<?xml version="1.0" encoding="utf-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>Disqus - Latest Comments for epugh</title><link>http://disqus.com/by/epugh/</link><description></description><atom:link href="http://disqus.com/epugh/comments.rss" rel="self"></atom:link><language>en</language><lastBuildDate>Fri, 12 Aug 2016 11:18:51 -0000</lastBuildDate><item><title>Re: Hearting, Liking, or Starring with Rails</title><link>https://codediode.io/lessons/1788-hearting-liking-or-starring-with-rails#comment-2834031049</link><description>&lt;p&gt;This was a really great article!  I have "stars" that are global, so didn't need the relationship stuff, but the tutorial was perfect!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Fri, 12 Aug 2016 11:18:51 -0000</pubDate></item><item><title>Re: How to access your Google Dive files with RubyDecember 17, 2013</title><link>https://www.krautcomputing.com/blog/2013/12/17/how-to-access-your-google-drive-files-with-ruby/#comment-2417108375</link><description>&lt;p&gt;Really appreciate the article, I'm building our internal dashboard "Thats Metricable", and a lot of the data I need is tied up in various Google Spreadsheets and other documents.    I am using OmniAuth for Google Logins, so I already have my authorization token, so I just do:  client.authorization.access_token = session[:google_token], and it works like a champ!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Sat, 19 Dec 2015 07:05:23 -0000</pubDate></item><item><title>Re: OSCON 2013: A Review</title><link>https://www.rodhilton.com/2013/07/30/oscon-2013-a-review/#comment-1068534918</link><description>&lt;p&gt;I attended/spoke at OSCON for about 5 years in a row, and when it moved away from Portland, kind of lost track of it as a conference, even when it moved back.  However, reading your article, that puts it back on my list for conferences next year!   I'll start looking for the CfP for speakers!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Thu, 03 Oct 2013 07:31:19 -0000</pubDate></item><item><title>Re: Yes, You Can Make Money with Open Source</title><link>http://blogs.hbr.org/cs/2013/01/yes_you_can_make_money_with_op.html#comment-770866979</link><description>&lt;p&gt;If you are going to build a business on open source, you need to A) make sure that open source solution is very wide spread in adoption.  Only 1 in 1000 (or 10,000!) companies will ever look for paid support of any kind.  B) make sure that if you are packaging up a open source solution (a la ActiveState) that the value-add is worth the money you are charging.  Either you've built in fantastic proprietary features, or you've done the hard work of gluing together disparate pieces together.  You need to be head and shoulders better then the free solution.  C) Make sure you aren't directly competing with the open source free solution in your offering, as you'll always lose out in the long run to "free".  &lt;/p&gt;&lt;p&gt;I see two viable paths: the go big model of providing certified distributions taken by Red Hat, Acquia, Cloudera, etc, and the solid reasonable growth rate consulting firm providing paid consulting services, like my own firm OpenSource Connections.   Trying to both sell a distribution and do services is a very difficult path to tread.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Thu, 17 Jan 2013 08:28:07 -0000</pubDate></item><item><title>Re: Shalin Says..., SolrMarc vs DIHMarc</title><link>http://shal.in/post/327246264#comment-219489855</link><description>&lt;p&gt;How much of this is because SolrMarc uses lots of meta programming to avoid tying you to a specific version of SolrJ client thats matches the version of Solr?  Lots of commands like this:&lt;/p&gt;&lt;p&gt;Class&lt;br&gt;commitUpdateCommandClass =&lt;br&gt; &lt;br&gt;Class.forName("org.apache.solr.update.CommitUpdateCommand");           &lt;/p&gt;&lt;p&gt;commitUpdateCommand =&lt;br&gt;commitUpdateCommandClass&lt;br&gt; &lt;br&gt;.getConstructor(boolean.class).newInstance(false); &lt;/p&gt;&lt;p&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Mon, 06 Jun 2011 17:39:51 -0000</pubDate></item><item><title>Re: TwigKit: Why Developers Should Become UX Designers</title><link>http://www.twigkit.com/blog/2011/05/20/a-call-for-high-quality-open-source-demo-data.html#comment-207821476</link><description>&lt;p&gt; Funny, a colleague of mine is doing a demo of rapid prototyping, and was looking for the exact same set of data!  I suggested weather data as somewhat unusual, but don't yet have a great source.  &lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Fri, 20 May 2011 09:34:38 -0000</pubDate></item><item><title>Re: An open source collaborative network</title><link>http://www.findbestopensource.com/article-detail/lucene-vs-solr#comment-88877187</link><description>&lt;p&gt;While I agree with the summary at a high level, it's not a 50/50 balanced choice between Solr and Lucene.  The cost of implementing search on Lucene are MUCH higher then the costs of Solr.  I would say "To get more control for very specific reasons, and I have bucket loads of cash to pay really smart Java developers, use Lucene.  Otherwise choose Solr".  We have helped numerous clients move away from Lucene based search to Solr specifically because it is so hard to match the features in Solr when you are rolling your own in Lucene.  Twitter had 8 really smart developers do their Lucene search, and the probably doesn't cover the 20 other folks who where involved!  &lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Thu, 21 Oct 2010 15:12:14 -0000</pubDate></item><item><title>Re: Fuzzy searching in SOLR with Sunspot</title><link>http://www.pipetodevnull.com/past/2010/8/5/fuzzy_searching_in_solr_with_sunspot/#comment-66199171</link><description>&lt;p&gt;Out of curiosity, do you have the lowercase filter defined in your schema.xml for queries?  That should work around the case issue that you had???&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Thu, 05 Aug 2010 09:53:33 -0000</pubDate></item><item><title>Re: Test your web browser for WebSocket support</title><link>https://jimbergman.net/websocket-web-browser-test/#comment-50329293</link><description>&lt;p&gt;I am using Safari 4.0.5 on Snow Leopard and websockets work (tested via using &lt;a href="http://PusherApp.com" rel="nofollow noopener" target="_blank" title="PusherApp.com"&gt;PusherApp.com&lt;/a&gt;), but this script reports that it doesn't work...   Chrome did report back that it worked.  I am looking for some JS that would check if websocket support was enabled.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Fri, 14 May 2010 07:47:37 -0000</pubDate></item><item><title>Re: Rails on XML, Part 1: Background</title><link>http://alpinegizmo.com/2009/01/03/rails-on-xml-part-1-background.html#comment-24604928</link><description>&lt;p&gt;We worked with eXistDB on a site that was powered using these set of technologies, and I would really have loved to have built the frontend in something more powerful, like RoR, and continued to use eXistDB as the XML database.  I really like eXistDB as a DB, but as a web dev framework, well XSLT and XQuery for logic parsing are too clumsy to render a pile o' html!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Wed, 02 Dec 2009 11:41:10 -0000</pubDate></item><item><title>Re: Dan Manges - CruiseControl.rb Growl Notifier Plugin</title><link>http://www.dan-manges.com/blog/25#comment-20658425</link><description>&lt;p&gt;Really love this plugin..  One thing was I had to run "gem install ruby-growl", which wasn't obvious from the install directions!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Tue, 20 Oct 2009 18:54:44 -0000</pubDate></item><item><title>Re: paperplanes. The NoSQL Dilemma (with a Happy Ending)</title><link>http://www.paperplanes.de/2009/9/8/the_nosql_dilemma.html#comment-16418054</link><description>&lt;p&gt;Do you have any sense of a successor to SQL that works across NoSQL data stores?   One of the nice things about SQL is that is more or less works on every RDBMS.  But in your post you mentioned I would have to learn either proprietary query languages.  If JQL worked across all the NoSQL data stores, then you could start building reporting tools that work with more then one NoSQL Data store!&lt;/p&gt;&lt;p&gt;On a somewhat related note, where do you think Solr and the use of Lucene search queries fits into the SQL/NoSQL world?   Solr has a schema, but you don't have to fill every column.  And can define them on the fly.  &lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Fri, 11 Sep 2009 09:28:40 -0000</pubDate></item><item><title>Re: File Uploads with Selenium</title><link>http://www.opensourceconnections.com/2007/06/06/file-uploads-with-selenium/#comment-21700035</link><description>&lt;p&gt;Well, can you hack around it?  In other words, can your remote server do something like \\someip\mydesktop\testfile.txt?   You are right that it's not a perfect solution, and adds a lot of brittleness.&lt;/p&gt;&lt;p&gt;I would argue however that your Selenium tests if they are being run in any sort of regular way should be using the same file over and over, and not relying on your local desktop anyway!   If you are running Selenium under any kind of Continous Integration environment, having a dependency on a local file will cause all sorts of nightmares!&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Thu, 16 Aug 2007 16:34:54 -0000</pubDate></item><item><title>Re: Eric Pugh To Present At VPTC 2007 Trends Seminar</title><link>http://www.opensourceconnections.com/2006/08/04/eric-pugh-to-present-at-vptc-2007-trends-seminar/#comment-21699935</link><description>&lt;p&gt;Will,&lt;/p&gt;&lt;p&gt;I found this page from [Ruby Cookbook](&lt;a href="http://www.crummy.com/writing/RubyCookbook/test_results/19041.html)" rel="nofollow noopener" target="_blank" title="http://www.crummy.com/writing/RubyCookbook/test_results/19041.html)"&gt;http://www.crummy.com/writi...&lt;/a&gt; that runs unit tests for code in the Ruby Cookbook.  I found a .is_a?(Date) command, that I think maybe was mixed into the classes.  Not sure if `is_a?()` is an actual standard method.   I haven't had a chance to dig furthur into it though.  Here seems maybe to be more about it [online](&lt;a href="http://codeidol.com/other/rubyckbk/Date-and-Time/)" rel="nofollow noopener" target="_blank" title="http://codeidol.com/other/rubyckbk/Date-and-Time/)"&gt;http://codeidol.com/other/r...&lt;/a&gt;.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Eric Pugh</dc:creator><pubDate>Tue, 24 Apr 2007 11:49:01 -0000</pubDate></item></channel></rss>