Slashdot is powered by your submissions, so send in your scoop

 


Forgot your password?
Close
typodupeerror
×
Science

SETI@Home For Linux 185

Posted by CmdrTaco from the truth-is-out-there dept.
Benny_Eggs writes "It's not the pretty screensaver version, but a SETI@HOME client for Linux is now available." For those of you unfamiliar with the project, Seti@Home is like Distributed.net, except instead of brute force encryption cracking, it searches radio signal noise looking for signs of intelligence.
This discussion has been archived. No new comments can be posted.

SETI@Home For Linux More

SETI@Home For Linux

Comments Filter:

  • This must become open source (Score:1)

    by Anonymous Coward
    I realize that there are reasons to keep this closed source to preserve data integrity, but we all know those reasons are dumb. anything you can do by hacking the source can be done through reverse engineering- and if someone reports a false positive, it's easy to check be running the data on a trusted computer.

    I sent a comment asking them to open source it, and I hope you did too.
  • This is not just an encryption contest, and much more is at stake.

    Not really. The problem here is that it's difficult to detect earth-like signals from other planets. (In other star systems.) According to the Sci.astro FAQ [astronomy.net], we couldn't detect a TV signal at a distance of 0.01 light years. (I'ts interesting that most of the signals that we transmit that _can_ be detected at that range we'd consider unimportant in a SETI search.)

    While this can probably detect an alien race transmitting a "beacon" meant for us to find, the chances of us finding earth-like leakage is minimal.

  • Well they don't even have the i586 one ready yet. I'm running the 686 one on my Cyrix 5x86/100 cpu with no problems (a 486).
  • Probably beta testers.
  • RC5 had source for a long time, but people kept modifying clients to give false reports.. not to mention use screwy compiler options which would result in broken binaries giving false reports.

    SETI@Home does have source available if you register as a developer if I remember correctly.

    --
  • There is no chance that people are going to find alien life this way - so why bother.

    And your basis for this statement is...??

    If there was alien life don't you think they'd alreay have tried to contact us??

    And what makes you think they haven't? Perhaps we just have not been listening?

    The laws of physics apply to the entire universe, not just here. If there's no such thing as faster-than-light space transport, then that applies to civilizations on other planets. That would leave something like radio as the only practical alternative for communicating (even if the communications turned out to be something as simple as a beacon - "Hello out there!"). So, perhaps there's a universe full of beacons, but we have not been listening for them. In that case, Seti@home could turn out to have a pretty dramatic payback.


    ...phil
  • Analysis of propogation loss of RF energy in free space, multiplied by effective radiated power, with some guesses about how weak a signal you can reasonably pull out of the noise before quantum effects step all over it.


    ...phil
  • The whole point of the 64 bit encryption is that 64 bit is used for all the millary and govenment encryption,

    I don't think you can support this statement. Skipjack is 128 bits. Triple-DES is 112, and easier to implement. I seriously doubt that you have access to classified information to show otherwise.

    when we crack the 64 bit encryption

    In 5 years? wooo.

    it is going to get all the governments and companys in the world extremely worried.

    Not likely. As I said, I don't think you'll find many governments using 64 bit encryption. The governments would do better to be worried about the NSA inserting backdoors in the encryption they do have (take a look here [counterpane.com]).

    Like I said, any political statement distributed.net could make has been made. Time to move on.


    ...phil
  • I'm indifferent on this myself, but there are alot of other things that should come before that. Kosovo, medical research, etc.

    If somebody could come up with a way whereby my spare CPU cycles would solve the Kosovo crisis, I'd have the entire company switched over tomorrow.

    As for medical research, there probably is some distributed analysis that could be run in this environment, probably some genetic research. But, unless you could convince me that the results would not be snapped up by some private corporation, then used to generate huge profits, then I'll donate my CPU cycles somewhere. That's an advantage to SETI - hard to determine a profit motive.


    ...phil
  • You can find a list here [mersenne.org].


    ...phil
  • Posted by jpepin:

    Or am I looking in the wrong place?

    Joe
  • I would love to have an xscreensaver [jwz.org] module for this.

    The question is, what kind of graphic visualization of the search should be done? What does their Windows client look like, anything?

    Perhaps the display should contain multiple kinds of data, maybe a map of the sky in one part of the screen, and a graph of area searched over time, or hits found over time, or something like that?

    Having it be an OpenGL hack might be interesting too: what if the "sky display" was a rolling sphere with a star-map drawn on it, with grid-cells that lit up for the area being searched, or the areas that had already been covered by others? (But maybe the SETI client doesn't work that way?)

  • I don't know what SETI@Home are doing, but certainly for distributed.net it's not hard to check up on people. You can generate a cryptographic "proof" that you've tried all the keys in a given block; then you sign all your proofs to say that they were all honestly generated. If you hand out a few duplicate blocks you can quickly detect if a given person is generating false proofs, discard all their data and publically humiliate them.
    --
  • A transparent firewall should work fine, as should a transparent http proxy. Unfortunately, for a network like mine, a http proxy has to specifically specified - and there's no way to do this with the current software.

    Any chance of proxy support? I've got about 15 UltraSPARC's that I'd love to put to some good use!
  • The FAQ says that it will work on a firewall that passes http. It says nothing about proxies.
  • Simple solution. Send the same data out to multiple clients. They should all come back with the same results. This is good scientific method anyway, it verifies that the clients are working and keeps human factors to a minimum.

    Regardless, the data should be analyzed many times by different groups to keep the results honest and error free. Its the Scientific method! :-)
    --
    Python

  • Need to correct myself.. It will not go thru a proxy. Most places use a proxy to serve up net content, not a strait-thru firewall..
  • There are several things that affect as to if a program can go thru a proxy or not:

    Does it support proxies?
    Does it support proxy authentication?

    Not all HTTP applications can go thru a proxy. It requires connecting to the proxy and telling it you want to pull another URL. YOU think about it some..
  • The only limiting factor I've seen is that it will NOT use a firewall of any type.. MAJOR limitation in my book.

    Back to distributed.net cracking, I guess..
  • Hey, that actually sounds like a REALLY neat project idea.. Do you know of any pages that actually have the math involed in this, and perhaps trajectory information of current known bodies?
  • As Jodie Foster in Contact says - there's no way they'd understand it. Which makes sense, as nobody on Earth understands it either.
  • They were plenty talkative with me. I admit that I did not have any questions. I just said, "here try doing the socket io like this and how about this for the console?" with working code snips from their code attached. The guy I emailed responded right away and said he'd check it out. Two days later he replied and said thanks, and that they'd incorporate the changes.

    Then I went off to work on something else...

  • http://huey.jpl.nasa.gov/~spravdo/neat.html

  • Or at least my suggestions even if they didn't use my code. It's up and running here and d.n is down the tubes. Too bad they never got v3 together. It was fun while it lasted.

  • How come they never release the source for any of these "idle CPU" clients/projects?

    ---------------
  • Big deal. How's the alien going to know it's porn? You're assuming that a) aliens know that there's something inherently private about a naked body and/or the act of procreating in our culture and b) what the Hell's going on in the picture anyway.
  • At one point during the development cycle, the source code for unix was available. I downloaded it at the time, and reported a few changes that had to be made to the code and the makefile to make it compile under Linux and AIX. It was mostly simple stuff like taking -lsocket out of the compiler options, and a few prototype changes, so I doubt I was the only one to figure it out. So, even though they might not have the source code right out there on the web page, it was available publicly at one point, and therefore anything but secret.

  • It's been nearly 2 years that the rc5-64 challenge has been running, and distributed.net has only covered like 6% of the keyspace.. taking 10 years to exhaust a single 64 bit keyspace does nothing other than to demonstrate that key-cracking big keyspaces is really, really hard.

    SETI@Home may have much worse odds, or it may have much better odds, we just don't know, and that makes it far more interesting. Attempt enough trillions of trillions of keys and you will crack it.. it's just counting, really.

  • Some do.

    GIMPS, [mersenne.org] "The Great Internet Mersenne Prime Search", is currently pullling more than half a teraflop [entropia.com] using idle cycles to find large prime numbers, and the source for its clients is at http://www.mersenne.org/source.htm [mersenne.org].

    --Joe

  • If there *is* a needle in haystack, and someone takes a nap on the haystack, Farmer John's Theorem clearly states that the needle will find either the left or the right buttock (depending on which hemisphere the haystack is located in) of the napper at the precise instant that said napper attains the Most Comfortable Position Possible within said haystack.

    What we need is to re-create the cosmic equivalent of the above described scenario, at which point Intelligent Life will undoubtedly find us, prick us in the rear and send us on our way, cursing and searching for a better place to take a nice, quiet nap.

    (yawn)

  • I think you may have missed the point here. If reruns of I Love Zorblak are indeed flying around in our corner of the galaxy then SETI programs have a chance of detecting them, whereas without such searches then we would not know that we are in the vicinity of such broadcasts at all. Your reasoning seems to be 180 degrees off target.
  • I just assumed there would not be a screensaver available for Unix platforms. However, I was peeking through the .txt files that are created after you run the client. What would stop us from creating our own visual display of the data? outfile.txt = interesting signals? work_unit.txt = the raw data Outfile contains things like: gaussian: peak=5.596647e-01 mean=5.280021e-01 ra=21.901 dec= 22.28 time= 2451246.16512 freq=1420360510.73 sigma=4.804805e+00 chisqr=2.706838e+00 fft_len=8192 chir prate=7.105420e-01 I'm no astrophysicist but I got enough from Contact that ra must mean right ascension (or something like that), dec == declination, time is time, there's the frequency, and the rest I'm a little unsure about. work_unit.txt contains things like: start_ra=21.888 start_dec= 22.16 end_ra=21.928 end_dec= 22.53 angle_range= 0.666 time_recorded= 2451246.16476 (Mon Mar 8 07:57:14 1999) subband_center=1420355832.96 subband_base=1420351560.50 subband_sample_rate=9765.62 fft_len=2048 ifft_len=8 subband_number=36 receiver=ao1420 nsamples=1048576 tape_version= 1.30 num_positions=22 [$num_positions coordinates] end_seti_header [DATA] Couldn't we figure this out and make a GTK/SVGALib app that polls the files and displays pretty pictures? And hell, that program could communicate with some type of master server. It would say, "Hey, I'm working on this part of the sky". And you could go to a web page and see a generated imagemap of the sky that contains color-coded regions of places you've checked, places everyone has checked, etc.. and you could click on those regions to see the peaks. And there could be a top-10 list of peak signals. This doesn't sound unreasonable to me. I'd be willing to work on the webpage end and contribute what I can to decoding the text files.
  • Why don't we get a mailing list together?

    The biggest hurdle will be decoding the data but I think that can easily be overcome.

    You said that your problems with the people at Seti were that they wouldn't give you any details about the data, right? Maybe we can find one of the source releases and go from there?

  • I just assumed there would not be a screensaver available for Unix platforms. However, I was peeking through the .txt files that are created after you run the client. What would stop us from creating our own visual display of the data?

    outfile.txt = interesting signals?
    work_unit.txt = the raw data

    Outfile contains things like:

    gaussian: peak=5.596647e-01 mean=5.280021e-01 ra=21.901 dec= 22.28 time= 2451246.16512 freq=1420360510.73 sigma=4.804805e+00 chisqr=2.706838e+00 fft_len=8192 chir
    prate=7.105420e-01

    I'm no astrophysicist but I got enough from Contact that ra must mean right ascension (or something like that), dec == declination, time is time, there's the frequency, and the rest I'm a little unsure about.

    work_unit.txt contains things like:
    start_ra=21.888
    start_dec= 22.16
    end_ra=21.928
    end_dec= 22.53
    angle_range= 0.666
    time_recorded= 2451246.16476 (Mon Mar 8 07:57:14 1999)
    subband_center=1420355832.96
    subband_base=1420351560.50
    subband_sample_rate=9765.62
    fft_len=2048
    ifft_len=8
    subband_number=36
    receiver=ao1420
    nsamples=1048576
    tape_version= 1.30
    num_positions=22
    [$num_positions coordinates]
    end_seti_header
    [DATA]

    Couldn't we figure this out and make a GTK/SVGALib app that polls the files and displays pretty pictures?

    And hell, that program could communicate with some type of master server. It would say, "Hey, I'm working on this part of the sky". And you could go to a web page and see a generated imagemap of the sky that contains color-coded regions of places you've checked, places everyone has checked, etc.. and you could click on those regions to see the peaks. And there could be a top-10 list of peak signals.

    This doesn't sound unreasonable to me. I'd be willing to work on the webpage end and contribute what I can to decoding the text files.
  • .... netrek has used 'em for years without trouble. However, the source _should_ be free, if only so people can contribute patches and see what _exactly_ they are running. You can download source to various netrek clients but those clients cannot connect to blessed servers, because the netrek world had this very same problem: clever folks would 'borg' their clients giving them auto torp det, auto plasma lance, auto phaser aim, etc..

    Just don't give the private blessing key to anyone.
  • Currently, their biggest bottleneck is that all of their data is stored on magnetic tapes that aren't in a library (read as, you have to manually stick them into a tape drive). I'll try to find the URL when I get home.
  • Linux is first, but I would like to have teems
    as d.net.

    OverLord



    linux-gnu:
    Work units sent: 1940
    Results received: 793
    CPU time: 1376 hr 04 min 16.2 sec
    Average CPU time per work unit: 1 hr 44 min 07.0 sec
    SunOS 5.5:
    Work units sent: 699
    Results received: 628
    CPU time: 433 hr 52 min 37.5 sec
    Average CPU time per work unit: 0 hr 41 min 27.2 sec
    unknown:
    Work units sent: 869
    Results received: 576
    CPU time: 4249 hr 01 min 47.2 sec
    Average CPU time per work unit: 7 hr 22 min 36.4 sec
  • Any one know of any big names going to implament this....I can imagine it now.. NASA's currently #1...=P


    I'm not sure what your getting at, but NASA bailed on any SETI research in 1992, or early 93. Basically, they had a program called "HRMS", the High Resolution Microwave Survey. Congress got wind of it, and in a flurry of budget cutting and snide comments about a "little green men", they canned the program.

    Luckily, the fledgling SETI Institute was able to get NASA to donate the hardware that had been built and used it for the first runs of Project Phoenix. Except for that initial burst of govt. hardware, Phoenix has been completely privately funded.

    I'm actually really happy that NASA won't touch SETI. It leaves the entire field open to enterprising scientists, instead of it being run by the rather fickle adminstration in NASA and Congress.

    Some individuals inside NASA might run the screensaver, but it won't be an agency-wide project, that's for sure.
  • Waste of time that it maybe, at least it's just idle time that would otherwise be wasted on most people's computers. Personally, even if the project doesn't find signals from intelligent life, perhaps some new cosmic elements could be identified, such as a new pulsar. They said that more radio telescope data is being collected than can be processed by the astronomers. Why not use the power of all those wasted idle cycles to help process that data (or factor mersenne primes or work on rc5, etc.)..
  • I hope they would not rely on a single client for the data analysis results on any given block. If they have any smarts at all they'd send the data to several client and compare the results. Erroneous data could be ignored.
  • Because then you'd find out that it's just a front for the NSA to crack non-critical PGP encrypted messages.
  • cause on the stats page i saw this

    Windows NT: 4.0:
    Work units sent: 86
    Results received: 47
    CPU time: 548 hr 33 min 42.3 sec
    Average CPU time per work unit: 11 hr 40 min 17.5 sec
    Windows 95: 4.10:
    Work units sent: 123
    Results received: 47
    CPU time: 340 hr 34 min 44.3 sec
    Average CPU time per work unit: 7 hr 14 min 46.9 sec
    Windows 95: 4.0:
    Work units sent: 90
    Results received: 30
    CPU time: 177 hr 08 min 55.2 sec
    Average CPU time per work unit: 5 hr 54 min 17.8 sec

    whats up with that?
    --

  • If they do find something you dont think that they would but the name of the owner of the machine that found it. That would a heck of a press release to be associated with!
    --
  • I too have been glad to participate in d.net but we've proven our point; now we need something interesting to solve that isn't so esoteric as a "Goloomb ruler". Finding space aliens does it for me!

    I'm moving over; if S@H could get a team stats engine going, I'll bet a lot of /.'ers will move to.
  • I thought the source code problem was due to restrictions of exporting encryption code to agents outside the USA/Canada. I realize that the algorithm is freely available, but as I understand the export restrictions, it's the source code that cannot be sent outside the country, not in electronic form anyway.

    As for non-encryption idle CPU tasks, the blessed client reason does seem to make sense, but only when combined with some sort of authentication mechanism.
  • It might be a bit of a stretch, but perhaps with enough interest, the distributed.net folks could help out the SETI folks. As many have stated in previous posts, it's only a matter of time for the 64bit encryption to be broken. While looking at encryption strength is important, there are more interesting things for that much computing power to work on.

    Unfortunately, there is no contest involved, and no guarantees of finding anything at the end, but there is always that chance...
  • Sorry, I didn't mean that you wouldn't be credited, what I meant was that there is no guarantee that anyone will find anything.

    I agree tho, just the chance to have my name associated with that find is worth it!
  • We don't need another idle CPU project to deminish the power behing more useful projects such as distributed.net.

    Nobody is forcing you to use your CPU idle time. It's voluntary. And how is distributed.net more useful? It's just a game of luck with no purpose, see who can find the key the fastest. We already know it can be done, it's been done before. Woopee-dee.

    There is no chance that people are going to find alien life this way - so why bother.

    Why bother? Because there is no chance that we are going to find alien life by sitting on our asses. Yes, the chances are small, but a small chance is better than no chance.

    We haveto think closer to home and work on getting the US government to change their policy on encryption.

    I'm indifferent on this myself, but there are alot of other things that should come before that. Kosovo, medical research, etc.

    If there was alien life don't you think they'd alreay have tried to contact us??

    Exactly! Say ET is out there somewhere trying to find us. ET beams a big signal saying "hey, I'm here". But what good will that do if nobody is listening for it?
  • If you think that mapping the distribution and trajectories of masses in our solar system (asteroid detection to avoid near-earth interaction or collision) is a good idea (and I agree), why not organize a public project using an open source-designed algorithm to analyze the data. I would certainly assign two of my four pcs to the task. Maybe all of them if someone really tried to convince me!

    Carping about SETI@Home is not the answer.

    There are enough idle pcs out there to do both jobs.

  • Within X ly, there are no civs X years ahead of us. If there were, we'd be flooded with reruns of I Love Zorblak. All of the trekies in the world with this software won't change that.
  • We'd find them easily with no need for this.
  • This is a serious advantage rc5 has over most of the other distributed computing projects--it's got a microscopic memory footprint.

    d.net's only looking at 64bits at a time.

    ok, that's facetious, but they're trying to decrypt a one-line message. Seti@Home is trying to analyse radio signals, and the unit of work is significantly bigger--350K on my machine. It's constantly doing fourier transforms of that and searching for peaks and trends. I don't think you could do that efficiently is much less memory.

    So yes, until everybody's got 128MB of RAM, this will limit participation.

    You could try the GIMPS project ( www.mersenne.org [mersenne.org]). It's not as sexy perhaps, but still new territory. My client weighs in at 1.6 MB.
  • /dev/audio isn't very smart about file formats--it's just a dumb device, so you pretty much have to give it raw audio data.

    look at the work_unit.txt file. It's got a header, followed by some printable-coded binary data. One has to figure out how they've does the character encoding.

    my guess is some sort of hacked uuencode. The faq says the portion out 0.25MB data blocks, and my work_unit.txt is 320K--about the right ratio for uuencodes 3->4 byte expasion.

    Then, it's unlikely to be in 16-bit stereo, so one has to figure out what kind of samples they used. Probably single channel but 8-bit, 16-bit, what? Or, goddess help us, some kind of float--scientists seem fond of that datatype, despite the portability problems.
  • I responded to the initial call for developers back in july. I fact, I think I saw the announcement here, but looks like Rob's deselected that article. :(

    Anyway, they did answer their emails in my experience, but it quickly became apparent they weren't interested in an Open Source development approach, so I told them I wasn't interested in contributing.

    Mostly they cited fears of data-subversion by hacked clients. As has been mentioned in other comments, there are ways around this, and in the long run I think they're worth the extra effort. Security through obscurity only keeps the honest people out.

    I think they still don't understand the benifits of open development. In my original contact with them, it didn't feel like a rational decision, and I've seen little evidence they've improved in this respect.
    This was something that always bugged me about d.net, too--they wanted my cpu, not my involvement.

    Sad, since I'm sure it would have been finished sooner. The basic engine seemed to be complete at the time. I think server funding was part of what held them up, but still.
  • Fabulous idea--too bad the data's encoded.

    Looks sort of like uuencoding, but I've not managed to parse it yet. Any ideas?
  • "Mars needs women!!!"

    Leonard
  • The FAQ for the client states that it only uses HTTP as a protocol and should have no issues with a firewall. If I read this correctly this means it should be able to use a proxy as well as work if you have a masqueraded network. Of course, I haven't tried this yet but will do as soon as I can.

    Leonard
  • Well this one, at least. About a year ago I volunteered to write the X11 screensaver for them. They agreed, and portioned off the project to me. Then when I needed to correspond with them concerning technical details they never answered their mail; it was as if I didn't exist to them. I also had some concerns with their code regarding portability to various Unix flavors, and I volunteered to write autoconf scripts for them. This too fell on deaf ears.

    Now suddenly the clients are available. I was very eager to be a part of this and contribute code, but they seem to have quietly shifted projects to other developers, and left the originals in the cold. What's the deal? I very badly wanted to run these clients on several of my boxes, but now I'm not sure how I feel about this project. Has anyone else fallen into their email black-hole?

  • Not a bad idea. If you plan to do this, I recommend you create a module for Xscreensaver. This is how I had planned to do the screensaver for them before they went silent.
  • Please see my comment above (they blew off their developers.) So maybe it wasn't written afterall. I just assumed that they handed it off to someone else.
  • No, I'm a professional research programmer. And I would have been fine if they found someone better suited to the task, had they at least answered my email to tell me so. But I did not know about Sun's involvement- it may have been a factor. Thank you for mentioning that.
  • There are several factors in play here which should lead to an X11 graphical display being produced eventually:

    a) More workunits are being processed by unix than other 'OSs' - why only provide the pretty graphics for a minority of users?

    b) A sizeable percentage of those unix users (esp. linux) are able to program and may offer their services to the project

    c) The unix people could collectively withhold their cycles until their demands are met :)
  • Although I agree it would hurt d.net stats with more and more powerful
    proceesors around it shoudnt be to bad also if like me you have access
    to mor then one machine you could fust run the new client on one m/c.
    We must search for E/T life, it is the one thing that could unite
    the planet
  • The whole point of the 64 bit encryption is that 64 bit is used for
    all the millary and govenment encryption, when we crack the 64 bit
    encryption it is going to get all the governments and companys in the
    world extremely worried. Then when that happens we have a go ayt 128
    bit. The exercise is to keep these organisations on their toes and not
    to give them an easy time
  • It's here [berkeley.edu]. Well, actually it's not, but you can give your e-mail address to be notified when it's released later this month.

    And it looks so nifty!!

    Mike
    --

  • Calibrate the zero-level by pointing dish at Redmond!
  • What's so cool about it. There are no graphics, a big drawback if you want to see whats going on. Plus, the code is not parallel. There is no source so we can't toy with it. I don't know, I just hope the windows version runs under wine...
  • Topic says it all
  • Four,
    if your interested, I can pass you the Win code and you can get an idea of the class structure and inner workings. Don't try to port it though...its about 9 months old and it's MFC!:0>

  • Actually the source for the SETI@HOME project was available over a year ago if you registered as a volunteer developer. I have 3 different version of the Windows client laying around my PC. If you contact the director, you may be able to get an older version, or even the latest version. The screen shot of the new version looks 100 times better than the version I hacked away on.

    I wonder if they have a KDE or GNOME version yet (hehe)?

  • Check the listing for top users. One guy has an average time per work group of 10.3 seconds. It took me 4 hours to do one. He's running it one something VERY nice..
  • Current activity of my box. Goodbye distributed.net, hello sETi.
  • Erroneous data will be ignored. If a match is found, the chunk of data sent to that machine will be tested by SETI itself to make sure there are no false reports. So while it may be possible to send back faulty data, with all the double checking that will go on, actually fooling them is basically impossible.

    -B
  • There is no chance that people are going to find alien life this way - so why bother.

    A wise man once told me:

    If you try something the answer may be no, but if you don't try, the answer is no. Those two words make a world of difference.

  • I'd be curious to know how you define "pseudoscience". By my understanding of the term, this meets none of the criteria.

    BTW, Carl Sagan was a CSICOP fellow and his enthusiasm for SETI projects is well-known.

    Mark
  • It is very possible that Sun offered some resources and personnel ( or their employees offered to help). Unfortunately, in that case, they may have chosen professionals rather than a student (I'm assuming that is what you are).

    In one respect, it was very unprofessional for them not to respond. On the other hand, a proven professional offering to code vs. a student is a wise business decision. (Considering their limited private funding)

    My $.02
  • Definitely beta testers. I'm as anxious as the next person...
  • Comment removed based on user account deletion
  • How about this, if we find Alien Life, the prize will be one Lustful and Steamy night with Jody Foster. I'll put all my p-450's to work asap :-)
  • I think there's a big deterrent to the vast majority of folks who would run this: it'll hurt their d.net stats!

    There are now several worthy causes to donate idle CPU to. I hope soon there will be more. I want to be bewildered by choices!
  • look at the bottom of the page,
    i686-pc-linux-[glibc|libc5|static]

    Yes, it runs on my P54C

    -- A wealthy eccentric who marches to the beat of a different drum. But you may call me "Noodle Noggin."

  • Are you sure about that? I could swear I read a while ago that it would. That anyone who had http trough a proxy would be able to use this.

    sure here it is! (from their FAQ)

    Are there any issues concerning the SETI@home software and firewalls?

    SETI@home uses the HTTP protocol, and should work through any firewall that allows outgoing Web traffic.

    http://setiathome.ssl.berkeley.edu/faq.html
  • I have a linux box (486) that just does masquerading for my machines at home here, I just fired up setiathome on my PII450 box and it connected thru my firewall fine, so I'm happy. :-)
  • well, duh... read the faq, it says they will have to verify the hit by analyzint the data themselves. Do you think they will announce they found ET when Johnny cracks the code and starts sending false positives?

    I don't think so...

    (not that I disagree that it would be nice to get the source, just don't follow the argument)
  • unless, of course, YOU are wrong.
  • Nuh-uh! Looking for primes is pretty cool, I say. At least we have proof that the primes are out there! :-)

    Besides, I can think of a lot of things I could do with a cool $50,000 [eff.org]!

    Then again, I'm a math nerd, so I'd be looking for primes anyway. Hey, I think I see one now...

  • O come on. People aren't going to be running their own unique hacked version of the client. We just want an open development environment so the software can be improved. There will be "official" releases that everyone uses from the main web site, just like every other free/open project (kernel, gnome, kde, etc.).

  • The same server I'm using to post book reviews of science fiction is now processing SETI data. There's something so perfectly enclosed about that!

  • Ok, this is great and all, but what happens one day when a well-written macro virus reverse engineers that big ass satellite and beams all of the internet porn to some life form out there? Intelligent or not, I'm sure they are going to have something to say back.
  • Heh, I'll tackle that one after work. Don't want the firewall guys coming after my ass. I push it buy coming here all the time.

  • How come they never release the source for any of these "idle CPU" clients/projects?

    They want to be sure that noone runs bogus clients. In something like this, one bad result could destroy the whole project. (For example, supposed the d.net client returned a no rather than a yes when it found the RC-foo key.)

    This isn't to say I completely agree... I'd probably prefer a system where a server checks out open-source clients to see if the're not bogus. (Throws some know problem sets at it every once in a while, and sees that it answers correctly.)

  • Distributed.net set out to prove something, and it succeeded. Since then, it's been proved again and again, most recently with EEF's dedicated cracker. Working on RC64 is now just flogging an expired equine. It's just not going to make the kind of political statement that RC56 made.

    As far as a tool set, d.net had a chance to get their distributed engine out, but they seem to have missed their window. Now, we've got several dedicated distributed efforts going, and d.net is rapidly turning into a fading blip.

    My only question is to figure out which machines to switch over from doing primes. Find a new prime, get your name in the history books (well, the math history books). Find a new civilization, get your name in the real history books forever.


    ...phil
  • At 13 megs don't expect to run this client except for bedtime. Don't expect to find extraterrestrial life until 13 meg level 2 caches come along either. Since it's still better than trying to crack passwords, I'll probably dump the rc5 client for a while.
  • The surest sign that intelligent life exists is that it hasn't tried to contact us.
  • There's something that worries me about this. There will probably be some kind of stats or rankings, and people will compete to get to the top. That's fine, but what if somebody develops a client that just drops the blocks so they can get to the top of that list? I heard something like that was developed for rc5.

    This is not just an encryption contest, and much more is at stake. I would hate for it to be ruined just by a bunch of stupid people.
  • My guess is that having the source for it would greatly increase the capacity of malevolents to send fake or distorted results back.

    I wish I knew of a good counterargument. Suggestions?

  • I've been watching SETI@Home for a while and have been dying to download the client. Sure, its fun to chip away at RC5, but we all know that RC5 64 bit will eventually be cracked given time.

    With SETI@Home, you can use your idle time for something that may have a greater purpose down the road.

    Bottom line is, brute force will eventually crack encryption. Woohoo. Increase the bits. Now that this new idea has come along, you have a chance to aid the search for intelligent life.

    And that my friends, is a big deal.
  • They'll be lucky! I have enough trouble finding signs of intelligence searching the FM band. I shudder to think how much worse it would be at a non-radio station frequency!
  • I am interested in doing this but i could not find any info on the site as to whether the screensaver version for win32 would be unobtrusive with other applications running in the background like say an FTP server. I would assume it is but... then again....

Slashdot Top Deals

Intel CPUs are not defective, they just act that way. -- Henry Spencer

Close