Dan Cameron - Latest Comments in Twitter: “Back to Normal”

Re: Twitter: “Back to Normal”

JaredB — Tue, 27 May 2008 19:24:10 -0000

The point is that it's ridiculous for anyone (from an "outisider's perspective") to try and talk about where the blame lies, because they don't know.

Anyone who knows enough about Rails (or software dev in general) to actually be capable of making those kind of calls would be aware of how little they know about the specific nature of Twitter's problems.

So, what you're left with is a bunch of people unqualified to comment on the issue but who (for some reason) feel the need to attempt to make themselves look "smart" by proclaiming that *they* know what the problem is. Gee, if only Twitter would subscribe to their blogs / read their Tweets, etc. this whole thing could be resolved so easily!

I am curious to hear about the "insider" comments you refer to though. Links? One that I saw just this morning (right before I posted my earlier comment, which is why it was fresh in my mind) was this one from Dick Wall, who works at Google and has talked with and seen the inner workings to a degree that most of the people trying to speculate have not.

I do understand the appeal and interest in the subject, and how it leads lots of people to theorize about what they *think* the problems could be, it just gets annoying when you see it so much, and when people talk about it as if it's an easy problem to solve and/or attribute it to one specific aspect (like the choice of language / platform) without any evidence at all.

Re: Twitter: “Back to Normal”

Dan Cameron — Tue, 27 May 2008 16:13:39 -0000

It's easier to blame the architecture than the architect because it's less personal.

In this case it's the only thing to fall back on, from an outsiders perspective, because they definitely have the money to build their app to scale and to add more infrastructure.

So, when you think of it that way it's easy to blame Rails even if Twitter won't admit it because of their loyalty.

Also, there has been plenty of "insider" comments attributed Rails to their problems.

Re: Twitter: “Back to Normal”

JaredB — Tue, 27 May 2008 13:28:22 -0000

Pretty sure that last comment was spam, but I'll respond anyway... :-)

While it's fun to play "armchair software architect", the fact is that kind of speculation is purely hypothetical since you and I haven't seen the details of how Twitter is built.

It's a common theme from lots of people who *aren't* familiar with the inner workings / design of Twitter to post about how Ruby (or, more specifically Rails) is the root cause of Twitter's ongoing issues.

However, from the relatively few external (non-Twitter) people who actually have been privy to those inner workings and designs, I consistently hear that that is not the case, and (additionally) that Rails forms a relatively small part of the overall Twitter picture at this point (and for a while now).

Re: Twitter: “Back to Normal”

theamoeba — Tue, 27 May 2008 03:55:32 -0000

i think that twitter is running on a poor core (ie: Ruby). I think that if they rewrote their core on something like php or python they wouldnt have as many issues.

Re: Twitter: “Back to Normal”

JaredB — Tue, 20 May 2008 19:49:11 -0000

Yeah, what's even more interesting is the fact that everyone is still using them. Probably a combination of there being no good competition and the fact that everyone (or so many) are already on Twitter and will be reluctant to change even if something better did come along.