</p><ol><li>Irregular verbs</li><li>Extreme adjectives</li><li>Silent letters</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=10eb1d1878db" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Memory and reading ]]></title>
            <link>https://medium.com/@iamkarlson/memory-and-reading-f410fc04f0fd?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/f410fc04f0fd</guid>
            <category><![CDATA[books]]></category>
            <category><![CDATA[memories]]></category>
            <category><![CDATA[developing-skills]]></category>
            <category><![CDATA[reading]]></category>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Mon, 30 Jan 2017 09:39:04 GMT</pubDate>
            <atom:updated>2017-08-30T20:27:12.374Z</atom:updated>
            <content:encoded><![CDATA[<h3>Memory and reading</h3><p>Last week I learnt 2 things:</p><ol><li>Most people don’t know how to write there/their/theirs/there’s their’s in a right way</li><li>If you going to remember something after reading you should stop every paragraph, remember main points and realize these points.</li></ol><p>First thing is easy to understand. Therese very similar pronunciation for all of them and a lot of people just have never thought how to do it right.</p><p>Second one has come to me from a book about improving reading skills. It’s in Russian but I think it should be mentioned: <a href="https://play.google.com/store/books/details?id=amXeBQAAQBAJ"><strong>Павел Палагин</strong>: Скорочтение на практике: Как читать быстро и хорошо запоминать прочитанное </a>Author claims that reading skill are combined from 5 different skills.</p><ol><li>Viewing</li><li>Aim</li><li>Attention</li><li>reading</li><li>Memorizing</li></ol><p>If you can achieve better results in each of them then you will get better results in reading. For instance, memorizing texts improves your reading skill because you’re not reading same text twice. <br>This idea follows me last few days. This idea is simple. When I go step by step through book I find more and more which is very similar to my thoughts, but I couldn’t realize what they are. I hope that author doesn’t mind that I put the most important points from book to my posts.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f410fc04f0fd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[dotfiles repository]]></title>
            <link>https://medium.com/@iamkarlson/dotfiles-repository-abd27ff4eb52?source=rss-453d18cfad74------2</link>
            <guid isPermaLink="false">https://medium.com/p/abd27ff4eb52</guid>
            <category><![CDATA[developing]]></category>
            <category><![CDATA[syncing]]></category>
            <category><![CDATA[powershell]]></category>
            <category><![CDATA[emotionally]]></category>
            <category><![CDATA[github]]></category>
            <dc:creator><![CDATA[Georgy Grigoryev]]></dc:creator>
            <pubDate>Wed, 25 Jan 2017 07:10:12 GMT</pubDate>
            <atom:updated>2017-08-30T20:26:54.879Z</atom:updated>
            <content:encoded><![CDATA[<p>Since I started using vim, I’ve found that too many software developers store their configuration files at <a href="https://github.com/">github</a>. It really awesome. Last year I wrote a post at <a href="https://habrahabr.ru/post/278757/">habrahabr</a> (russian) how do I synchronize settings around multiple PCs or in other words my bootstrap procedure for new PC. It wasn’t so popular at habrahabr. I think that’s because I didn’t know about storing configuration files at github but many other people did. <br>What is a purpose of storing configuration at github? You can easily set up your new PC just by checkout repo and run init script. For example, I make <a href="https://github.com/iamkarlson/dotfiles/blob/master/vim/createLinks.ps1">simple script</a> to create symlinks to ~\.vimrc</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/953c68d2330a13259eaa6cdba4e9c635/href">https://medium.com/media/953c68d2330a13259eaa6cdba4e9c635/href</a></iframe><p>Now, I sync most of my setting still by Dropbox folder, but I going to move these settings to github.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=abd27ff4eb52" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>