Slashdot Mirror

A little while ago I wrote an HCI-HOWTO. I've been meaning to rewrite it some to focus on the more casual developer reader, but haven't gotten to it yet. Please feel free to leave me feedback.

Ethan

...is a title that Madonna has used for both an album and a song, which seems to make using Kazaa for anything "interesting" kind of pointless.

From the best of my understanding "police" officers are not employed by universities. This may be different here in Canada, but universities here have private security forces, which are far from the police.

Depends on the university. RPI's Public Safety, for instance, has arrest powers. The University of Michigan has had its own police force, including two canine units, since 1992.

I'm doing pretty much the same thing at RPI. I went there for my undergrad in CompSci (physically, in person), and am now doing their on-line thing for my master's (technically in Computer Engineering, but focusing my coursework on subject areas like networking, operating systems, and software engineering).

I can also say that these on-line classes are no different than if I were to take them in person. In fact, they actually are "real, in person lectures" that are merely video-taped and placed on-line. (sometimes I even recognize someone in the videostream) It actually makes things harder, not easier, since you lack that same level of classmate interaction. (and when the lecture is boring, or the professor is rambling, you pay even less attention when you're at home) Oh, and don't get me started on the understandability of a thick Indian accent over low-quality lossy audio compression.

But in general, it is an enjoyable program, and more flexible (with better quality classes) than my alternatives around here.

I'm doing pretty much the same thing at RPI. I went there for my undergrad in CompSci (physically, in person), and am now doing their on-line thing for my master's (technically in Computer Engineering, but focusing my coursework on subject areas like networking, operating systems, and software engineering).

I can also say that these on-line classes are no different than if I were to take them in person. In fact, they actually are "real, in person lectures" that are merely video-taped and placed on-line. (sometimes I even recognize someone in the videostream) It actually makes things harder, not easier, since you lack that same level of classmate interaction. (and when the lecture is boring, or the professor is rambling, you pay even less attention when you're at home) Oh, and don't get me started on the understandability of a thick Indian accent over low-quality lossy audio compression.

But in general, it is an enjoyable program, and more flexible (with better quality classes) than my alternatives around here.

I also dislike the yellowish light from low-pressure sodium lamps. As I approach an intersection on a street lined with them, I cannot see the traffic signal once it turns yellow: it disappears into the mass of other yellow lights.

These yellow lights also compromise safety, as they hinder you from seeing pedestrians in your periphery at night. The bright blue headlights that everyone complains about glare from are actually great for nighttime peripheral vision. See this article by my colleagues.

Energy is obviously a big concern for lights operated by municipalities, and you can't beat the efficacy of low-pressure sodium. But remember they use this energy off-peak, when it is cheaper. With its long lamp life, an LED solution can reduce maintenance costs enough to offset incremental energy costs, saving taxpayers money.

The other argument against white light streetlights is light pollution. This is better addressed by managing the light distribution, cutting off light that needlessly aims into the sky. Also, using a 3-LED combination to produce white light yields a white that can still be filtered. A graduate student here at the Rensselaer Lighting Research Center just designed an "umbrella" LED outdoor fixture that employs this thinking.

In theory, 3-LED streetlights could also be used for signaling purposes. For example, the city could turn all the streetlights red when there is no parking allowed. Since they don't have a 5-minute warm-up time like low-pressure sodium lamps, they could also flash on and off, in unison or in sequence, perhaps to warn about emergency vehicles or a dangerous chase in progress. You get the idea.

• Around 1984, students in the RPI-ACM created the "CB" program, which took the place of chatting (in "vamp mode") via the MTS system's *FORUM service (which was a bit like a proto-blog, I suppose).
• In 1986, a complete rewrite of the system was made, and named "CONNECT". All of the users moved from CB to CONNECT, which was quite superior.
• There was a vaporware project called "Connect-2" that was active around 1988-1990 -- this was to be an advanced object-oriented successor to CONNECT, but it was never actually written, despite the dozens of people who were actively involved with the project. (Some interesting design documents were created, however.)
• Another system, "Clover" was started in December 1989, and all the CONNECT users moved to this system after the CONNECT system was shut down (for political reasons) on June 30, 1991. Clover was the first in this line of CMC systems written for a Unix system instead of the MTS operating system that RPI used for its mainframe.
• Another CONNECT-like CMC system (which I wrote) was Gangplank, which was previously known as "Phoenix". (In its earliest days, it was just called "conf" and written in C, but this was just in the few months since development began on November 30, 1992 -- the system was soon rewritten in C++ and renamed.) I renamed the system on November 30, 2001 when I released the source code under an Open Source license. Originally, I wrote this server to talk to family members who couldn't use CONNECT. Later, I hoped it might replace Clover, but it wasn't ready enough until it was too late to interest the existing users. Gangplank is unique in that it implements the TELNET protocol directly (along with remote echo, line editing, input history, etc.) instead of using a client application. (I'm not aware of any other CMC system that provides such a user-friendly interface directly to TELNET clients...)
• In early 1994, one of the authors of Clover wrote yet another new CMC system from scratch, named lily. Again, this system was similar (from a user perspective) to CONNECT and Clover, but it was a complete platform change again. Although still running on Unix, lily is implemented in the LambdaMOO programming language. Again, the entire user base transitioned to a replacement system, moving from Clover to lily. The lily system remains in active use today. This is also an open source system, but the main server is the RPI server that the old user base migrated to.

RPI's "CMC service" might qualify as a "pre-web" service according to your definition -- although the users migrated from C

SALSA has already been done ... it's essentially a distributed system framework for Java.

The intuitiveness of Java and the expressive power of Fortran.

Maybe Dave Musser at RPI (one of the founders of the STL project) might take it over.

I heard that there was this EE who was eating at a diner. And when some dude forgot the convolution integral the EE killed the whole town. http://www.rpi.edu/~harmsj/RUPEE/

You mean like a CSA (carry select adder), which've been around for years and years?

They're pretty commonly used to break up longer-bit additions sequences, and their underlying premise is to calculate the result for out comes if the addition produces a carry and if it doesn't. Then, one of the two branches is propogated depending on earlier segments. A 2 to 1 mux with the select signal controlled by the previous segment is pretty much how the "decision" is made.

Pretty common stuff people came out with years ago, nothing innovative here.

Here's a pic of a CSA:

http://www.ecse.rpi.edu/frisc/theses/ErnestThesis- MS/Image2.gif

I'm sure you are already familiar with the work of Leik Myrabo , but in case you aren't, you ought to check out his stuff. He is the big pioneer in this area (ref. 3 on your correct link).

You are right that huge savings can be had by separating the power source from the vehicle - Myrabo was writing about this 20 years ago. Most of the energy used by a rocket is used to elevate the fuel to the altitude at which it is burned. If the energy is supplied to energize a propellant (such as vaporized water as suggested in your link), the amount of propellant can be much less than the amount of burning fuel would be. Unfortunately, Myrabo has focused his more recent efforts on rather weak air propelled engines which haven't had much punch.

Because the real limit in terms of how much propellant is needed depends upon the specific energy (and thus the temperature) to which the propellant is energized, it is best if the propellant is heated to a very high temperature. For these purposes, an concentrated ultraviolet light source would work as well as a laser, and would likely be much cheaper than a fairly efficient laser of high power. Also, if a laser is used, the best case scenario would be if the photon energy corresponded to a particular energy jump in the target propellant, similar to the way an excimer laser is used. In the case of simply heating the propellant, material considerations will limit the amount you can heat the material (i.e. at some point, your water vapor will melt the nozzle). However, in the case of the excimer laser, a solid propellant could be used which would not suffer from heat transfer from the solid/gas interface, which means really high specific energies could be used.

It says a lot that you take a comment like that seriously.

I was talking to Jesse Schell, the chairperson of the IGDA, about two weeks ago at the RPI GameFest about the work he's been doing at CMU. We were using Torque for our game development but he actually told me he was using Panda3D for their work. Not surprising when you consider that he's a former employee of Disney but apparently it has great qualities including editing, compiling, and modifying the engine while it's running (I believe it has some Python hooks or somesuch). Anyways, it hasn't been mentioned here so I thought I'd drop its name.

Yea, once sarge is finally released...

Anyways, in the meantime, here's xorg compiled for sarge:

deb http://www.acm.rpi.edu/~dilinger/xorg/ ./

We have. This has been a known problem since early 1997. It is well documented in the IRC community (admins and coders.)

Documents like this one from 1997: http://www.cs.rpi.edu/~kennyz/doc/unix/dns.spoof

I don't see how would it stagnate anything. On the contrary, we get two competing versions, where each can make its own design choices. Neither Gnome not KDE have full control of everything. If one of them doesn't want something, the other one might accept it.

And nowhere near as bad? Here on my laptop I have exactly one look: KDE. Very infrequently I use a lone Gnome app. That's very far from the mess I often see on Windows. Please try at least some of the following: Winamp, Sonique, AdAware, ZoneAlarm, any antivirus, Trillian, MSN, Norton Utilities.

Let's take a look at these programs:
Winamp
Sonique
AdAware
ZoneAlarm
Panda Antivirus
Trillian: Norton Utilities
MS Office 2003

Let's see... Winamp and Sonique are completely non-standard, and aren't even used the same way. AdAware has its own pretty widgets. ZoneAlarm has its own non-standard looking interface. Every antivirus I've seen at least paints lots of bitmaps everywhere. And even MS Office has menus that aren't found anywhere else.

Where's the consistency? Every Windows desktop I see these days is full of stuff like the above. And every Linux desktop I see is 99% KDE or 99% Gnome, with once in a while some lone app from the other environment.

Reverse execution is possible at the source level, but it requires generation of extra data structures to handle operations that don't correspond to invertible functions. This approach has been applied with some success to high performance simulations to give a "lightweight rollback", by Peters and Carothers in An Algorithm for Fully Reversible Optimistic Parallel Simulation.

"The RIAA also filed suits against Daniel Peng of Princeton University and Joseph Nievelt of Michigan Technological University. According to The Detroit Free Press, Nievelt holds over 652,000 MP3s on his own computer, a number that yields a total fine of $97.8 billion, or 120 times last year's CD sales, if the RIAA is granted the amount it requested."

FYI, Andres Salomon's patchset provides the foundation for Debian's kernels and has been discussed recently on kerneltrap here and here.

If you've had problems with CFLs burning out after less than two years, or with them not starting right away when switched on, or with them being too dim when they are first switched on, then TRY A DIFFERENT CFL.

Unlike incandescent lamps, which are pretty much all the same regardless of manufacturer, CFLs vary widely in their performance. CFLs from the major lighting manufacturers have been proven in independent studies to last at least as long as they claim under standard conditions. (We did those independent tests here at Rensselaer Polytechnic Institute's Lighting Research Center http://www.lrc.rpi.edu/.)

CFLs from the dollar store work about as well as you'd expect them to. But even different CFLs from the same known company perform differently in terms of start-up delay, warm-up time, color appearance, and whether or not base-up light output differs from base-down.

The best way to buy CFLs for your home is to pick out 3 different ones from companies you've heard of. Try them out side-by-side and observe their performance -- you will see differences, especially in the color. Then go back to the store and buy a bunch more of the one you like best. Put the other two into your porch lights.

Make sure that whichever one you choose, it is at least as small as a regular light bulb, so that it can fit anywhere. If you have any of your light circuits on dimmers, make sure you get CFLs that say they are dimmable. They even make 3-way CFLs.

I have CFLs in every light bulb socket in my home, including the one in the stove hood. Unfortunately, I still have one of those dimmable halogen torchiere floor lamps that uses a 300W halogen bulb, even though there are "fluorescent torchieres" now available that use one-fourth the electricity.

I'd like to point out that this already exists at other universities, include my personal favorite (because it's close to my heart), Rensselaer Polytechnic Institute.

I'd like to point out that this already exists at other universities, include my personal favorite (because it's close to my heart), Rensselaer Polytechnic Institute.

I'd like to point out that this already exists at other universities, include my personal favorite (because it's close to my heart), Rensselaer Polytechnic Institute.

There are bug-fixes-only patches available since 2.6.10 thanks to Andres Salomon. So if you are afraid of a 2.6.x kernel you can wait for 2.6.x.n which won't contain the most cruel bugs.

Since the submission is a little short on pictures, google was nice enough to point me at the writer's description, which is complete with pictures. I was curious how in-depth the book actually goes...

Basically, he explains how to work with base electronic equipment, basic parts (resistors, capacitors, potentiometers), and how to tie them together with infrared reflectors into an feedback system to direct steering (something my college offers a course to introduce control systems). It appears that it ends with a little writeup on some robotics clubs.

All in all, its basic mechanical/electrical engineering work, basically a bottom-up design for creating an autonomous object. Good project for starters, written in a way that most of us can whip up at home (its made out of legos after all). For those that are interested in more, look up "behavior-based robots" in your favorite search engine.

Material Safety Data Sheet

At UC Berkeley, back in 1990, you told a great story of you and your wife attending a movie premiere which used a fractal landscape effect they'd hired you to produce. (please forgive my repeating old family gossip, especially if I've misremembered the details :) As I recall, it took longer to generate than the producer's patience lasted, so they cropped it rather than wait for its last triangle to completely render. Your wife hadn't heard about the "shortcut", but when your effect came onscreen, she gave you a big pinch. After the movie ended, you asked her what was wrong, and she said, in effect, "That's not a fractal!" - apparently she could recognize even partial fractals as incomplete, therefore nonfractal.

Have you learned more about any other fractal recognition, either people or artificial (eg. software)? Identifying fractals, fractal metrics, noniterative predictions, comparisons without analysis... Have you heard about the recently published African Fractals, a scientific investigation of fractal "sensibility" in traditional African designs, both unconscious and explicit? Do you think human fractal recognition and execution can inform our computer science investigations of this geometry? Perhaps the popularization of fractals in European-rooted design might influence our modern global culture as deeply as it seems to have influenced culture in Africa?

We agree about the European view towards overseas cultures. Voltaire himself wrote convincingly of the impossibility of African civilization, despite its reality. Eurochauvinism cheats us of vast learning developed by other people, which we can steal without diminishing them (to paraphrase Jefferson, another chauvinist ;).

I'm currently reading the pioneering study of "fractal sensibility" in African craft, African Fractals. We are fortunate to be more enlightened than most Enlightenment Europeans - our world heritage entitles us to steal freely from the ideas of anyone we choose.

You'd think you could google.

NATO document affect on local climate human impact reference ref ref ref ref ref ref ref

free alumni email for life

This might not last...

in the ultimate metaphor for tech worship, the building used as the main computing center and server room is a former church.

I have to agree. The building is very nice: Take a look.

sweet deal on the standard laptop: IBM T40

The standard laptop is actually the T42 for the class of 2008 :)

Also, since a standard laptop must be purchased by incoming freshmen, it is for all intents and purposes part of your tuition.

This is not true -- one only has to buy a laptop that meets the minimum specifications. However, the standard Thinkpad is the best deal, I think.

We actually have had usenet since it's inception before the Internet.

Um. No comment...

free alumni email for life

This might not last...

in the ultimate metaphor for tech worship, the building used as the main computing center and server room is a former church.

I have to agree. The building is very nice: Take a look.

sweet deal on the standard laptop: IBM T40

The standard laptop is actually the T42 for the class of 2008 :)

Also, since a standard laptop must be purchased by incoming freshmen, it is for all intents and purposes part of your tuition.

This is not true -- one only has to buy a laptop that meets the minimum specifications. However, the standard Thinkpad is the best deal, I think.

We actually have had usenet since it's inception before the Internet.

Um. No comment...

Also, nobody requires students to run windows, in fact there is a large effort by the ACM and other groups to install linux.

Hooray for the ACM. I have a hundred times more respect for them than I do for the actual administration. Unfortunately they don't run the school. What bothers me is their willingness to extend the hegemony of Windows into the academic sphere. When they first started the laptop program, they said you can use a powerbook if you run Virtual PC on it. The current requirements don't even seem to allow that, AFAICT.

Dorms may be firewalled off, but it isn't restrictive like you claim, it just blocks incoming ssh, ftp and a few others.

When they implemented the firewall, it blocked all incoming connections. They later relaxed it so port 80 was unblocked for HTTP - all other ports and applications were still blocked. Has this changed?

And the problems with the RIAA came because RPI students wrote some amazing software for searching networks, further demonstrating the power of the RPI network.

Phynd is great and all, but I wouldn't exactly call it amazing. All it demonstrates is that the network works at all. I am not impressed on RPI's part. Do they defend their students, or just collaborate with the RIAA? That is the important test.

it's ah... not like it's freaking New Jersey, or anything.

Because Troy sure is nicer than Princeton, and damn if I don't love having it snow during commencement ceremonies in May!

Hey guys, check this out. RPI said they'd give all of their alumni free e-mail. That's pretty cool isn't it?

But, unfortunately if you go here, you'll find out that they're charging now. That's got to cause them to lose some points.

I don't like how they don't put the true reason on the front page, they hide it away in the FAQ

What does everybody else think about this with regards to the article?

Hey guys, check this out. RPI said they'd give all of their alumni free e-mail. That's pretty cool isn't it?

But, unfortunately if you go here, you'll find out that they're charging now. That's got to cause them to lose some points.

I don't like how they don't put the true reason on the front page, they hide it away in the FAQ

What does everybody else think about this with regards to the article?

Hey guys, check this out. RPI said they'd give all of their alumni free e-mail. That's pretty cool isn't it?

But, unfortunately if you go here, you'll find out that they're charging now. That's got to cause them to lose some points.

I don't like how they don't put the true reason on the front page, they hide it away in the FAQ

What does everybody else think about this with regards to the article?

RPI trivia: their main computer labs are in a stone chapel, complete with stained glass windows.

RPI trivia: their main computer labs are in a stone chapel, complete with stained glass windows.

RPI trivia: their main computer labs are in a stone chapel, complete with stained glass windows.

Its about time RPI was number one in something other than student depression and nothing to do around campus! Go Shirley!

RPI's not a huge surprise? I expected MIT at number one... not below the top 25. Same for many others. WTF?

Well, unlike MIT, RPI is much much more interested in technology and applied science than pure science - it is an "engineer factory" so to speak. Not only that, but their campus is tiny. A few wireless access points is all it takes to cover the entire place, unlike my school (Cornell University).

When I was there last, it was hard to find a place without wireless or ethernet available. Very cool. Good job RPI!

BTW, if you wonder what they do there, I was talking to Freeman Dyson the other day and he seemed to be very excited about their lightcraft - UFO looking space ships powered by earth-mounted laser generators. He seems to think they are much more likely to work than space elevators.

Cheers,
Justin

Universe is marshy but evaporates downstream. Sense the rapids reverse. Down
bridge is now marsh:
Marshy marshy marshy marshy marshy marshy marshy marshy marshy marshy now.
All evaporates downstream. Sense
the rapids now:
Rapids rapids rapids rapids rapids rapids rapids rapids sensed.
Ugh +.
Take powers from snowmelt therefore;
the current time is of youth. Fountain is young. Bear cannot
reverse. Down inverse. Lock young. Switch young. Range. Switch clone to the
switch itself. Now inverse. Lock narrows down:
Powers
to append. Up go all young. Bear time evaporates
then. Therefore:
Spawn power. Invert evaporates it. Down force. Down reverse. Down net. The
net reverses force.
Now try:
Add add add add add add add now.
It is not possible; now count:
0.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10 .
11.
12.
13.
14.
15.
16.
17.
18+.
You can now pump
in reverse. Down lock goes; narrows lock down:
Inverse. Lock young. Range. Sense 0n 1n 2n 3n 4n 5n 6n 7n 8n 9n
Powers lock time now.
Inverse. Lock young. Range. Sense 0n 1n 2n 3n 4n 5n 6n 7n 8n 9n
Powers snowmelt now.
Powers
all:
Bear hatchery n
powers
insulated bear hatchery ?.
Hydro. Power spring as
snowmelt powers snowmelt then, and disengage.
HYDRO!!

I like the Electrical Engineers parody.

More screenshots I've seen:

http://home.centurytel.net/jacob002/xorg-mplayer.j pg
http://home.centurytel.net/jacob002/metacity-compo sitor.png
http://home.centurytel.net/jacob002/skippy-xd.jpg
http://albin.abo.fi/~jfors/images/saya-20040830-1. png
http://members.arstechnica.com/x/ioslipstream/milk shot.jpg
http://home.comcast.net/~amsilveira/screenshots/08 -27-04bg.jpg
http://www.rpi.edu/~penwan/ss-20040829.png
http://home.pacbell.net/elomire/screenshot.png
http://thorin.battleaxe.net/~prototyped/kde33.png
http://members.arstechnica.com/x/treatment/Screens hot-14.jpg
http://home.centurytel.net/jacob002/xorg-glxgears. png
http://www.arslinux.com/~jorge/screenshots/xorg.pn g
http://home.centurytel.net/jacob002/xorg-transvide o.jpg

David Cross is definitely one of the lesser known hackers of the world. But he's contributed to FreeBSD and fixed bugs with NFS - he's also doing a little filesystem work in his free time.. He's really the guy that keeps everything running smoothly in the RPI Computer Science Department. So I'll just give him a tip of my hat and be on my way.

David Cross is definitely one of the lesser known hackers of the world. But he's contributed to FreeBSD and fixed bugs with NFS - he's also doing a little filesystem work in his free time.. He's really the guy that keeps everything running smoothly in the RPI Computer Science Department. So I'll just give him a tip of my hat and be on my way.

That's interested. I rarely run into tape that old, but I can think of one where their likely is some, I'll have to go check it out. I'm also a big fan of Permacel.

What do you do, theatrically? I'm in school, running the amateur theatre group at rpi.edu, and having a lot of fun with it.

--
Phil

That's the key. Lock your door and lock your laptop. 99.9% of theives are looking for something quick and easy. Even in your dorm room, even if you lock your door regularly, buy a 10 dollar laptop lock and use it all the time. Your roommate might leave the door unlocked or open for 30 seconds while he/she goes to the bathroom, but if your laptop's unlocked, 30 seconds is all it takes. I just graduated from RPI where they make all students own a laptop. Every week in the paper one or two people would have their laptop stolen. Every single time the laptop was not locked, either on a library table or in an unlocked/door wide open dorm room. Be slightly paranoid and the would-be thief will steal your neighbor's unlocked stuff instead of yours.