Genome@home - Page 5

Undertoad · 10-20-2001, 12:30 PM

I've been following the G@H message board and there is much confusion and unhappiness in the G@H community.

It seems that there have been problems in some of the work returned. A lot of work is going to be thrown out and a new stats calc method is going to be implemented.

Now, 27% of TeamCellar's work is going to be thrown out. This is a massive number compared to most of the others. This may be because I didn't install the client properly on some of my boxes - I'm not sure, and the discussion doesn't make it clear. And it may be that Lisa's multi-CPU box returned results in such a way that they couldn't figure out whether it was bogus or not. I'm not sure and the discussion doesn't make it clear.

Having results thrown out when we did nothing really wrong is alarming. The sense I get from the message boards is that the project is being poorly run. The loss of stats is one thing, but even pointless work for the science is another and if I'm cranking away at it and wind up losing a quarter of the work, that's just sad.

Bottom line here is that I'm going to stop running the client myself, at least until it goes out of beta and there is some agreement that it actually works. Unless there is any disagreement, I'm also going to remove the tracker on the front page.

lisa · 10-21-2001, 08:30 PM

Well, that's okay... Mine needs to stop running soon also. The machine will soon be needed for other purposes. So, we had our run.

10-20-2001, 12:30 PM	#61
Undertoad Radical Centrist Join Date: Jan 2001 Location: Cottage of Prussia Posts: 31,423	I've been following the G@H message board and there is much confusion and unhappiness in the G@H community. It seems that there have been problems in some of the work returned. A lot of work is going to be thrown out and a new stats calc method is going to be implemented. Now, 27% of TeamCellar's work is going to be thrown out. This is a massive number compared to most of the others. This may be because I didn't install the client properly on some of my boxes - I'm not sure, and the discussion doesn't make it clear. And it may be that Lisa's multi-CPU box returned results in such a way that they couldn't figure out whether it was bogus or not. I'm not sure and the discussion doesn't make it clear. Having results thrown out when we did nothing really wrong is alarming. The sense I get from the message boards is that the project is being poorly run. The loss of stats is one thing, but even pointless work for the science is another and if I'm cranking away at it and wind up losing a quarter of the work, that's just sad. Bottom line here is that I'm going to stop running the client myself, at least until it goes out of beta and there is some agreement that it actually works. Unless there is any disagreement, I'm also going to remove the tracker on the front page.

10-21-2001, 08:30 PM	#62
lisa Etherial Join Date: Apr 2001 Location: CA Posts: 153	Well, that's okay... Mine needs to stop running soon also. The machine will soon be needed for other purposes. So, we had our run.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)