In Internet Parlance: LOL
January 31st, 2008 by EyeOnWinerDave wants Twitter to brief him on their new hardware. Thinks it makes sense because the President sometimes briefs Congress.
I’d like to really understand what’s going on behind the scenes at Twitter, Inc. They say they’re confident the new infrastructure will hold up better, I’d like to understand why. Can we have a meeting, with a few people from the tech community who actively use Twitter and a few people from the company, to be briefed on what’s going on. The same way the President briefs Congress when there’s some kind of international crisis.
The answer to this is “no” for so many reasons. Further, even if it’s “yes”, the only reason Dave would be included in that group would be the “squeaky wheel” syndrome.
“Hello there, I see that your cleaning solution is being marketed as ‘new and improved’, I’d like to schedule a time that some other consumers and I can be breifed on the changes.”
“Your gym was recently remodeled, I’d like a briefing on why you believe this to be a more efficient setup. Me? Oh, I’m a member of the gym!”
NPD in full effect here. Twitter has thousands upon thousands of members… what is it, exactly, that makes Dave think he’s one of the handful that would actually be selected? The only thing he does better than the rest of the Twitter community is bitch and run smear campaigns.
…come to think of it, that might be enough.
January 31st, 2008 at 9:57 am
Not sure which is more annoying, the idea that a change in the hardware infrastructure at Twitter constitutes an international crisis, the fact that Dave apparently wants the authority to issue subpoenas to compel testimony from people who provide him with free services, or the idea the Dave’s background and education might somehow qualify him to even have an opinion on the matter.
I did like the reference to “those of us who are engineers or would-be engineers.”
January 31st, 2008 at 9:59 am
I commented on Dave’s blog: (it will be removed… and not because of my words… but because I’m betraying my multiple personalities and he’ll read htis and delete me on priniciple.) I wrote:
Twitter used to be hosted by a hosting company called Joyent. When they reached sugnificant size they elected to bring up their own infrastructure. They swtiched over recently when the service went off line over a weekend.
The latest clue is that Twitter expectes the Superbowl to really stress the system so they have contracted with Joyent again for extra capacity. This split infrastructure is probably the reason they needed to re-configure the systems… so some of the work stays local and some opverload gets dispatched to infrastructure at Joyent.
NOTE: All of this is based upon a recent Joyent blog post and it IS pure conjecture.
If twitter survives the superbowl and doesn’t go down then they have truly scaled and the money invested in infrastructure was well spent. if not, then they will need to invest in more infrasatructure engineering talent to carck the code on the next wave of adoption.
Thinking that twitter can be replaced by protocols misses the essential problem. Large systems that experience network effects always require new levels of expertise that have not been tested by any but a small handful of talent.
Scale anything to a great degree and it will break or deliver service levels that are unacceptable.
During these times of crisis the last thing you want to do is add more “engineers” to the process… especially engineers that use the service and just want it fixed. So, wait and shop for alternative services.
Google out engineered Yahoo for search. Apple is out engineering MS for user experience on a laptop.
Can anyone out engineer twitter for this context sensitive chat environment? It will take a lot of hardware and talent to keep it running. And that means money. I’m amazed they keep scaling and haven’t considered how to make money yet.
PS> This all applies to wordpress.com and they are amazing, IMHO.
January 31st, 2008 at 10:06 am
Asking Dave is sit on a “Users Counsel” would be like asking the Unibomber to tour post offices.
His experience in helping companies is well documented. The list of companies he has “helped” include:
Audible (recently purchased by Amazon): Dave had a heated confrontation with the CEO at a social function.
Apple: Dave got a level of service that no other customer gets. He would expect us ALL to demand nothing less than what he got by simply writing to Steve Jobs and sharing the text on his blog.
UserLand: He owns this one but let other run it. Then they asked him to share the spoils from selling off a little asset for $2.15M. He re-focused the attack on the lawyers used to negotiate.
PodShow: Almost became a principle of the start-up but got thrown off the island before it was too late. Now they are the poster child for New media greed, in Dave’s view.
Mahalo: “JC complex” of interest there. But Dave’s just trying to help them.
O’Reilly: Dave is “covering ETech” with a press pass from O’Reilly. End of an era.
January 31st, 2008 at 10:24 am
Remember how well Weblogs.com scaled when Dave was in charge? He knows shit about scaling. If he did he probably wouldn’t try to duplicate wget’s functionality with a custom scripting environment that has to be emulated in real time to run on Intel.
And Twitter has already been out-engineered. By every other company who has ever done IM. AOL, Google, Yahoo, Microsoft, etc. They all handle billions of messages per day without a peep. Twitter to date has had a bit over 600m updates. AIM will do that before lunch every day.
January 31st, 2008 at 10:36 am
Can I see every AIM message someone a user sent out today on a webpage?
Can I track a conversation between two distant AIM users?
Twitter is distinctly different… a hybrid of blogging and messaging. It’s a paradigm changer.
And AIM was funded by a profitable on-line service… remember those damn CD’s everywhere? Google used ads. Yahoo too. Microsoft just applied Monopolgy 2.0 to get MSN to scale. Twitter is the little guy trying to meet payroll and find capital.
Dave is no VC consultant. He’s hasn’t run a successful company since he sold Living text to Symantec. That’s another way for Twitter to get profitable. That’s the Web 2.0 model for achievement.
How the hell does wordpress.com do it?
January 31st, 2008 at 10:47 am
IM and Twitter are different enough, but the challenges are quite similar. The individual Tweet pages aren’t the problem, should just be a single very quick DB call, it’s the fact that when one user posts a message it might have to be pushed out to thousands of people.
WordPress.com does it without Rails. And really inefficient messaging.
January 31st, 2008 at 10:50 am
I laughed out loud when I read Winer comparing a Twitter outage to an international crisis.
McD: There’s no technical difference between Twitter exchanging messages in a browser and AOL exchanging them in an IM client. Twitter, given its technical challenges, shouldn’t be having these downtimes. They must have shitty or horribly overworked coders.
January 31st, 2008 at 11:17 am
Bull: I worked for the hardware vendor that supplied AOL. They had the capital to buy racks and racks of servers to make AIM scale.
If twitter asked the VC’s for a few hundred million they could re-architect the infrastructure and scale as well. But in the process Evan would loose his company. H’s going to wather this phase of growth on the cheap and cash in when someone buys the business for the user base.
Most problems of scale face the constraints of funding, people and process before the technical issues are even a factor.
The Twitter Ruby meme is pretty superficial. It reminds me of:
Ebay Java Amazon what ever Google Linux
People, process and finance come first. Get those under control and engineering fixes the bottlenecks.
January 31st, 2008 at 12:04 pm
There’s a big difference between sending a message between two individuals using a single protocol and sending a message by one of several protocols (web, sms, IM, etc) and broadcasting it to thousands of users also using multiple protocols.
At the simplest level, the number of hits to the website to view messages would be similar to the digg or slashdot effect. Meanwhile, it’s also being pushed out to hundreds or thousands of recipients via SMS & IM.
January 31st, 2008 at 12:18 pm
AIM has tons of mobile and web users. But they are a big company and can afford the servers like McD said.
I pick on Rails here because this is exactly the kind of service it’s not made for. There is not much to the front end of Twitter, the magic is all in the backend. And the magic is apparently running slowly. Even 37Signals rewrote a key part of Campfire (a web chat service, sort of related here) in C because Ruby couldn’t keep up with real time chat when a decent number of users came online. If it can’t keep up with a simple chat program, there is no way it will keep up with Twitter.
January 31st, 2008 at 1:28 pm
Instant messaging and IM-like web services don’t require AOL-sized resources to function with acceptable levels of performance. ICQ was developed by four young Israelis in the mid-’90s. It grew at a much faster pace than Twitter and was bought by AOL two years after its founding for $400 million.
There’s a big difference between sending a message between two individuals using a single protocol and sending a message by one of several protocols (web, sms, IM, etc) and broadcasting it to thousands of users also using multiple protocols.
Twitter is centralized. Where’s the broadcasting?
People, process and finance come first. Get those under control and engineering fixes the bottlenecks.
Not always. Look at Technorati.
January 31st, 2008 at 1:30 pm
Dave won’t provide Mahalo any links to update HIS OWN PAGE on their site because he’d be “working for free”. Responding to Sean Percival’s comment about cleaning up a spam link on Dave’s own Mahalo page and suggesting Dave might drop by and suggest some links, Dave wrote:
= = = I assume you’re paid a salary and get benefits and stock options to do this work.
If so, when you ask me to suggest some links, you’re basically asking me to work for free, right? So your stock can be worth more as well as the other shareholders of your company?
Now I do this for a lot of companies, I know when I’m doing it I’m being a chump, but if I love the product, I do it anyway. But…
Sorry, not only don’t I love your product, even if I did, I doubt I’d help the company. = = =
buuuuuuut he expects TwitterCo. to drop to their knees and offer up people and time to sate his obnoxious, intrusive, utterly-irrelevant curiosity?
It’s not like Mahalo’s asking him to edit, say, Perez Hilton’s page on their site, or be a regular unpaid contributor to their site in general.
You’d think, ego-driven as Dave is, he’d be interested in any prominent representative collection of DaveFacts out there. And really, I’m sure he IS, but he’ll put out a word to his minions and have THEM do the work for him. All UNPAID, of course!
Nevermind that he’s frequently asking for assistance, links, ideas, comments on his own blog, Twitter, and elsewhere. Any of those people getting paid, Davey? It’s OK to work for free if they’re working for YOU for free? Is that right?
So typical Dave does one thing and expects everyone else to abide by entirely different rules, depending completely on how much he benefits from them.
How does our planet sustain his ego again? And more distressingly, imo, who are all these chumps sucking up to him and doing all his dirty work that don’t grok that they’re getting played, like, daily?
January 31st, 2008 at 1:36 pm
Twitter is centralized. Where’s the broadcasting?
If twitter is centralized, basically everything they do is “broadcasting.”
If you look at it as a webservice, this seems bizarre. But if you get your twitters from your phone or IM, it is much more obvious.
January 31st, 2008 at 1:46 pm
The really interesting thing is that his ’scripting’ environment won’t work with google talk because of the ‘interfaces’ - but he seems to think that other XMPP servers will have a different ‘interface’ even though it is XMPP.
How do you break something like that to someone like Dave?
January 31st, 2008 at 3:36 pm
Twitter is centralized. Where’s the broadcasting?
That one server needs to handle lots of hits as people view the page, just the same as any other web server. The broadcasting comes in when it also has to push that same message to hundreds of IM and SMS recipients. With traditional IM, one message goes to one recipient (or a few in a multi-user chat).