Twitter in FailPOOL
Twitter - whether you know what it is or not is the one company (next to Yahoo!) that has been in the news in 2008 for its ability to fail. What started out as the golden child for social media experts and early adapters now face constant struggles to stay alive. Let’s face it… when your designers need to get cute with multiple ERROR screens because there are several ways to kill the system - that’s a problem! Sending out a hot vet assistant telling me my dog has cancer does not make the issue any easier to swallow. The choices are to operate or to put it down. Note: multiple hot assistants wouldn’t help either.
What irks me about the Twitter platform is that it’s being used as a live communication device. A few of my friends have subconsciously stopped emailing me and have instead turned to direct messaging me on Twitter to get my attention. That’s fine for recreational purposes or as updates to their status on a microblog but when things are important or when things need my immediate response, Twitter is still failing miserably at this.
In the last week or so, I’ve noticed overall updates have been down. Whether that’s people abandoning the platform or Twitter’s throttling their usage, I don’t know.
I’ve noticed with the most recent Twhirl upgrade, Twitter has dropped its hourly API quota from 70 down to 20 requests per hour. First off, just because you managed to keep your system alive (barely) on life support, doesn’t mean it’s still good or useful for the general public. If email only allowed 3 gets an hour or if cell phones only allowed access to calls and data within some throttled time blocks during the day, would it still be useful for communication? Would anyone trust it enough to rely on it knowing it’s in a constant state of cripple? The answer should be “No” to the above 2 questions.
I’m really baffled as to why Twitter think it’s okay to scale it back to this level of service when its indirectly competing with other forms of communication. It’s like a rat competing to be the messenger in a room full of pigeons. Let’s be real here… the only time I would personally accept a delay in live communication to the point of crippling the experience would be if we’re doing live chat with someone who lives close to the Sun. Being that it takes light 8 minutes to travel between Earth and the Sun, I guess I would have to settle for delayed responses even though I’m sure engineers would try to find a way to make that closer to real time than Twitter’s proposed solutions.
Maybe they need to hire some real startup people - you know, the ones who have built extraordinary things without the luxury of doing so in a funded startup. I fail to see how $15 million (and more) can’t create a stable system… at the same time, maybe they should just shut it down and start over. Sometimes it just makes more sense to buy a new car than trying to fix a lemon.
If you like this post the please consider subscribing to our full RSS feed.
Save to del.icio.us |
Digg This! |
Share on Facebook |
reddit
Stumble It! |
Wei on June 24th 2008 in Website Reviews










Henri responded on 24 Jun 2008 at 9:27 am #
Amen!.. Is it time to move away from twitter? I think so.
Calvin responded on 24 Jun 2008 at 9:54 am #
From all the information I’ve read, it sounds like that they are starting over. Scaling down their service is just a way to keep the lights on until they can use their new architecture.
Don’t get me wrong, I’m not defending twitter here - they really haven’t done a good job of managing site performance. I’m skeptical that 15 million will fix the problem either (without creating others).
Brian Culler responded on 08 Aug 2008 at 10:41 am #
I’ve said this all along. Twitter is an IM service wrapped up to look like something else, with some fancy opt-in membership rules. It should have been built from the ground up as an IM protocol.