Cook, typically a verbose sort online, hasn't mentioned anything about his new job on Twitter or any of his other frequent outlets. Could he be abashed about his new assignment? Yahoo is no one's employer of choice in the Valley these days. But given his noisy departure from Twitter, it's not likely he had many options. And neither did Yahoo.

Five out of six Summize employees — all engineers — are joining Twitter. They proved their mettle by helping Twitter stay online during the announcement of Apple's iPhone — the kind of technological epiphany that sends Twitter users into a frenzy.

Sure, Twitter could have hired the likes of Summize CTO Greg Pass. But the tools of traditional recruitment — resumes, references, interviews — are time-consuming and mostly useless. And it's difficult to reward employees adequately, even with pre-IPO stock options. Do you really think a director-level Twitter employee would get the equivalent of $3 million in salary and stock? By buying a startup, and taking in its employees, a company can dole out outsized — but well-deserved — stock rewards, and skip a lot of HR rigmarole.

Cisco has long pulled this maneuver, buying promising networking companies and then pumping their products through a readymade sales force. At Google, employees are now muttering about how the only way to get rich is to leave the company, form a startup, and get bought by their ex-employer. That Twitter's pulling such a sophisticated trick at such a young age shows that Williams is even cleverer than we thought.

But it's clear they're talking about Cook, who they identify, rather insultingly, as "a former systems administrator." The post brags about "a recently enhanced staff of amazing systems engineers formerly of Google, IBM, and other high-profile technology companies." That, too, is an obvious dig at Cook, who's mostly worked at startups.

But the friend of Blaine who emailed us about Twitter's outage puts the blame on an "operations guy" at Twitter, whom he describes as a "fucking moron." He writes:

The whole story is that it takes more than just Blaine to keep Twitter up and running and whether servers are up, properly configured and not running hot definitely doesn't fall under the developers' responsibilities.

The other part of the dispute was whether Twitter needed to be rewritten in another programming language. Perhaps, but that wasn't the real issue in scaling, according to the Cook camp.

We're utterly unqualified to evaluate the technical arguments here. But the back-channel badmouthing that's going on here? We're experts at that, and we rate it utterly delicious. As fingerpointing goes, this Twitter battle takes the prize.

Languages don't scale, architectures do. Some languages are faster than others — a given operation costs less. Faster means cheaper, it doesn't mean more scalable. Perl used to be slow. Now it beats JoCaml with the bestest concurrency. What was Perl built for? Parsing text. Does it mean that you can't build Wide Finder with another language? Does it mean that you couldn't build Wide Finder to scale out to a trillion documents with gawk? If you answered "yes", go back to the start of this post and read again! :-) If you're still answering "yes," try reading some more.