Newest Content
Latest NYC Coverage
NY Web Tech Directory
Most Popular Content
Press Releases
New Interviews
Recent comments
- Re: Amtrak's Express 2107 Microblogging Express Service (video)
1 hour 42 min ago - Re: Amtrak's Express 2107 Microblogging Express Service (video)
2 hours 40 min ago - Re: Amtrak's Express 2107 Microblogging Express Service (video)
2 hours 56 min ago - Re: Photo Sharing Site Photwo Goes Up For Auction
7 hours 16 min ago - Re: Photo Sharing Site Photwo Goes Up For Auction
8 hours 41 min ago
Twitter Might Not Need a Business Model But Apparently They Need More Servers
Written by Allen Stern - January 15, 2008
Earlier this month we wrote about Twitter and their lack of a business model. Some believed that they don't need a business model at this point, just a massive userbase. People are beginning to rely on Twitter (which is a good thing for them) but will massive outages on key dates cause users to look elsewhere?
For gadget bloggers, today is the super bowl. It's the day that they workout using the Wii for all year long. They don't sleep, they eat lots of pasta, no beer, etc. Several key bloggers noted that they would be using Twitter for live coverage of the JobsNote. But unfortunately, Twitter died. And even now, after JobsNote is over, I still get this message, "500 Internal Server Error".
Nicholas at Valleywag has some of the info from their perspective on the outage. Adam has some additional info over on Mashable. An official note from Twitter has been posted, "Twitter is currently experiencing some slowness related to the massive number of updates around Steve Job's keynote at MacWorld. We're working on it and will keep you updated!"
If a massive userbase is the key to success for Twitter, unfortunately they just failed. Guys, you might want to make sure the servers can handle SXSW, cause all the people suffering from Twitterdiction will be there!
Our Partners
I definitely agree with you Allen, I really had high hopes for the twitter experience during the keynote, but instead I was really frustrated trying to refresh it and see it if works, and ended up writing a rant post against twitter http://technozzle.com/?p=40
I never get why Twitter scales so horrible. Is that Rails? (Sorry for that! :-))
But all in all, I wonder with Services like Amazon's ec2 or Joyent - how can you not build an app that scales easily with more hardware.
There are so many startups doing this and they easily survive massive traffic and most of them even share all the goodness on their blog, etc..
I wonder what exactly the problem is for Twitter. Like where is the bottleneck at - I'm curious.
I suspect that the answer is simply that Twitter, by design, has a far higher write to read ratio. Compare it to a blog site; it encourages writing, but not to the degree that Twitter does. I wish I had stats, but I expect that their push design (send to IM and SMS) inflates the write/read ratio (since receiving updates probably don't visit the site as often) but also potentially adds another set of problems (but not as badly as e-mail).
I too was thoroughly frustrated by the Twitter-out this morning. I don't know why Twitter is so prone to crashing but I do know this: Someone's going to take away all their customers if they don't get their issues under control, and soon.
If I were a coder I'd be working on a "better Twitter" right now.
And yes, I know, there's Pownce, but for whatever reason nobody likes it.
duh - it's cuz scoble's on twitter!
What is your thing with Scobble?
Like, what exactly does he do?! =)
good q - what does he do?