• Trying searching the web for a term. Especially good for technojargon. Google is a very good search engine for this purpose. Example:

  http://www.google.com/search?q=dmca

• Try an online dictionary. Merriam-Webster is useful and free, and the Oxford English Dictionary is available to on-campus users. webopedia claims to be an online dictionary for computer and internet technology and has a definition of bottleneck that matches how Rich used it.

From Anti-dmca.org:

The DMCA is the Digital Millenium Copyright Act, passed by the U.S. Congress in 1998, supposedly to update copyright law for electronic commerce and electronic content providers. Unfortunately, this law is very poorly written, and is now regularly used by corporations to restrain the three primary concessions of copyright and otherwise prevent free speech activity.

1. People need to stop taking this stuff so personally. If you are in the top 10, even if you are (gasp) #1 on the list, no one is calling you a bad person. I'm sure that Rich and everyone else truly believes that you are a beautiful and unique snowflake. As someone who has made the list a few times myself, I just don't understand how anyone can be offended by the discussion taking place here. The numbers don't lie - if you are sending or receiving a lot of bits, that's it.

2. This discussion is *important*. It's about the regulation of a shared and limited resource. Those two factors combine to make the problem one of balancing. What Rich has asked for from the beginning is input on how best to perform that balancing. The same sort of discussion is taking place all over the country, as university administrators realize that bandwidth is expensive and will (usually) be filled as quickly as it can be purchased. For all of the failings Brandeis may have, I'm very glad that this school's approach has been one of conversation and debate, as opposed to adminstrative decisions imposed from on high. So let's try to be a little more mature about it, shall we?

3. If you want to take part in the debate, it is your job to familiarize yourself with its terms. I am relatively confident that none of the posters have used jargon in order to cloud the issue or to enhance their egos. Terms like DMCA, bandwidth, and bottleneck all have fairly standard definitions. Explaining them every time they're brought up is a waste of time, especially in a forum that is, by necessity, technically oriented. You're college students (maybe grad students, for all I know) - demonstrate some of that intellectual curiosity that got you admitted to the 'Deis in the first place and look something up.

4. Sorry this is so long. I'm a COSI major, but I'm also applying to law schools for next year. So you get a treatise that is not only crammed with technical jargon, but is also mind-numbingly long.

http://news.excite.com/news/r/011130/07/net-tech-filesharing-verdict-dc

http://www.theregister.co.uk/content/6/23107.html

http://slashdot.org/articles/01/11/30/0537210.shtml

I think it's overly simplistic to say that people are being punished for "making use of a resource." The warning e-mails and access restrictions come about because, practically speaking, some people's usage negatively affects other people. When their bandwith usage overwhelms others', I think that some sort of a line has been crossed. Brandeis provides us with a resource in the form of bandwith, but that does not mean it's an unlimited resource. That's why it's fair for the school to try to ensure equal access to everyone.

Remember Rich's discussion of Morpheus/KaZaA traffic and his fun neo.php script, I decided to start a catalog. This catalog has two purposes:

1. To allow people to search for and access files that are shared by other campus users over Morpheus/KaZaA.
2. To allow people to search for and access files that are available on the local Windows network.

• My computer does a service scan of all the dorm subnets, looking for computers with open port 1214 (Morpheus/KaZaA).
• This list is fed into a spider that goes and fetches each page and parses the data into a MySQL database.
• The MySQL database is indexed for full text search.
• A web interface gives everyone a convenient access method.

• No metadata searching. You get filenames only, no ID3 tags in music files
• No short words. MySQL 3.x requires you to have 4 characters or more in the search. Don't try looking for just "U2" because you won't find anything.
• No booleans. MySQL 4.x supports this, but I don't know if that release is stable enough for me to upgrade.
• Finally: no guarantees! If someone has turned off Morpheus (or their computer) since the time of the scan, their files will no longer be accessible.

On the edge of your toes? Here is the system: Boogle
Happy hunting!

Despite the beta status, this is very impressive, and more importantly, cool! :) Keep up the good work and let me (and anyone else interested) know what we can do to possibly help you speed up the process to reach a fully functional inter-college search engine.

But do a search on Google for Sharescan.

As far as the actual implementation goes, why are you using a database? That's serious overkill for this application. You're better off just dumping URLs into a file and then grepping it (after appropriate shell-sequence related paranoia, of course). That'll get rid of your short word and boolean problems, as well as save you memory and CPU time. I've also got a faster script to generate the file list, if you're interested.

Faster scripts are good, although mine is pretty snappy right now. I wasn't looking for polished, just workable. And I was hesitant to do this until I saw the RPI site, which is so blatant, and seems not to be having any problems. Besides, what if you want to find a copy of Sheffries Phandemonium, where else will you turn? :-D

I used a DB because 1) I'm no good at Unix stuff but know PHP, 2) It seemed like with an expectation of around 100,000 files it would be appropriate and 3) I just had one running already. Does a nice bubble search or whatever, gives rankings, all very easily.

I'm totally not against improvements, and I figured that for Samba stuff I would end up integrating Phynd, since its all there already, and, if I'm getting my programs correct, it is the one that is opt-in, where you type in your computer name and then select shares to catalog. Or is that the other one? But that way, its voluntary. Whichever one is the voluntary one is the one I would use. Not that you aren't being voluntary in just leaving open shares on your machine...

Seperate note: I use my server to run several web sites that I would like to keep up. If people at UNet find a problem with this service, please contact me instead of just dumping my network connection. Thanks. ;-)

Philip Bond, the Commerce Department's under secretary for technological policy, said cyber-pirates steal an estimated $12 billion worth of technology and goods a year, according to the Business Software Alliance. American leadership in computers and software is "very much at stake" because of piracy, he said.

And if Brandeis offered education discounts like most other schools out there, this might help a bit also with the piracy problem...if I could get Mac OS X 10.1 for FREE like EVERY OTHER COLLEGE STUDENT, I wouldn't have had to download the fricken 650MB image from a warez site. I mention this in a public forum because OS X 10.1 was a free update, so realistically I broke no laws, but it was annoying as hell. And hey, if you actually had to pay for Photoshop, I know that I, for one, would just start using The GIMP.

Boogle 2.0 is brought to you by me, Danny Silverman, and by Peter Williams and Nat Budin, who basically did all the hard work on the SMB support. Also thanks to Tim Hickey for letting me work on this project for COSI. The system is running on my Mac OS X box. The spiders are running manually right now, since everyone is leaving anyway and it seemed dumb to have them crawling with no one around. I don't plan to update the file lists until after break. With that said, get 'em while they're fresh. And please try to stay as legal as possible. ;-)

Technical correction: Gnutella is not the same thing as Morpheus/Kazaa.

UI suggestion: Uncheck SMB if USER_AGENT is anything but MSIE for Windows.

EFF Counsel Lee Tien presented a sort of updated version of http://www.eff.org/IP/P2P/Napster/20010227_p2p_copyright_white_paper.html at USENIX. It's worth noting that the EFF *lost* important bits of both the 2600 and MusicCity cases within the last few weeks. Still I think you have a reasonable defense on fair use grounds, since it's limited to local campus. You don't qualify for the safe harbor for search engines because you lack a posted copyright policy, agent, etc. You can start on the road to that by linking to http://www.brandeis.edu/copyright.html.

Copyright holders annoyed by things found by boogle should be instructed to send a standard notice (note that the law provides a boilerplate that should be followed for a complaint to be taken seriously; Stanford has a nice outline at http://www.stanford.edu/group/itss-ccs/security/dmca.html) including the IP address, filenames, and verification of copyright ownership to copyright@brandeis.edu.

"If another University (RPI) uses it and boldly publicizes it through a website"

Um, RPI isn't doing anything. phynd is an individual student project, just like yours.

Removed gnutella reference (was trying to make it easier to understand, guess it just complicates things.)

SMB suggestion implemented.

http://www.brandeis.edu/copyright.html Linked to.

Brandeis U does qualify as a safe harbor, right?
As for the RPI, can't find that comment on this page, but I think I was just saying that its blatant, not that its university-supported.