Still running Seti classic. Don't like the new stuff at all. Have completed 8,000+ work units and am seriously considering leaving when they shut down the classic side of the house. Any other "old timers" feel the same way?

---

I have not been with S@H as long as you have, but I did run S@H Classic for a couple of years. I didn't like it much, as it required several 3rd party tools to run it smoothly.

BOINC is one single tool that just works. I have no problems with it. But then, hey, I always like new software.

---

I having been running SETI in some form since before you started. I changed over to BOINC nearly immediately after it was made available. BOINC is far and away the best platform under which to run SETI.

---

Classic is over six years old. How many programs are you using right now that are that old?

Hi, Since they have turned the search for life in the cosmos into the search for university research cost savings I have no interest in running "Boinc" not that it has ever worked anyway. The server is down again.

---

Hi, Since they have turned the search for life in the cosmos into the search for university research cost savings I have no interest in running "Boinc" not that it has ever worked anyway. The server is down again.

No, everything's running fine!

02.08.2005 12:10:52 230 Computation for result 26fe05aa.10476.13216.897148.177_2 finished
02.08.2005 12:10:52 231 Starting result 28ap04aa.15206.26657.767346.70_3 using setiathome version 4.18
02.08.2005 12:10:53 232 Started upload of 26fe05aa.10476.13216.897148.177_2_0
02.08.2005 12:10:56 233 Finished upload of 26fe05aa.10476.13216.897148.177_2_0
02.08.2005 12:10:56 234 Throughput 793600 bytes/sec
02.08.2005 12:17:12 235 request_reschedule_cpus: project op
02.08.2005 12:17:13 236 Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
02.08.2005 12:17:13 237 Reason: Requested by user
02.08.2005 12:17:13 238 Requesting 0 seconds of work, returning 1 results
02.08.2005 12:17:15 239 Scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded

(all times are UTC)

Hi, Since they have turned the search for life in the cosmos into the search for university research cost savings I have no interest in running "Boinc" not that it has ever worked anyway. The server is down again.

You seem to have missed the point that SETI is not funded by Berkeley.

I too ran seti classic for many years and accumulated more than 11K units. I think that the policy of NOT converting the classic work accomplished to the new boinc unit measurement method is a very negative demotivator. Enough so that I think that the project will loose many participants. Starting everyone from scratch feels like getting slapped in the face. Sort of, "thanks, but no thanks". There are certainly enough statistics available to convert this with a high degree of accuracy. I've seen some discussion regarding some 'cheaters' that exist in the community. There should be plenty of data available to detect who they are/were and remove credit that they don't deserve. I'm sure that this would take a lot of work and computing power though. But demotivating the masses for the acts of a few is NOT the way to keep people involved, particularly when there is no other form of 'payment' other than this work unit accomplishment.

Some people have remarked that boinc is technology that "simply works". This has NOT been my view. There have been many improvements, but it is far from fully debugged. I have a work unit sitting on one of my machines right now that will probably be lost because the boinc server locking mechanism is broke. Some of the comments you see suggesting that old software is not as good as new software simply spins me up. It reminds me of the Oracle sales people during the Y2K days. They were pushing people to get off 'legacy' code when in fact there is a boat load of code within the Oracle environment that was more than 20 years old (at the time). This is true of virtually every major software system out there. There are still routines in Windows 2003 that come from the original MSDOS! Software is an evolutionary process, not a replacement. Even if the code is completely rewritten, the methodology is usually very similar if not identical.

Boinc installation is an area that needs a lot of improvement. One should not need to be able to write startup scripts in order to automate it (unix/linux). The idea that thousands of people need to do it on their own is simply mind boggling! What a waste of time and talent. And how about those that don't have either the time or the talent? I guess that's a population that will never use boinc as it stands today.

One thing that I have noticed in the short time that I've been running boinc is that the new measurement technique seems to favor new hardware. Well over 95% of my completed work units end up having the claimed credit reduced. My guess is it because all of my machines are in the 500 to 800 mhz range. But I also notice that credit is much less on the linux and unix machines. It's not like these machines are busier than the windows boxes either. This is something that I will need to watch and investigate further.

Way back when, I also installed SetiQueue to take care of the communication issues associated with dial-up. Boinc has the capability to handle the caching of work units, although it is poorly documented. So far, the closest replacement is BoincView. It certainly will display much more information that it knows about. Unfortunately, it has some serious flaws where it can easily miss client transfers and does not have the capability to recapture or validate any of this history with either boinc messaging or the user and work unit information available on the seti website. It is also much more cpu hungry and creates a lot more network traffic too. The statistics, graphing, and history recording functionality of SetiQueue is what I really miss.

I'm sure that there are others that have issues with moving to boinc as well. The bottom line is that the people in charge of the seti project should probably resolve most of these issues before they shut down classic processing. Granted, some of this involves third party software. But they should not loose sight of the original goal. Seti needs outside help in order to get the processing done. If some of the third party functionality is needed to keep people involved, then I guess this needs to be addressed before they pull the plug. Once the people are gone it will be very difficult to attract them back.

Yep, fully agree to these feelings. Running Seti now since 2002 withot any problems, I am suffering with Boinc some serious problems with up- and downlading WU's. OK, I admit to be a tech dumbo, but hey, they want our computer time. So shouldn't they feel responsible to provide an easy to use system? Electric power is quite expensive here in good ol' and bancrupt Germany. So I am thinking seriously to switch my three machines of. Win a little bit - loose a little bit.

---

There is another reason for starting everyone with 0. Not everyone has access to the email account that is used for classic or their password. So those people cannot transfer their account data.

The amount of work to detect the cheaters and remove them is large, and the work force is tiny.

There really is not a good conversion from classic WUs to CS. The problem is that different versions of the classic software did different amounts of work on the science (remember the screaming when 3.1x was forced on everyone?)

S@H does not run as well on machines with small L2 caches (512 Kb seems to be the minimum that staves off thrashing) - even a new CPU with a small L2 cache suffers. It is, however, reasonably fair in that everyone that crunches the same WU and gets a valid result gets the same credit.

The devs are aware of the setup issues (maybe not all of them) and are working on solutions (this is at least part of the reason that classic is still running).

---

BOINC WIKI

I too have been running Seti@home for several years (couple months more than the original poster) and also DO NOT like the BOINC software. I am a security freak and a speed freak when it comes to my computers. I liek them running at their best all the time. Or atleast with very little drag on the CPU. The Classic software NEVER gave me any problems. I read one post that said it needs several 3rd party addons to run smoothly....HUH??? I do not use any 3rd party addons and it has run fine since May 99.

I have to say this about the BOINC software though, I do like the new interface and the screensaver is very cool! I just can't deal with all the CPU usage and will most likely drop the service if that is not improved.

one more thing....DAMN! 8000+ units processed since July 99? You must have a Cray or several computers running. I have 4 running it for most of the time I have run Seti and am only just approaching 1500 WU processed.

Anyway, I vote for the Classic software!

Koz

---

I have to say this about the BOINC software though, I do like the new interface and the screensaver is very cool! I just can't deal with all the CPU usage and will most likely drop the service if that is not improved.

?
What CPU usage?
On my system BOINC generally shows 0% usage.
Seti uses whatever is available- as other programmes need CPU time, they get it. No problems.

---

Grant
Darwin NT

The response lag the some people are seeing is because BOINC projects can do a log of file IO. I found that turning off either read checking or write checking on all of your virus scanners and malware scanners will do a great deal towards improving the responsiveness of the computer.

---

BOINC WIKI

I`m also an entousiastic SETI-classic user (3000+) but I don`t like the new BOINC stuff at all and called it quits. BOINC seemed to go messing with my computer and being a systems manager that posed to great a risk for me.

Sad though.

Success with the BOINC thing, but I think you`ve just killed SETI.

---

I can't even get this new software to work, let alone let it run to discover a plethora of bugs that everyone else seems to be complaining about. If I can't get this to work, and the Classic software is getting dropped, then I will have no option but to also discontinue as a client. This saddens me because of the reason I am part of the project - not to accumulate credits as if anyone actually cares - but to just be part of something that may turn out to be one of humanity's greatest discoveries. If a few people are cheating the system - so what? I can't use software that doesn't work. Whoever has received public money to construct this inferior software ought to be very dissappointed in themselves... I'm very dissappointed.

As for seti@home project managers... please reconsider retaining the Classic as an option... you're infringing on my freedom of choice and the penalty will be that you'll simply loose your clients.

---

PLEASE let us have our OLD SETI. I have just about 7200 Work units. I started in 1999 I believe with a computer that took 64 HOURS to do 1 work unit. Today I can turn them in 3 hours. I wanted to get to 10,000 Work Units.
I tried the NEW program and it about burnt my system up. I run OC'd anyway but this was Nuts.
I would atthe very least like a link to the "OLD" SETI...

Am I wasting my time with Classic now???

---

Am I wasting my time with Classic now???

Unfortunately: Yes. Berkeley already shut down the account creation and it'll take just a few months until Classic is shut down.

Try BOINC. It isn't that difficult it looks like. Install it, attach to Seti@home, make a few settings in your General preferences and BOINC does everything for you (downloading, uploading, reporting, caching etc).

---

since 29-Jun-2004:

If a few people are cheating the system - so what?

This is a science project. Corrupted data isn't acceptable.

I can't use software that doesn't work.

The software does work, so what's your problem?

... you're infringing on my freedom of choice ....

They are what?
I hope you eventually get a better grip on reality.

---

Grant
Darwin NT

If a few people are cheating the system - so what?

This is a science project. Corrupted data isn't acceptable.

I can't use software that doesn't work.

The software does work, so what's your problem?

... you're infringing on my freedom of choice ....

They are what?
I hope you eventually get a better grip on reality.

Well... this is hardly constructive in overcoming a transition period that, if you peruse the message board, I’m not the only one experiencing problems in.

As for the first comment… the logic that follows is that SETI@home has been collecting corrupted data for years, potentially since its inception, and is therefore must be scientifically invalid… I was certainly not under the impression that the corruption would be so bad and if it were, then I would have hoped that such a critical contamination would have been addressed long ago. Given the limited resources of the SETI program, I accept that new software should be developed that can enhance data quality with much greater efficiency than what has been done in the past… but no methodology is impenetrable to confounding elements – as any good scientist would know – especially when the methodology explicitly relies upon many of use mere humans interacting with the analytical process. These confounds need to be weighed against the ability to maintain the altruistic spirit of computer users world wide in allowing an external group of researchers to access (in many cases) our personal electronic space. Welcome to science… where methodological ideals must be balanced with real world practicalities.

As for the second comment… my problem is, as if this were not clear, is that I can’t get the software to work!! While I do not have much in IT know-how, I have enough experience to download software from the Internet and even try to navigate around technical integration issues… I even had technical support personnel at my workplace try get it to work. But this is not in the spirit of the SETI@home project… I was under the impression that it was supposed to be accessible to everyone in order harness the analytical power of many many small processors, not just the educated elite. Thus, the user interface needs to be easy to access, easy to install, and easy to maintain – a characteristic of the Classic software – which to me, is not immediately apparent with the new software. Otherwise, the entire methodological approach needs to be revised.

And the third comment… well, which reality do we want to live in? Last I checked, we Westerners live in a sociopolitical ideological environment that sanctions freedom of choice… so much so, it is written into constitutions and we are in fact in a war in order to ‘protect’ this ideology, and import into other cultures that have not traditionally consisted of this ethos. The notion of ‘experts’ authoritatively decreeing to the populace – which in this case, socially, politically and economically supports the program– about how this knowledge will be constructed, is unrealistic, and most people familiar with the sociology of scientific knowledge (SSK) will know this. Until the software is developed to a point where anyone interested can participate then, given our sociopolitical ideologies, I think it is inappropriate to force an analytical tool upon clients who are ultimately providing the power to do this project. If this methodology is insufficient, then perhaps efforts should be honed on generating greater political support to fund a ‘supercomputer’ to analyse this data consistently, which the whole SETI@home concept was designed to overcome.

Hey… if you don’t like our participatory democratic model where people have choices, then perhaps try live in China, where authorities do dictate what, who and how are done. Otherwise the reality is, we have choices and as they become limited, the last choice may be simply to not participate. I hope the SETI@home researchers are able to balance the conflicting ideal of scientific rigour with methodological practicality – but I would think shoving one approach down the throat of those who are prepared to simply help out and participate, is a not particularly rational or practical strategy.

In any case… this is my opinion… like it or not, but this is the world of peer review.

---

Am I wasting my time with Classic now???

Unfortunately: Yes. Berkeley already shut down the account creation and it'll take just a few months until Classic is shut down.

Try BOINC. It isn't that difficult it looks like. Install it, attach to Seti@home, make a few settings in your General preferences and BOINC does everything for you (downloading, uploading, reporting, caching etc).

---