Has any new spike/gaussians been found lately?

Everything is strangely silent on the results front... everyone still seems to be focusing on credits/server outages.

Can anyone point me towards any new utilities which datamine the new BOINC results we are creating for personal research?

---

Overclock with the MSI G31M3-L and Intel E8600 3.33Ghz
Intel D865GLC Socket 478 Motherboard
~How To Overclock The Eee ASUS 1005HA Netbook To 1.9Ghz~

> Has any new spike/gaussians been found lately?
>
> Everything is strangely silent on the results front... everyone still seems to
> be focusing on credits/server outages.
>
> Can anyone point me towards any new utilities which datamine the new BOINC
> results we are creating for personal research?
>

I noticed the same thing, so I posted a gaussian singnal thread in Number Crunching a few days ago called Gaussian Signal.

Thanks!

I am a SQL DB programmer who loves doing this type of work. What would you think about a website where we could post this type of information?

I wrote a similar database in Microsoft Access for the old SETI client. It runs as a pseudo screensaver that every thirty minutes runs a query comparaing data spikes/gaussians on a two week window for the target year and the year before (same two week range). It yielded some real interesting results. But since my cluster (10 machines) could only get WU data for what it was given, the data was incomplete. If we had a way of sharing this data (outside of SETI official channels, since we are not allowed to search it) with each other we could have some real fun!

---

Overclock with the MSI G31M3-L and Intel E8600 3.33Ghz
Intel D865GLC Socket 478 Motherboard
~How To Overclock The Eee ASUS 1005HA Netbook To 1.9Ghz~

> Thanks!
>
> I am a SQL DB programmer who loves doing this type of work. What would you
> think about a website where we could post this type of information?
>
> I wrote a similar database in Microsoft Access for the old SETI client. It
> runs as a pseudo screensaver that every thirty minutes runs a query comparaing
> data spikes/gaussians on a two week window for the target year and the year
> before (same two week range). It yielded some real interesting results. But
> since my cluster (10 machines) could only get WU data for what it was given,
> the data was incomplete. If we had a way of sharing this data (outside of
> SETI official channels, since we are not allowed to search it) with each other
> we could have some real fun!
>

Hmm, that sounds like a good idea Reuben. I'm not sure if you meant that I help in setting up the site though. I live outside the states right now so this would not be something practical. Setting up the site is not so hard, but administrating it is another matter. Also, the traffic and service restrictions that I have in place on my system would make this very difficult. But like I said, I'm not completely sure what you mean about the website. I'm suprised that Berkeley doesn't have a solution for this. [edited]

> I'm not sure if you meant that I
> help in setting up the site though. I live outside the states right now so
> this would not be something practical. Setting up the site is not so hard, but
> administrating it is another matter. Also, the traffic and service
> restrictions that I have in place on my system would make this very difficult.
> But like I said, I'm not completely sure what you mean about the website. I'm
> suprised that Berkeley doesn't have a solution for this. [edited]
>
I have a site already (my company is a webhosting operation). The complexity comes in with result authentication. I can make the site accessible to the public, but uploading results by inviation only. The public will be able to download the master database anytime. I just need to get: 1) Interest, 2) A standard container(Access MDB), 3) A group of users willing to upload the result files in a container on a montly basis.

The whole point of the subproject is this: you find a spike at coordinates (X,Y,Z) for time = N. You fire up the public database and do a SQL search for spikes at any variation of X,Y,Z,N. Found a similar curious match? Good, then report it as a responsible SETI amateur scientist! What harm could come of 100 users getting up in arms over a series of spikes. I know seeing the users who whine online about credits getting drowned out with reports of spikes would be music to my eyes. :)

What do you think?

---

Overclock with the MSI G31M3-L and Intel E8600 3.33Ghz
Intel D865GLC Socket 478 Motherboard
~How To Overclock The Eee ASUS 1005HA Netbook To 1.9Ghz~

> > I'm not sure if you meant that I
> > help in setting up the site though. I live outside the states right now
> so
> > this would not be something practical. Setting up the site is not so
> hard, but
> > administrating it is another matter. Also, the traffic and service
> > restrictions that I have in place on my system would make this very
> difficult.
> > But like I said, I'm not completely sure what you mean about the website.
> I'm
> > suprised that Berkeley doesn't have a solution for this. [edited]
> >
> I have a site already (my company is a webhosting operation). The complexity
> comes in with result authentication. I can make the site accessible to the
> public, but uploading results by inviation only. The public will be able to
> download the master database anytime. I just need to get: 1) Interest, 2) A
> standard container(Access MDB), 3) A group of users willing to upload the
> result files in a container on a montly basis.
>
> The whole point of the subproject is this: you find a spike at coordinates
> (X,Y,Z) for time = N. You fire up the public database and do a SQL search for
> spikes at any variation of X,Y,Z,N. Found a similar curious match? Good,
> then report it as a responsible SETI amateur scientist! What harm could come
> of 100 users getting up in arms over a series of spikes. I know seeing the
> users who whine online about credits getting drowned out with reports of
> spikes would be music to my eyes. :)
>
> What do you think?
>

It sounds good, but I think that you mentioned a couple of problems: 1. Getting enough people to upload result files. I for one would not have a problem uploading the results files. 2. Using query language for the searches. This I think would be something that will limit the amount of users 3. The issue with the container. If I'm reading what you're saying correctly, this would be a problem with the data access objects.

I agree that people do tend to talk about their credits just a little too much. Sometimes I wonder if people are forgetting that they're actually looking for alien signals.