Comments on: Twitter goes down (again) but will it soon be counted out for good? http://www.pcpro.co.uk/blogs/2009/08/11/twitter-goes-down-again-but-will-it-soon-be-counted-out-for-good/ Blogging in the real world Sun, 12 Feb 2012 15:02:20 +0000 http://wordpress.org/?v=2.8.4 hourly 1 By: Jeff Dickey http://www.pcpro.co.uk/blogs/2009/08/11/twitter-goes-down-again-but-will-it-soon-be-counted-out-for-good/comment-page-1/#comment-100903 Jeff Dickey Thu, 13 Aug 2009 15:29:32 +0000 http://www.pcpro.co.uk/blogs/?p=6736#comment-100903 Have any outside folks ever done a reasonable audit of Twitter's infrastructure, technology and so on? I remember about a year and a half ago, when Twitter was going down more often than a Bangkok "independent businesswoman", there was a real shoutfest in some of the technical blogs that were looking askance at what Twitter had done to make Ruby on Rails scale kinda-sorta-enough. Numerous respected writers weighed in on all sides - those saying that Rails COULDN'T do what Twitter really needed to (in its current form, at least), others that Rails wasn't the problem so much as Twitter's implementation, and various other brushfires. Rails pseudodemigod DHH weighed in, having been described as berating the Twitter folks savagely for what they'd done to his 'baby'. Things died down after Twitter managed to stay up for more than a couple of days at a time, but the impression I had at the time was that there were still some very critical open questions about the entire architecture and infrastructure in the minds of (at least self-styled) experts. I don't know that a real audit could be done without spilling more of the Twit's IP into the public arena than they'd reasonably be comfortable with. But I DO believe that, if it's going to remain viable as a platform for other people to build on top of (and even sometimes sell to real, paying customers), that there needs to be several hecks of a lot more openly known than there seems to be now. Have any outside folks ever done a reasonable audit of Twitter’s infrastructure, technology and so on? I remember about a year and a half ago, when Twitter was going down more often than a Bangkok “independent businesswoman”, there was a real shoutfest in some of the technical blogs that were looking askance at what Twitter had done to make Ruby on Rails scale kinda-sorta-enough. Numerous respected writers weighed in on all sides – those saying that Rails COULDN’T do what Twitter really needed to (in its current form, at least), others that Rails wasn’t the problem so much as Twitter’s implementation, and various other brushfires. Rails pseudodemigod DHH weighed in, having been described as berating the Twitter folks savagely for what they’d done to his ‘baby’.

Things died down after Twitter managed to stay up for more than a couple of days at a time, but the impression I had at the time was that there were still some very critical open questions about the entire architecture and infrastructure in the minds of (at least self-styled) experts.

I don’t know that a real audit could be done without spilling more of the Twit’s IP into the public arena than they’d reasonably be comfortable with. But I DO believe that, if it’s going to remain viable as a platform for other people to build on top of (and even sometimes sell to real, paying customers), that there needs to be several hecks of a lot more openly known than there seems to be now.

]]>