[UCLA-LUG] static IP's in d

[Bill Helfman]

Note: This isn't from my email address. Either reply to linux@linux.ucla.edu
or witten@torsion.org

> > UCLA damn well better be in the "business" of providing things to make
> > students doing their homework happy. Or else, guess what.. students will
> > take their "business" elsewhere.
>
> So remind me again why placing a test server on a specific PORT requires a
> static IP?  Additionally, if you'd read the second part of the message you
> would have noticed that there is a means to temporarily get a static IP
> for classes that might actually have a reason for requiring a static IP
> (though unless its for profs/TAs to check servers.

I think you're missing the point. But it's probably my fault for not making my
point very clear. It's not the static IPs I'm worried about; it's the whole
trend of clamping down on what students can do on the dorm network. As I said
in a previous message, the old policy was that you had to have a static IP
address in order to run most network services. The new policy not only
prohibits static IP addresses, but also requires hoop-jumping and paperwork
*in order to run a Linux machine*. It's that last part that I'm concerned
about.

And yes, UCLA is perfectly entitled to make policies like this.. the same way
*any* corporation is entitled to treat their customers like so much cow
manure. Likewise, I'm perfectly entitled to advise prospective students to
consider these policies before choosing to attend UCLA.

> I don't really see much
> justification for that but *shrug*) so obviously they realize that
> students might want to conduct the "business" of their homework-- they
> just don't want students conducting the business of, say, a major FTP
> site. (They probably haven't installed load levelling, bandwidth
> throttling and the other goodies that let real ISPs support high-traffic
> sites)

(Actually, check out the policies on www.resnet.ucla.edu. They specifically
mention the possibility of such traffic shaping.)

There are a lot of legitimate things that a student can do with network
services. Not everyone with an FTP port open is going to be running a warez
site. While requiring "approval" before students can run said services isn't
exactly the worst thing the STC's ever done, it isn't very supportive of
academic learning, either. Do you think people would be as prone to play
around with new protocols, new servers, or new programs if they have to fill
out a form in triplicate and get it co-signed by two university administrators
every time they want to do so?

You know what.. this issue is *very* similar to the whole U.S. cryptographic
export rigamarole. Everytime a business wants to sell a product that uses
encryption to an overseas client, they must go through all kinds of forms and
checks and approval before they can even hear whether their product might
possibly be approved for export. THIS STIFLES INNOVATION. It holds back the
U.S.,  while foreign countries without braindead governments can innovate and
export all the encryption their little hearts desire.

So to bring this whole analogy full circle, requiring UCLA students to get
approval before they can open a TCP port on their personal computer will be
just as restrictive and stifling for education as export controls are for
cryptograppy. And guess what? The bright students will go to universities
where they don't need approval everytime they want to blow their digital nose.
And they'll learn more about Computer Science than their UCLA counterparts, as
a result.

> > With the magic of dynamic DNS services, www.linuxvalue.com can point quite
> > nicely to a dynamically-allocated IP in the dorms. So there goes their
> > alleged "reasoning" for disallowing static IP addresses. But this is to be
> > expected, as the STC has never been a real big fan of reasoning.
> 
> *sigh* The point of the STC policy (that you've obviously missed) is not
> that it is ultimately useless in the face of someone who REALLY wants to
> run a server-- they're covering their ass for when RIAA decides to sue
> UCLA 'cause some idiot was running a public MP3 server, infringing
> copyrights left and right and managed to piss of BMI (like that poor kid a
> couple of days ago). Since UCLA owns the networks they are (to some
> extent) liable for what people do with that network.

Well, it's debatable as to whether they'd really be held liable. But I think
you're right -- a lot of UCLA's motivation for the policies has to do with
ass-covering, whether they would really be held liable in court or not. But
surely a happy medium can be struck between ass-covering and allowing students
to do legitimate research and homework. Some might say that this
"fill-out-this-form-and-we'll-get-back-to-you-about-whether-you-can-open-a-TCP-port"
approach is just that happy medium. I happen to disagree.

> ---
> "There is an ineffable quality to some technology, described
> by its creators as cocinnitous, or technically sweet, or a 
> nice hack-- signs that it was made with great care by one who
> was not merely motivated but inspired. It is the difference
> between an engineer and a hacker"
> 	- Neal Stephenson, The Diamond Age
>
>  byron@mbi.ucla.edu * Byron Ellis * Chris Lee Lab * x44026

Dan Helfman