Neil's News

First Academic Paper

10 July 2009

My paper on Differential Synchronization has been accepted by DocEng 2009, the 9th ACM symposium on document engineering. In a couple of months Google is sending me to Munich to present it.

Here's a word cloud of the paper as generated by Wordle:
[Word cloud of Differential Synchronization paper]

Update: While working on the final draft of the paper I discovered a math error in the central scoring formula for fuzzy matching (apparently division is a different operator than multiplication, who knew?). The bug was introduced three years ago and has been quietly distorting results ever since. But interestingly nobody had noticed because MobWrite is both best-effort and self-correcting. This is not the first error I've discovered in MobWrite which the system was actively hiding from me. Now the fun part begins: When I fix it, how many other users of the Diff/Match/Patch library will experience failures due to having calibrated their systems to the wonky formula?

< Previous | Next >

Legal yada yada: My views do not necessarily represent those of my employer or my goldfish.