UNIX SIG Notes By Christopher J. Fearnley Highlights from the D.C. Linux Conference: Donald Becker of NASA described his group's efforts to maximize price/performance on compute intensive "supercomputer" class problems by using off-the-shelf Linux boxes in a parallel computing ethernet environment. Performance is only somewhat lower than with real supercomputers. Dr. Gregory Wettstein of the Roger Maris Cancer Center explained how Linux was able to dramatically improve patient care at their facility by providing a reliable distributed peer-to-peer network. Dr. Wettstein suggested some "marketing tips" for Linux. I have modified one (I guess I should have taken notes :) which describes Linux as "a cooperative venture developed on the global information networks." And I have to quote another of Dr. Wettstein's phrase-tools: "I would rather spend 10 hours reading someone else's source code than 10 minutes listening to Musak waiting for technical support which isn't." Vance Petree of Virginia Power described how Linux replaced a minicomputer in the task of collecting vital summary data from six remote sites. And unlike the mini, Linux hasn't been the cause of any lost data packets since it took over the job! Mark Bolzern of WorkGroup Solutions (the FlagShip people) announced that his company is so enthusiastic about Linux that they are marketing Linux (not just Flagship) on the assumption that Linux's market growth will, as a side effect, spur FlagShip sales. Bob Amstadt, the founder of the wine project, spoke about the Windows emulator project. Apparently there are presently only about six wine developers. They each have there own special interests which means that parts of wine are working pretty well and other parts much less well. It does sound like most of the wine infrastructure is in place. But a lot of debugging is left. Since the funding for a paid intern fell through, wine is still alpha software. Don't expect it to reach production level (including being able to run your favorite windows apps) until more developers volunteer or are hired by as yet nonexistent funding. Amstadt feels another six months at least. The consensus at the conference seemed to be that Linux works, works well, and is rapidly taking over responsibility for those mission-critical applications for which the commercial alternatives have proven to be inadequate. Obviously there was a lot of networking going on, etc., etc., One conversation that hit home was on whether Linux is easier to administer than say DOS or Windows. Several argued that Linux is easier because you don't have so many designed-in limitations to fight with (can you say TSR conflicts?). This hit home when I returned to Philly to yet another Windoze job (arghhh!) where I had to manhandle Windows into submission (yet again) and barely survived. Put simply: Windows doesn't work, but Linux does. Perhaps the only problem with Linux is that reading the manuals is still important (oh, and it can't do everything ... yet). Finally, just in case you didn't see a copy of the January Linux Journal. I should quote Linus' answer to Belinda Frazier's question "Do you have any new hopes for Linux?" Linus responded: "I think my ``plan'' says something like ``World domination. Fast.'' But we'll see." I guess Linus isn't satisfied with simply having the best OS around, now he wants it to take its rightful position at the lead of the OS pack. I think I agree with him! At next month's (January 18) meeting I will talk about "X Resources."