- Jul
- 18
- 2008
Is Twitter Responsible For Rails’ Image?
By: Robert Dempsey | Tags:Personally, I think not, however let’s take a look at the subject.
Who are the people in an organization that decide which programming languages and frameworks to use? Is this a new company? Do they have infrastructure and custom applications in production? What magazines, blogs, and reports do they read? Who do they talk to? How much do any of these people know about tech, or more specifically, programming? What experiences have they had in the past? Let’s dig a bit deeper.
Purely Business
A CEO or other non-IT executive in a company probably doesn’t care, or know, what programming language or web framework was used to develop the applications they use. A quick search of the Harvard Business Review came up with zero results for “ruby on rails,” and the Wall Street Journal says that Rails is used by “boutique outfits.” While the WSJ isn’t correct (we have many enterprise-level clients using Rails), a CEO might pay attention, giving it a second thought only when their CIO says, “and we’re going to build it in Rails.” While the CFO and COO probably don’t use Twitter, the VP of Marketing might, however I doubt that they know that Twitter uses Rails, or that much of Twitter doesn’t use Rails at all. Conclusion: most non-IT C-level executives likely don’t use Twitter, and don’t know about Rails.
IT Executives
The main industry rags and sites for IT executives include ZDNet, eWeek, and InformationWeek. ZDNet had a great article on LinkedIn using Rails for a Facebook app that serves 1B pageviews/month, which is on the first page of “ruby on rails” search results, along with a number of articles saying that Rails can scale and to not use Twitter as the benchmark. eWeek’s results discuss startups working to improve the scalability of large Rails applications and how companies scale internally. The first page of Information Week search results don’t mention Twitter at all. TwitDir tells me that there are 1610 Twitter users (of the 2,136,236 that it knows about) that have “CIO” in their description, 734 with “CEO,” and 6223 with “COO.” Compare that to the 59,764 with “IT.” A quick check of Google for “rails scaling” returns 6,960,900 results, the first page of which mentions Twitter two times, both of which are positive. A search for “ruby scaling twitter” returns a mere 429,000, two of which are from TechCrunch. Conclusion: few IT Executives are using Twitter, or attribute it’s scaling issues to Rails.
Developers and other IT Folk
I’ll come out and say it plainly - in the development world, the debate of what text editor you chose, the programming language to use, and the web framework you want, can all turn into religious wars. Those who are not fans of Rails have claimed that Rails can’t scale and Twitter is proof. Those who are Rails fans say that Rails can scale, and that Twitter is an example, among many. I’m not here to flame anyone, so I’ll blow out that match.
Conclusion
To reiterate my opinion on the subject, Rails can definitely scale if you know what you are doing. And Twitter and its downtime? Twitter is not the benchmark.
What do you think? Does Rails have a negative image due to Twitter’s issues? Or, is all it just meaningless noise?
Keep up with everything going on with Atlantic Dominion Solutions by subscribing to our monthly newsletter.
Share this post
Related Posts
You can leave a response, or trackback from your own site.
11 Responses to “Is Twitter Responsible For Rails’ Image?”
On July 18th, 2008 at 10:03 am Derek Gallo said:
I think when it was new Twitter gave off that impression because the app was still new and people didn’t have its history to judge from so they just assumed that a widely used application must be backed by solid developers and architects.
Any developer or IT professional who is worth anything should be able to realize by now that it must be Twitter’s architecture rather than Rails. I am not a Rails developer but any competent team should be able to work out their problems after this long, regardless of platform. I can’t think of hardly any new significant features they have put out since their launch and with 12+ engineers they should have been able to rewrite the whole thing by now.
On July 18th, 2008 at 11:14 am G B Hoyt said:
I think the most impotant factor to consider in any new technology is not scalibility. That’s asking a question for a time when servers had had set physical limits. As virtualiation becomes more mainstream, the old ruby maxim “computers are getting faster, people aren’t” becomes a bigger factor in application development than scalibility. To keep things on track, a better question to ask in regard to twitter being a rails app is “Does rails allow twitter to develop something different and innovative?”
On July 18th, 2008 at 11:23 am Adam Fortuna said:
Anything that can be used as an association with failure usually will be by those trying to make arguments against it. Any hugely popular site that experiences downtime reflects badly on their language choice; whether justified or not. Take MySpace for example. It was running ColdFusion when it was in the top 10 trafficked sites on the web; but due to the downtime CF was marked as the reason (above all others). Seems like an argument that will just keep repeating itself with new languages.
I don’t know what the non-IT world thinks of these kinds of issues in scaling, but anyone in IT worth their salt should know there’s more to it than purely language and framework choice.
On July 18th, 2008 at 1:37 pm Tristan said:
Yes, the archirtecture for twitter isn’t appropriate, it was designed as a blog platform, but it’s really a messaging problem, not a CRUD blog.
RoR is GREAT at making CRUD blogs, havent seen anything else done that was worth using. I am starting to suspect the reason that 37signals doesn’t do new features is because anything that isnt CRUD in RoR doesnt work.
Twitter is proof of my suspicion.
Are there any good non CRUD apps written in ror that do scale?
On July 18th, 2008 at 2:15 pm Web 2.0 Announcer said:
Is Twitter Responsible For Rail’s Image?…
[...]Is Twitter Responsible For Rail’s Image? Personally, I think not, however let’s take a look at the subject.[...]…
On July 18th, 2008 at 3:29 pm abenamer said:
I think everyone would like to have Twitter’s problem. Where do I sign up?
On July 18th, 2008 at 5:26 pm Joe Grossberg said:
“Are there any good non CRUD apps written in ror that do scale?”
Yes. YellowPages.com Justin.tv Hulu.com Scribd.com
Only CRUD sites by a very liberal definition, and more traffic than Twitter.com according to http://rails100.pbwiki.com/Alexa+Rankings
Also, Robert, you have a typo in your header: it should be “… for Rails’ Image …” — i.e. with the apostrophe after the “s”.
On July 18th, 2008 at 9:19 pm Robert Dempsey said:
Thanks Joe - I have updated the title.
On July 19th, 2008 at 6:20 am Robert Dempsey weighs in on the F5 Ruby Scale thing « CodeHappy said:
[...] Robert’s latest blog post asks if Twitter damaged Rails reputation with all it’s downtime etc. He comes to the conclusion that it probably didn’t, but he also says that, as so many have, the Ruby scaling problem is a moot one. “ZDNet had a great article on LinkedIn using Rails for a Facebook app that serves 1B pageviews/month, which is on the first page of “ruby on rails” search results, along with a number of articles saying that Rails can scale and to not use Twitter as the benchmark.” [...]
On July 19th, 2008 at 12:13 pm Senthil Nayagam said:
for good or bad, twitter brought the attention to performance,
some of the decisions made while designing rails was taking into account deficiency in MRI Ruby(the only ruby distribution then), threading , GC, memory usage etc.
now we have more ruby runtime options than we ever had
we have covered a lot of distance from since the early days, FCGI, SCGI, mongrel and now mongrel enhanced ones(evented mongrel, swiftply) and passenger
we now have chance to undo all those non-enterprisy image issues we had. I am personally working on reviving actionwebservice and completing the unfinished agendas
On July 22nd, 2008 at 2:24 pm Thom Parkin said:
Well stated, Robert.
This argument is almost like (borrowing a very old and over-used analogy) judging a book by the method used to print it.