Hi All

Not really sure what category this query should go under on the forum so posting it here instead - Apologies if its wrong.

I used to run SETI@HOME as a screensaver way back when (Must be 10 or so years ago now) but this phased itself out over time with family concerns/new hardware etc.

The recent media coverage of HD164595 served to remind me about this project and I've signed up again and happily crunching away ;)

This whole experience of with SETI and my interest in Stats/Tech got me wondering as to the growth (or atrophy) of the project as a whole by way of active users and active hosts computing tasks for the project, so with that in mind; I'd like to start collecting these figures myself, preferably from source rather than via a third party stats collector however looking at BOINC documentation, I should be able to get this info via http://setiathome.berkeley.edu/stats/tables.xml but it seems to be lacking what I'm looking for.

Whilst the total number of hosts is exposed via a compressed file (I think), I'd rather not have to download this file and extract it, then process it to account for active hosts (and the same for users) each day if I can help it - Ideally I'd be happy with a simple web call that would give me what I am looking for - Its much nicer on everyone's resources ;)

So all that being said - Is there any way I can get this information directly from the SETI@HOME project or is it simply just a case of needing to do the aforementioned processing or try to obtain statistics via a third party source?

Ultimately this information will go on a sub-page of my website that would simply show graphs of these trends, hopefully collecting it for many years.

I guess what I am looking for is (hopefully) a "lightweight" way I can grab this info, similar to what seems to be available on http://stats.free-dc.org/stats.php?page=proj&proj=sah

Kind Regards,
Simon

The only way to do it, just like all the other statistics sites, is to download the big file there, host.gz and unpack that. And think of it this way, true statistics sites download that data from all of the BOINC projects, which depending on whether they count in alpha and beta sites can be quite a bit of them.

Of course, not all have the amount of hosts/users that Seti does. But it's still a hefty amount. And to download that daily, you're required to have an ISP that allows you to do that.

Don't think about page scraping, that's frowned upon here and will cost you a blocked IP if found out. Not that it matters much because the data you're looking for cannot be scraped from a page here, it truly is only in the host.gz and users.gz files.

---