
Abuzz
» see abuzz.com as it lived, on the wayback machine
~ the obituary ~
When I had a question in those years I posted it on a Turkish forum and waited. Somebody who knew would eventually turn up, or nobody would, and either way the mechanism was a thread and a lot of patience. Abuzz was built on the theory that the waiting was the problem and that software could route a question to the person most likely to answer it.
The product was called Beehive and it came out of MIT. Everyone in a group described what they knew, the system learned from what they actually answered well, and an incoming question got delivered to a short list of people instead of being posted to a room. Answers were rated and archived, so the second person with the same question got it immediately.
It started in 1996 in Cambridge, Massachusetts, aimed at corporate intranets, which is the right market for it. Companies have exactly this problem and very little else that solves it. Then came the web version, the same routing engine pointed at strangers rather than colleagues, and that is what The New York Times bought on 15 July 1999 for $30.1 million, mostly in its own stock.
Nothing about this was in my orbit, and even if it had been, the answer to "who knows about this" on the Turkish web of 2000 was a forum with a signature block under every post listing the hardware you owned. Mine listed mine. The routing was done by whoever happened to be reading that evening. Worse in every measurable way, and it is the version that survived.
~ the case file ~
| Born | 1996, in cambridge massachusetts, as beehive |
|---|---|
| Peak | bought by the new york times on 15 july 1999 for $30.1 million |
| Killed | 2001, after a $22.7 million write-off the previous quarter |
| Lifespan | five years, two of them owned by a newspaper |
| Made by | abuzz technologies, out of mit |
| Killed by | the dot-com crash, and not enough people asking |
~ what the times wanted ~
Martin Nisenholtz ran Times Company Digital and had spent 1999 consolidating around fifty separate online properties into one unit. His pitch for Abuzz was that it would be a resource for the paper's growing audience of like-minded people and for the advertisers who wanted to reach them, which is a sentence from 1999 in every particular.
The real logic was traffic. A newspaper site in that year had a reading problem: people arrived, read one article and left. A question-and-answer network gives them a reason to come back and a reason to type something, and typing something is how a visitor turns into a user. Every publisher on earth had this thought at roughly the same moment and most of them bought a company to act on it.
~ the write-off ~
It did not work. The service ran on nytimes.com and never reached the density that makes a question-and-answer network worth opening, which is the failure mode all of them share. A question sitting with no answer under it teaches the next person not to bother asking.
The accounting is the clearest record of what happened. The Times took a $22.7 million write-off on Abuzz in the fourth quarter of 2000, three quarters of the purchase price gone inside eighteen months. Then it got worse in a specifically dot-com way. Because the planned New York Times Digital public offering was pulled when the market turned, the company owed former Abuzz shareholders $25 million in cash instead of the stock they had been promised, and paid it out in the first quarter of 2001. Flatiron Partners was among those who collected. By April 2001 the trade press was writing about Abuzz in the past tense.
~ who won instead ~
The idea was not wrong. The timing and the ownership were. Yahoo Answers launched in 2005 with no expert routing at all and an open door, and it got the density Abuzz never did. Google Answers tried the paid-expert version and closed in 2006. What beat both of them was volume plus a search engine indexing the archive, so that most people arriving at an answer had never asked anything.
What Abuzz had was a good model of who knows what, built in a decade before there were enough people online for the model to have anything to chew on. Routing works when you have a company behind it. It does not work when what you have is a newspaper's readership and a launch date in 1999.
~ leave a rose ~
an anonymous tribute. one rose per visitor per day.
@}-->-- ... roses left here
~ leave a wreath ~
a rose fades by morning. a wreath stays on this grave for good, with your name on it.
from $1, paid once through stripe. it keeps the graveyard running: the hosting and the nightly digging. adopt it and the headstone reads "maintained in memory by" you. the groundskeeper plants new wreaths at the next tending, usually within a day.
» one email a week: new graves and the anniversaries of the dead. subscribe
other community: Answerbag, Enciklopedio Kalblanda, Tiexue, CGSociety
also died in the 2000s: Adobe Authorware, BitMagic, Windows File Protection
~ leave a note ~
visitors before you have left these graveside notes. anonymous welcome.