20 January 2009
On the heels of last week's tech talk, here's MobWrite 3. Now with guaranteed synchronization and many other improvements. When demoing it at Google I like to yank my Ethernet cable in mid-edit then watch the system recover gracefully. There's a lot more to come, but I think the synchronization protocol is now stable. Everything else is just add-on features.
Talk amongst yourselves. Be nice. (This instance of MobWrite is running on my dev server, so it may go up and down.)
A couple of follow-ups from previous stories:
Last week I posted a photo of an apparently dangerous electrical panel inside the engine room of the S.S. Jeremiah O'Brien. Documentation shows that the O'Brien runs on 110v DC and the limit of lethality is approximately 50V AC or 120V DC. So it would sting like crazy, but probably won't kill a healthy person.
Last May I posted a picture of a plane floating after an emergency water landing, as depicted on a safety pamphlet. I had claimed that this was impossible for planes which have engines under the wing. Here's a video showing what normally happens when such a water landing is attempted. The only successful water landing in history was in 1968 when a DC-8 (which has very small underwing engines) landed in San Francisco Bay. Thus it came as quite a surprise when Flight 1549 landed uneventfully in the Hudson River last week. Thus I take it back, it's not impossible, just improbable.
It would also be interesting to objectively look at who should get what fraction of the credit for executing the successful landing: the pilot (who has received all the credit) or some software engineers in France who wrote the flight-control systems. Airbus has been the leader in fly-by-wire controls (though they got off to a rough start (interestingly that was an A320, the same as the one in the Hudson)).
And finally, here's a screen shot of a Google search for US Airways Flight 1549 a few hours after the incident. I love that Google's freshness is better than the up-to-the-minute flight status ticker.