Myself along with other I've spoken to do hope you reconsider your plans on stopping this project.

Let the data pile up until there is a more effective way to sort it. Machine learning, AI, advanced scripts, etc.
Is it other reasons?

I never thought that this Seti@home would accept the idea of giving up because its too much.

They were not analyzing any data--just stockpiling it. Not clear if there were any computations deciding what to keep--or whether tor not it was an effective culling. if they cannot answer this to we unsophisticated proles then They deserve to go out of business.
.

Take Sky. Divide into pixels. Store data from each observation, frequency, type, strength, polarization, and others in the pixel.

Question what pixel had repeated signal.

Answer, sort, sort, sort, sort.

Unfortunately sorting can only be done on data that isn't changing. Hence they have to stop filling the many many terabyte size database. Also consider just how huge the indices are going to be and how long it is going to take to build them. Unfortunately this isn't a problem that can be broken down and sent out to many processors, it is a linear problem.

Take Sky. Divide into pixels. Store data from each observation, frequency, type, strength, polarization, and others in the pixel.

Question what pixel had repeated signal.

Answer, sort, sort, sort, sort.

Unfortunately sorting can only be done on data that isn't changing. Hence they have to stop filling the many many terabyte size database. Also consider just how huge the indices are going to be and how long it is going to take to build them. Unfortunately this isn't a problem that can be broken down and sent out to many processors, it is a linear problem.

It is very common to create what is called a "snap shot" of the operational DB that is then used for data analysis. This, in theory, could allow the volunteers to continue to pileup data.

I am presuming that is technically feasible. I am also presuming that the time the staff spends keeping Seti@Home volunteers running reduces the amount of time spent on thinking/pondering/imagineering about the data analysis step.

While I miss being able to process S@H data I also would very much like to "see" if there are any Seti signals hidden in the observations we have processed.

Tom

---

A proud member of the OFA (Old Farts Association).

To get a better understanding of what is being done, and what the problems with the data (both in terms of size and complexity) it is well worth reading through David Anderson's posts about Nebula:
https://setiathome.berkeley.edu/forum_forum.php?id=1511

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?