« September 2004 | Main | November 2004 »

October 29, 2004

The Art of Lego

Something I happened to stumble across on Wednesday. Written by one of the guys at MIT who was creating (I think!) the thing-that-turned-into the Mindstorm, it contains some good tips for creating Lego robots, as well as some good tips you can use to make your robot sturdier. I know we need some help doing that ;)

Anyway - The Art of Lego is available as a PDF. Next time your robot hits a table leg you'll be happy you read it.

Posted by rwatkins at 02:53 PM | Comments (0)

occam/Mindstorms blogs

Found on transterpreter.org:

We asked the students taking part in the Cool Stuff in Computer Science project if they would be willing to blog their experiences using occam on the LEGO Mindstorms. So far, I (Matt, slinker wrangler) have been writing about CSCS as it progresses this term on my own weblog. Another Matt (we have around 700 of them in CSCS this year) has started keeping track of things over at blog.qualityerrors.net. His first post is a back-fill for the last few weeks.

As we see more blogs on this topic, we'll mention them here, and try and get a coherent page of links together (or perhaps a blogroll somewhere on the margins of this page).

Thanks, and welcome, Matt!

(How else am I supposed to cross-announce this, anyway? I'm quoting my own blog post on another blog... something's wrong here.)

Posted by mjadud at 02:21 PM | Comments (0)

October 28, 2004

Redcone Robot News

For those of you who are enjoying the work we're doing with the Mindstorms, you may be interested in tracking Redcone Robot News, a weblog that seems to track all things robotic, from hobbyist robotics to research and industrial development.

Posted by mjadud at 05:31 AM | Comments (0)

October 23, 2004

Talk: 28th Oct 2004, 19:00 in COLT2...

.....well I think it's COLT2 anyway.

Nick Ryan will be talking about some cool ubiquitous computing research he's involved in, including (apparantly) "sharing with us some cool toys". That's enough for me to turn up anyway ;-).

Meet at 7, the talk should last an undeterminate amount of time and then we'll go off for a few drinks afterwards.

If you've not come to CSCS before, then pop along and see what we (and the department) have to offer you! The talks are a great idea to know what other people do in the department as well as getting to meet your (future) lecturers. New faces are welcome and although our initiation ceremony leaves the rugby clubs' one looking boring, it's worth the effort.

See you there,

Rob

Posted by rwatkins at 02:36 AM | Comments (0)

October 15, 2004

Player/Stage

Now, it's true that most of you won't be particularly excited about this... unless you are particularly excited about this.

Player/Stage is a set of open source tools for developing robotics applications. There's really no reason we couldn't make use of this kind of technology in our own robotic explorations.

Of course, assuming that Christian, Damian, and I have infinite time to implement things...

Posted by mjadud at 03:19 PM | Comments (0)

October 14, 2004

The CSCS Moodle site

Moodle is an educational content management system. This means "it stores stuff and whatnot for people teaching things and stuff."

We've decided to give it a go for managing resources and discussion for CSCS this term. Who knows.

Perhaps it'll make it's way to the navbar of this page; for now, it's here.

Posted by mjadud at 05:59 PM | Comments (0)

October 04, 2004

What's going on this term?

What's going on this term in CSCS? The plan is to have a rotating, 3-week schedule; each cycle will include

  1. A mini-lecture and play time,
  2. A whole evening of play time, and
  3. A talk.

On the first week of a cycle, we'll introduce a challenge or two for you to undertake with the LEGO Mindstorms, and some programming tips that will help you work your way through the challenge. The second week is pure play time, and that will give you time to explore the ideas more fully without us talking at you. On the third week, we'll have faculty from the Computing Laboratory coming in to talk about their research---something which you, as undergrads, typically don't get a chance to hear. Our lineup for this term includes:

  1. Nick Ryan talking about ubicomp and wireless gadgets,
  2. Ian Marshall speaking on biologically-inspired robotic control, and
  3. Peter Rogers talking about the whizzy work he does with graph drawing.

If these are anything like the talks we had last year, they should be enjoyable and thought-provoking.

The Mindstorms work should, if all goes well, end with an open competition that we'll do some media spinwork on, and that should be a nice way to cap off some fun explorations in playing with these little LEGO robots.

Our goal is to build up to a point where we can tackle the larger robots we've purchased this summer for use in CSCS. These larger bots have ultrasonic rangefinders, a video camera, compass, touch sensors, WiFi, and a lot of computational power (32MB of RAM, 64MB of flash, 128MB of SD for storage, and a 200MHz processor) for robots their size. These bots all need to be built, so that will be a project we expect (some/all) CSCS will get involved in as we roll into the next term.

We've got space and kit for a few more bodies; this year is going to be an experiment for all involved, but we learned quite a bit after last year---we (Brad, Christian, myself) are excited to see what comes of it, and hope that you enjoy it as well. By all means, tell us if you're not!

So, for those of you who were there last Tuesday, keep your eyes out for anyone else who wants to be on the cutting edge and help us explore the use of some fun toys here in the department.

Posted by mjadud at 05:01 PM | Comments (0)

October 03, 2004

Software Patents

Kodak, through purchase of another company's "IP portfolio," has acquired some really bogus patents, and used them to sue Sun Microsystems over the very existence of the Java programming language. I don't understand how patents 5,206,951, 5,421,012, and 5,226,161 could have ever been issued. For example, the second of these three is titled "Multitasking computer system for integrating the operation of different application programs which manipulate data objects of different types"; it was filed in May of 1993, and cites earlier patents from IBM as related work; one of these, "Method and apparatus for dynamic invocation of utilities", seems to describe remote procedure calls.

Are RPCs patentable? Are they an invention? That begs whether the procedure call was an invention, which was modeled largely on the mathematical concept of a function.

Over at Groklaw, we have comments to the effect of

[Europe: ] Think about it carefully, because this is exactly what happens when you adopt a system that rewards the Kodaks of the world for such behavior and penalizes Sun for years and years of expense and sweat and toil and creativity by robbing them of their due reward, not to mention removing any motive to ever do such innovative things again as long as they live.

without mention of the fact that IBM could have, under the same ass-backwards logic as Kodak has applied in leveraging their patents, sued Sun out of existence before Sun Microsystems even left Stanford.

Not that this is a good reason for not going to court, but if IBM, Xerox, and a number of other companies were to really start leveraging their "portfolios," we'd have the legal equivalent of World War III. Apparently, these massive portfolios are no longer a deterrent.

Software patents: good, or bad? Is software patentable? Where should Europe go with this? As CS students, this seems like a rather pertinent issue... depending on whether you intend to stay in this field when you graduate, of course.

Posted by mjadud at 09:50 PM | Comments (1)