Because I first brought the subject up here, I should be the first to say
that things at the Berkeley server seem to be quite screwed up for the
moment. For the past week, Berkeley has been transmitting the same 115 work
units over and over again to all 350,000 people that have subscribed to the
program.

I first found out about this snafu a few days ago, but I wanted to verify
that what was being said on sci.astro and other newsgroups was true before I
mentioned it here. It appears to be true. These 115 work units that are being
recycled have now been processed approx. 20,000 times each. Because of that,
I've turned off my computers for the moment. (BTW, I've found that you can
uninstall the SETI client and later re-install it with no harm; it resumes at
its last checkpoint. The current data file is not affected by the uninstall).

Things being screwed up is something that we're all familiar with. I do
believe however that the SETI folks are making a fairly substantial mistake
by not publicly announcing their failures. This project is as much
psychological as it is technical -- and it depends of the good will of a
great number of people.

The problems seem to stem from several sources. The Sun servers that are
handling the project seem to be too small. Sun Microsystems is apparently
donating larger ones that are in the process of being installed now or very
shortly. Secondly, data flow from Arecibo may not be going as smoothly as
desired. And there may be bugs in the client software that runs on the PC.
All of this is supposition, however. The people at SETI@Home are being very
closed-mouthed.

Under any circumstance, there seems little reason to keep recalculating the
same data for the moment. Nonetheless, I'm still enthusiastic about the
experiment and I will resume my participation when they seem to be ready to
go again.

Wirt Atmar