« November 2003 | Main | January 2004 »

December 06, 2003

Forthcoming Meetings

SUMMARY
=======
WHEN : 1:00 PM (Sunday)
WHERE : Woody's (Parkwood)
WHY : Discussion of all things Microsoft-y and A.O.B

Also there is no meeting this Thursday - other stuff has meant not everyone is able to make it.

Posted by rwatkins at 09:29 PM | Comments (0)

December 05, 2003

Ad-hoc links

If we pursue the idea of ad-hoc networked gaming "in the wild," some of these links might be of interest. If you come up with other links, send them to myself or Stephen (sh68), and we can add to this list, or post another one.

Again, this stuff might just be hard.

But, we might be able to play with it. Keep on diggin', and see what you come up with.

Posted by mjadud at 04:43 PM | Comments (0)

Funding opportunity

Next meeting re: possible funding: Sunday, Dec 6th, 1:00 PM, Brian Spratt room. Email Matt if you want in on some pizza.

It's easy to get blinded by money. In discussing this opportunity, making sure we're making good choices for the longevity of the CSCS group is important.

That said, we decided to spend some bandwidth (mental, physical) thinking about wireless gaming. In particular, we weren't thinking so much about things like "Hey! I could play Tetris with my friend wirelessly!", but more like "Hey! What if me and 30 people were running around in the woods with WiFi PDAs?"

Or an open field. Or a dungeon. Or wherever you run around with WiFi PDAs.

There has been some good discussion along these lines on the list, but I chatted with Ian Utting about our initial brainstorming; he basically said this would be hard. In particular, WiFi devices, by nature, expect to connect to the strongest signal, and partner with them. So doing a completely ad-hoc, wireless network where all the participants are freely communicating with each-other will (potentially) involve developing all the protocols and networking code (at a relatively low level) ourselves.

Hm.

Doesn't sound good. But, I haven't even googled on questions along these lines. I don't know how hard it would be to develop this kind of code; perhaps someone out there has an alpha toolkit for doing ad-hoc WiFi networking? I suspect they don't.

We didn't spend much time discussing how WiFi PDAs could integrate with our BotNet project, which might be something we can think about as well. There were a few other ideas as well, which might warrant some additional consideration.

I'm not gunning anything down, but I'm starting to worry about the difficulties involved in doing an ad-hoc networked game using wireless PDAs. It might be very hard---as in, low-level network hacking hard, which might not be the best thing for the CSCS group. Nor does it hit on any of my areas of comfort, which generally covers pedagogy, domain-specific programming languages, the design process, and a smattering of AI/Alife.

Hm. This hasn't provided much of a summary from the mailing list, but you can go back and read the messages again, now that you know their context.

Posted by mjadud at 02:27 PM | Comments (0)

Cobalt RAQ opportunity

"What? A rack of kobolds?" --Ed

In speaking with our sysadmins and D. Barnes, it has been made clear that we can make use of a Cobalt RAQ for a CSCS project server. We would be root on the machine.

Now, before you wet yourself, there are a few catches. First off, the machines have been heavily firewalled until now, because the department didn't trust them in their "out-of-the-box" configuration. So, they have been accessible from hadar only. If you don't know what this machine is, it doesn't completely surprise me, as it's a large, multi-processor machine mostly used for research.

So in moving the RAQ(s) out from behind hadar, we're going to be come responsible for the security of our machine. Currently, there's a short list of people who will have root on this machine:

Now, I don't think either Christian or I want to be the sysadmin on this machine (I can tell you for sure that I don't want to be the administrator!) I do suspect that there are some of you who would enjoy this, and furthermore, I suspect there are some of you who would like to learn what goes on in being a sysadmin.

So drop me an email if you'd like root, and would like to be part of the server admin team. One of the most important things we'll be responsible for is monitoring the usage and traffic in and out of the machine; in short, make sure we're neither hacked nor being used to propagate a hack. While I trust the CSCS group... I don't. That's rule one of being a good sysadmin, I think: trust noone.

Let me know if you want to play, and which category you consider yourself in: quasi-experienced, or learner. We can't all be root at once, but I'm interested in seeing that everyone who wants to have some responsibility gets a chance at screwing everything up horribly.

Posted by mjadud at 01:54 PM | Comments (0)

Backfill from December 4th

Two things came of yesterday's meeting.

  1. Cobalt RAQs
    The department has a few Cobalt RAQs. We've been offered one (or two?) to play on. There are some details to consider if we take the dept. up on this offer, however.
  2. Funding opportunity
    There was some initial hesitancy, partially for a number of (good) reasons related to the source of the funding, it's potential impact on the CSCS group, etc. Also, some of the hesitancy came from the fact that it's hard to just "BRAINSTORM! NOW!" A lot of activity is taking place on the mailing list; I'll backfill a little, and summarize some of what's been said.

I'll come back and edit this post to backfill hyperlinks to (to-be written) posts.

Posted by mjadud at 01:28 PM | Comments (0)

December 04, 2003

Today, December Fourth

SUMMARY
=======
WHEN : 5:30 PM
WHERE : Multimedia Lab
WHY : CSCS infrastructure, brainstorming meeting

The Cool Stuff in Computer Science group is meeting to discuss two things this evening. One is to try and finalize some infrastructure concerns for next term's CSCS project:

/blogs/cscs/archives/000289.html

The second item on the agenda is the possibility of the CSCS group applying for some soft money from Microsoft:

http://research.microsoft.com/collaboration/university/europe/rfp/curriculum/

This second point may be of broader interest, as it will involve thinking about what the group explores next academic year (beginning Sept 2004). However, the direction of this group is heavily influenced by the interests of the undergrads (and a few grads) who are currently invested in participating; we will not be committing to anything (financial or otherwise) that sacrifices the integrity or interests of the (currently committed) CSCS group.

Posted by mjadud at 11:29 AM | Comments (0)

December 01, 2003

Bored much?

Bored? Need something to do? Why not work your way through a course on the Philosophy of Quantum Mechanics. Or an introductory course in Programming Languages. An Introduction to Comparative Politics? Better yet, Urban Design Skills: Observing, Interpreting, and Representing the City.

A fascinating repository of information and ideas is being captured by the MIT OpenCourseware project. Poke around it sometime.

Posted by mjadud at 11:46 PM | Comments (0)

BCS Programming Competition

I don't know if competitions appeal to the CSCS crowd, but who knows. Personally, I'm not as keen on this competition as other competitions I've seen, as the BCS competition requires that you use C, C++, or Java. I personally am much more productive in other languages. Then again, I've never claimed to be a professional programmer.

The British Computer Society (BCS) Programming Competition offers a realistic platform on which information technology professionals, students, and academics can compete on equal terms and demonstrate their skills in a public arena.

The competition languages are C and C++ (Visual C/C++ version 6); Java (Standard Edition) and Visual Basic (version 6). The Eclipse Platform (version 2.1.0) will be available for the first time in 2004.

The Java version used in this year's competition will be 1.3.1 or later. Please note that Visual Basic will be available on a limited basis. We hope also to offer C# in 2004 on similarly limited basis. Please note that if demand for Visual Basic and C# is high, late registrants requesting these languages may be asked to choose another language.

Each team has free access to a personal computer and standard software, and the results will be judged against hidden test data by a panel of judges appointed by The British Computer Society.

Deadline for applications: 31 January 2004
Further information at: http://www1.bcs.org.uk/bm.asp?sectionID=932

Posted by mjadud at 10:43 AM | Comments (0)