It would be great to hear the scientific detail of why its not considered worthwhile to continue to process new data. After all if such transmissions were sent out you would hardly expect to see them over such s short period of time.

Does Seti now have sufficient in house processing capability due to technological improvements to simply not need this distributed computing approach?

or are the CO2 emission from generating the power to run this project simply not worth it?

Can the actual reasoning behind this decision please be explained as im sure may thousands of users would like a clear scientific explanation.

---

I would like to know also, I feel very badly let down by SETI, and I'm not the only one.

The answer is that they have so much data to sort through as the second phase of the search it has been considered expedient to stop generating more possible hits for the time being. With an ever increasing number of potential hits from our screening (billions of possibles) it was becoming obvious that something had to be done, three choices:
1 - Throw it all away;
2 - Stop generating new potential hits and do a massive correlation exercise on the existing mountain, then improve the screening algorithms to stop so many of the "obviously wrong" being detected;
3 - Continue generating potential hits and try to do the massive correlation task on the ever growing database, then try to improve the algorithms while using the old ones.

1 Is less than desirable for so many reasons that it landed on the cutting room floor early in the debate.
3 Doing the sort of correlation that is being talked about would mean that every time a new potential was found one would have to re-run a large part of the correlation again (and there were millions of potentials generated every week).
2 Is left as the least bad option - we don't have any work to do, but the project is able to work off a fixed database.

So far Nebula has only looked at a small corner of the total database, but the first look-see has thrown up some issues with the actual correlation algorithm that have to be ironed out before the mega runs can take place. Under the current correlation process this needs access to the whole database and a supercomputer so is, sadly, not really suited to distributed computing.

What we can look forward to (I hope) are the new screening algorithms that will allow us to re-start crunching in a similar manner to that we've been used to over the last ~20years.

---

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

The answer is that they have so much data to sort through as the second phase of the search it has been considered expedient to stop generating more possible hits for the time being. With an ever increasing number of potential hits from our screening (billions of possibles) it was becoming obvious that something had to be done, three choices:
1 - Throw it all away;
2 - Stop generating new potential hits and do a massive correlation exercise on the existing mountain, then improve the screening algorithms to stop so many of the "obviously wrong" being detected;
3 - Continue generating potential hits and try to do the massive correlation task on the ever growing database, then try to improve the algorithms while using the old ones.

1 Is less than desirable for so many reasons that it landed on the cutting room floor early in the debate.
3 Doing the sort of correlation that is being talked about would mean that every time a new potential was found one would have to re-run a large part of the correlation again (and there were millions of potentials generated every week).
2 Is left as the least bad option - we don't have any work to do, but the project is able to work off a fixed database.

So far Nebula has only looked at a small corner of the total database, but the first look-see has thrown up some issues with the actual correlation algorithm that have to be ironed out before the mega runs can take place. Under the current correlation process this needs access to the whole database and a supercomputer so is, sadly, not really suited to distributed computing.

What we can look forward to (I hope) are the new screening algorithms that will allow us to re-start crunching in a similar manner to that we've been used to over the last ~20years.

Thanks for posting this. When SETI returns, my computers and I will be there to support it!
Stay safe,
John

---

What we can look forward to (I hope) are the new screening algorithms that will allow us to re-start crunching in a similar manner to that we've been used to over the last ~20years.

Take your time. I will be there, in a few months or a few years. Just bring back SETI to us...

You make good points.

3 - Continue generating potential hits and try to do the massive correlation task on the ever growing database, then try to improve the algorithms while using the old ones.

3 Doing the sort of correlation that is being talked about would mean that every time a new potential was found one would have to re-run a large part of the correlation again (and there were millions of potentials generated every week).

Another reason why option 3 wasn't doable, I suspect, was that the distributed part of the project was starting to crumble under its own weight. The servers needed more and more manual interventions to continue sending and accepting work. It must've taken ever more time away from developing Nebula and doing the actual science: they couldn't have continued for very long.

---

Gazing at the skies, hoping for contact... Unlikely, but it would be such a fantastic opportunity to learn.

My alternative profile

I joined Seti over all other projects because, even if just a tiny chance of success, success could prove more fruitful than all the other projects put together.

---

I agree with you.

---

Realistically, I could see this coming because without advances in communication; well radio has it's limits wrt prospects for ET engaging in interplanetary communication. Communication with the moon (Apollo mission) had time delay people would rather not have. That said with quantum enmeshed particles being used for some secure communications now; until there is other technology to be on the search for as a possible work around to the time/distance problem...

Big question, probably many are thinking is updates to the BOINC software given many projects are now using it? We have a working platform now, but if hibernation continues long term, some projects might find need for updates in the years ahead. Especially if computer advances require software updates to continue to function well. Processor updates from Intel/AMD or Windows updates from Microsoft could necessitate software updates to avoid potential issues. New version of Windows and one's utilities stop functioning without software company provided support/newer versions is a prime example....

---

Development of the BOINC platform should continue as normal, despite the suspension of SETI as a project. The processes are separate, although there is some overlap of personnel between the two.

BOINC announced a new recommended version (7.16.11) earlier this month: this is available for Apple, Linux, and Windows. Many of the changes were required to cope with major changes in the Apple platform, confirming the points you make about the necessity of responding to external pressures.

SETI has paved the way to the BOINC platform, with 50 odd projects using it. I am now running projects dealing with Covid-19 instead of SETI@home.
Tullio

---

Since I joined Seti@home in 1999, I have apparently provided 454 quadrillion floating point operations. By my calculations, the same could be achieved in under 5 hours on an Nvidia GeForce RTX 3080, recently announced. Makes you think.

Julian

---

Nvidia has bought ARM, the holder of the Risc CPU patent rights. AMD has bought Xilinx, maker of FPGA chips. I have used a Risc minicomputer running Berkeley UNIX.
Tullio

---

T_T

I was joined SETI at 2000 or 2001. and lost my account 2times.
at last seti project stopped...sad. is this fail to find?

I hope we could find ET somedays.

or..if someone already found ET!?
so no need to try find!?

Read the very first post in this thread https://setiathome.berkeley.edu/forum_thread.php?id=85267&sort_style=6&start=0, then read the whole of the Nebula thread https://setiathome.berkeley.edu/forum_forum.php?id=1511 and you will see what is happening.

---

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

Seriously upsetting.

Just curious - read through the Nebula, most well over my head, but got me to wondering...
Though considering how long SETI ran, someone must have thought this (or it is totally wrong and irrelevant?):

Has anyone talked the radio telescope people into pointing their dish(s) at the furthest solid planet in our solar system that should reflect Earth's signals, and see what kind of data is returned by reflection off of it?
This would provide a sample of very strong signals to analyze and write algorithms to properly find real signals in.
This stuff is so far over my head - so if my question is meaningless - don't bother replying... but if it might have helped, something to think about.
Someone probably thought of this on the first day... LOL

Been with SETI since July 2000 - it has been fun, optimistic, and interesting...
Thanks and I will be ready if something comes up - like a Nebula analyzing BOINC app... ?

Todd

---

Just curious - read through the Nebula, most well over my head, but got me to wondering...

... Has anyone talked the radio telescope people into pointing their dish(s) at the furthest solid planet in our solar system that should reflect Earth's signals, and see what kind of data is returned by reflection off of it?
This would provide a sample of very strong signals to analyze and write algorithms to properly find real signals in...

That's a very good thought and a good question...

As it happens, I believe that s@h has indeed detected signals from our various space probes that are drifting across our solar system.

We also have a multitude of signals detected that are inadvertent radar/radio reflections off aircraft flying around various parts of the world! A large part of the Nebula work is to develop new algorithms that can reliably distinguish between the many signals that have been found that originate from Earth, vs another multitude of signals that are random noise, vs anything else found that is not Earthly and not noise.

The filtering out of all the known stuff to then leave only what is truly unknown is the really hard bit! Hence the very real research work developing Nebula...

Been with SETI since July 2000 - it has been fun, optimistic, and interesting...
Thanks and I will be ready if something comes up - like a Nebula analyzing BOINC app... ?

Todd

Hey! Good going.

Yes, this is still a world leading interesting project. One that has evolved from the early days of whether something such as Boinc was even possible, to now deeply investigating the signal riches that we have found.

There is a lot more to come yet.

There are just the minor details of researcher time, research grants/funding, and developing Nebula or a parallel successor to guide how we search further and farther!!

Keep an eye on the forums for "what next"!


Keep searchin',
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

Hi Todd,

With limited resources Radio Amateurs are doing something in that way since the Seventies. They are using the moon as a reflector to bounce fast telegraphic signals back to earth, the so called Moon Bounce. Also ionised atmosphere by meteor showers are successfully used, Meteor Scatter. Mars and Venus have been used to reflect signals back to earth in order to test communication equipment for amateur radio satellites.
As an initiation test, using simple reception equipment, you can hear signature signals of objects entering the atmosphere by tuning on 143.050MHz frequency of the Graves surveillance radar located in Dijon, France. http://www.itr-datanet.com/~pe1itr/pdf/The%20143.050MHz%20Graves%20Radar%20a%20VHF%20Beacon.pdf

Juan,
(amateur radio call sign M0WWA)

Joined with Pentium II back in 3rd of Apr 1999.

Thank you for good times everyone! :)

---