In my life as a full-time employee, I ran a membership management system for an international non-profit that organized over 40,000 volunteers at the local, state, and national levels in almost 30 countries. It coordinated physical material deliveries (part of the system was a full shopping cart installation), managed electronic documents, and delivered news to all those volunteers. I built the system myself from scratch (minus said shopping cart), and ran it on my own for almost three years.
When I wasn’t coding it or maintaining the server, I provided tech support for the office staff and customer support (including phone calls) for volunteers.
I was the only technical person in the whole place.
I can count the number of times that system went down as result of anything other than a hardware issue on one hand.
Much as I love it, I fail to see how a minimally configurable 140-character obtuse message board can fail so often!
Update: Alright, it’s back up. Good recovery this time, but you still have to reset the “Days since an accident” sign back to zero.
soooo…. what do you think is the culprit? News reports recently suggested that user base was as large as US Population. Does that give a clue or are you aware of an architectural defect?
BTW… I also think the timing is ironic with considering the recent GMAIL fail. Just noticing.
I honestly don’t know, Ethan, I was simply commenting on the frequency of the failures, using a piece of my own professional history as a slightly comical juxtaposition. I can’t tell if your comment was said with sarcastic inflection, so I apologize if I came across as pompous. My wife always says I’ve got decent material, but my delivery sucks. Oh well.
Obviously it’s not a big deal, I just find it funny how often the outages occur. Google outages always make news because they are so rare. Obviously Twitter doesn’t put the same emphasis on availability.
Let’s say that the failure was due to stress levels due to massive usage. Their analytics surely foreshadowed this scenario. This huge user base didn’t appear out of nowhere, after all. Their accountability for the outage remains, regardless of the reason.
Were I the boss, the vast majority of my resources would be directed towards infrastructure. 2 reasons. First, branding: I’d want to do everything in my power to get rid of the “Twitter’s out…again” stigma.
Second, With an offering this focused, all other projects are utterly worthless if your service isn’t 100% stable (and since you never get 100%, you keep striving).
BTW… I also think the timing is ironic with considering the recent GMAIL fail. Just noticing.
Given the rarity of such events from Google, I see what you mean.
No Pompous overtones taken….
I was simply admiring the elephant in the room as well. With announcements coming at the same time that twitter was about to turn a profit, it amazes me as well that the failure frequency is so frequent.
As a fan of your blog for some time now, I also thought that you had identified something obvious and was posing the question to your audience to see if we too caught it. >.<
That being said, as a New England native sarcasm is a natural trait!
Whew, glad to see my delivery isn’t completely off!
I was simply admiring the elephant in the room as well. With announcements coming at the same time that twitter was about to turn a profit, it amazes me as well that the failure frequency is so frequent.
100% agree. The failure rate is mind-blowing for a service riding this large of a success wave. Given the massive consumer interest (and consequential investor interest), their road map should be crystal clear:
1. Strive for 100% service availability
2. Profit!
3. All the unimportant BS that’s likely getting in the way of Step 1
As a fan of your blog for some time now, I also thought that you had identified something obvious and was posing the question to your audience to see if we too caught it.
That’s exactly what I was doing, though without as much initial confidence as I should have, given our discussion here. Through my experience, I like to think that I’ve got a firm understanding of the tension between business and technical needs at play in almost every business. So from where I sit, it seems to me that any business or technical dunce can see what needs to be done right now.
The fact that it keeps happening means that someone at the top doesn’t get it, which means it’s probably like the places I left to work for myself.
And thanks for being a fan, Ethan! Here I thought it was still just me and my cat.