Slashdot Mirror

Reminds me of the MIT AI Olympics and other such events. For instance, the attempt to build a robot that, without touching the ground, can find a red circular 'puck', pick it up, and put it in another specified location. These contests have been going on for years and some of the universities have done well -- but to my knowledge, none have pulled it off yet.

It'll be interesting to see the level this science (pertaining to nanotech and AI) has got to outside the University research community ...

This whole issue of software patents and related issues with copyright is at the heart of the GNU/FSF movement. Currently, most everyone in the Open Source movement is still playing fair so that the Free Software advocates seem overly zealous. However, in the not-so-distant future we may well be glad that some people held the torch.

Or any other design that completely removes or makes optional all system abstractions that do not contribute to performance for your application space. Read this link at MIT

It's Not Linux (INL), but MIT's Exokernel would likely be appropriate for this. They post numbers that seem to be as meeningless as IBM/Mindcraft/etc's.

I don't know what the fsck you are smoking but it must be damned strong to ever coe to the conclusion that *any* release Win95 is more stable than the MacOS?

For me, the issue is not strictly one of crash rates, but also of the likely consequences of a given crash. Even if early versions of the MacOS did crash more often than Win9x (and I'm far from certain of that), the worst consequence of a Mac crash was generally a quick reboot. Win9x has a pretty good chance of devouring itself in the course of crashing, or of recovering from a crash.

MacOS seems to be stable enough to use as a web server (as witness this site), but even if it isn't, at least it's not a self-eating OS.

--

Check out the mozilla-fonts package I've put together. It's a complete set of Times, Helvetica, and Courier typefaces for Netscape, in the big, voluptuous sizes your eyes know and love ;-)

Get 'em right here!

In my book, the free software definition is stricter than the open source definition.

An example of an Open Source license that doesn't guarantee my freedom is the new, Open Source Qt license. I cannot incorporate small bits of Qt code into my own projects, nor can I do a number of other things. Yes, I am aware that RMS and a few others have finally called it free enough for them; it is still not as free as GPL. Freedom comes in levels.

To give an example where the main advocates of each clashed (if people argue that we should listen to authority instead of common sense): the Open Source includes thing like the original Apple community license before Apple fixed it. Free software advocates didn't consider it free enough.

how can you make a robot smaller than the smallest possible computer core?

Maybe you can find something better than circuits etched on silicon surfaces. Tom Knight at MIT is looking at how to get bacteria to perform useful computations, using genetic engineering methods that have become well understood. You can mail-order custom DNA sequences, graft them into cells, and get the ribosomes to synthesize the proteins you want, if you're smart enough to design proteins. Eric Drexler, generally recognized as the guy who formulated the concept of nanotech, wrote one of his early papers on the possibility of engineering proteins as a step to a more complete form of nanotechnology.

The nanotechnology literature (see Engines of Creation and Unbounding the Future) talks about placing atoms at specific locations as you build up a molecular machine incrementally, in a process called mechanosynthesis. If this works (and I'm not aware of any technically sound arguments that it wouldn't), it might become possible to build almost any object whose existence doesn't violate the laws of physics. At least, it would become possible to build a lot of different things we can't build today.

how the hell do you tell them what to do? ...

how big are these things going to have to start out, since the first generation must contain all sets of code for all generations of nanomachines?

There is a lot of excellent information about nanotechnology at Ralph Merkle's site at Xerox PARC.

While something like this can't be expected to work in realtime on every home computer, it may be able to be calculated on the developer's machine, and a memory (brain) dump of the resulting, evolved AI could be used. The pre-rendered aspect comes into play here.

The datafile of the AI that would be passed along to the home user would be unreadable, perhaps, even to the developers.

rtfm.mit.edu contains some very good FAQs from Usenet on Genetic AI.

> [How *does* one copy *.c to *.c.old in Windows?

One creates a new folder, calls it "Project Name 990909" (or whatever), copies the files to be archived, and pastes them into the new folder.

That's how you make a copy of each file in a different folder, with the same extension, the thing that you do in unix with cp *.c "Project Name 990909" But that isn't what was asked for. What was asked for was to make a copy of the files in the *same* folder, but with a *different* extention. That's an entirely different operation, with entirely different results, especially on a system like Windoze where the result of clicking on a file depends on its extension. I suspect that there is no way to accomplish this task in Windoze, other than changing the names of the copied files one by one.

This is typical of mouse-based UI's. The number of distinguishable mouse gestures is vastly less than the number of distinguishable things you can type on the command line. So a GUI can only do a tiny number of things compared to what a command line can do. But this tiny number is large compared to the even tinier number of things that can be done during a demo, so a GUI demos as being as powerful as a command line, and easier to use with zero training. Since most software gets sold on the basis of demos, this means that if you want to sell your app, it needs a GUI, even though this means that the set of things you can do is far poorer than the experienced user would like.

Andy.Latto@pobox.com

Since our scientists are unclued on Quantum Physics (not claiming I am either), we have yet to scrape the surface on Brain Science.

But, if you want to research this kind of stuff, try http://www-bcs.mit.edu. Hrm...

I would love one for reading news or watever on my back on a radioLAN in the park, but for a proper werable I wold stil prefer the Private eye, or the M1.


-- We plunge for the slipstream the realness to find

It sounds just like chaffing & winnowing, except using DNA instead of packets.

The FFTw benchmarks put a 300MHz G3 just marginally ahead of a 300MHz P-II for double precision FFT's. The P-II may double it speed for single precision, I don't know what the effect on the G3 is. My experience is that the P-II compares favourably with the R10000 and 21264 for single presision FFTs. However for other mathematical calcs, e.g. anything involving division, square roots, or frequent trig calculations which cannot be tabulated, the RISC cpu's still win hands down.

AI is a much broader term. "Intelligence", in this sense, simply means problem solving ability. Old-school AI - expert systems, minimax algorithms, theorem provers, and the like - tries to mimic abstract reasoning. New-school AI - then work of folks like Rodney Brooks - asks how biological organisms interact with their environment to get things done and tries to apply that to robots and computer systems.

Expert systems are type of old-school AI - so no foul in calling this AI.

Encrypted web proxies.
See https://lm.lcs.mit.edu.
--Scott

While communications technology improves at an amazing rate, the Bell companies are stuck in the 19th century. Their primary goals appear to be increased profits through elimination of skilled union jobs, investment in anything but their core business, and maintenance of the status quo in services and pricing.

How difficult would it be to bypass them? I'm thinking of something like the Ricochet radio modems, except at much higher data rates.

In fact, the class was 6.805. Here's the current homepage, while last year's material is online here. (A division of labor between MIT and the Law School's Berkman Center.)

This article is either a hoax or poorly researched, see MIT's response in 1990.

The course is actually 6.095 and the correct link is here. If you have time, there are many resources located at MIT. You could search their collection of webservers or go straight to the source at the lab of computer science. Good luck, but keep in mind that it's hard to find a substitute for real books.

The course is actually 6.095 and the correct link is here. If you have time, there are many resources located at MIT. You could search their collection of webservers or go straight to the source at the lab of computer science. Good luck, but keep in mind that it's hard to find a substitute for real books.

Much of the reading materials for 6.085, a college course taught jointly between MIT and Harvard Law School, are available on line here. This was a great class and the reading materials are quite thorough. Topics included privacy issues, free speech issues, and computer crime.

If you want to send off a letter to your state
representatives protesting UCITA, I've transcribed
the Microsoft Word documents detailing the current
list of NCCUSL representatives, as well as the
suggested draft letter, at http://www.pdos.lcs.mit.edu/~cananian/U CITA. There's a LaTeX-formatted letter there, too:
just download, edit, print, stamp and send.
--Scott

Just to follow up -
Sun Tzu's Art of War.
Sun Tzu lived circa 400-320BC
Buddha was ~6th century BC.
Not sure of any influences Buddha had on SunTzu.
-cpd

This is a redundant comment, but necessary because while this information is available elsewhere in this thread, its much deeper in the tree of comments/replys and this information should be located closer to the root of the tree.

---------------------
Remember the LPF
by John Allsup

The organisation (or what ever there is of it these days) that was about this type of thing, is the League for Programming Freedom.

see http://lpf.ai.mit.edu/

This is also a good (in name at least) place to look as to where to start setting up such a legal fund. What WOULD be a good idea (if the funds could be generated for it) is to sort out the legal situation in as many countries as possible, and let everybody know where they stand w.r.t the law of their land (I'm a UK citizen).
-------- John Allsup email: jda570@bham.ac.uk

Felix

According to this history of the case IPIX even insists that its own threats are copyright, and "any dissemination, distribution, retention, archiving, or copying of the communication is strictly prohibited" But there's a copy of the original email on this excellent (and scary) patent watch site at MIT.

The most interesting thing is IPIX's belief that it "owns the copyright in the format it utilises", and that therefore it has a share of the copyright of the data-file of any image in that format, which it can use to restrict how that data-file is used.

From Dersch's (IMHO) staggeringly mild and reasonable summary of the story so far, it appears that they are still trying to push this claim, which is like Microsoft claiming copyright and distribution rights over every document in Word format.

In this case we might be lucky because IPIX didn't invent the format.

But think of (say) the MPAA claiming such a veto on any file using their new music format. In fact, under the new laws against script-kiddies even describing such formats might become actionable, as abetting the theft of copyright content.

This is a nasty can of worms and it's important for all of us that Dersch sees off IPIX with no compromises.

According to this history of the case IPIX even insists that its own threats are copyright, and "any dissemination, distribution, retention, archiving, or copying of the communication is strictly prohibited" But there's a copy of the original email on this excellent (and scary) patent watch site at MIT.

The most interesting thing is IPIX's belief that it "owns the copyright in the format it utilises", and that therefore it has a share of the copyright of the data-file of any image in that format, which it can use to restrict how that data-file is used.

From Dersch's (IMHO) staggeringly mild and reasonable summary of the story so far, it appears that they are still trying to push this claim, which is like Microsoft claiming copyright and distribution rights over every document in Word format.

In this case we might be lucky because IPIX didn't invent the format.

But think of (say) the MPAA claiming such a veto on any file using their new music format. In fact, under the new laws against script-kiddies even describing such formats might become actionable, as abetting the theft of copyright content.

This is a nasty can of worms and it's important for all of us that Dersch sees off IPIX with no compromises.

The organisation (or what ever there is of it these days) that was about this type of thing, is the League for Programming Freedom.

see http://lpf.ai.mit.edu

I'll side with the last group of people. The idea that you would need "permission" from a third party to distribute software you have written yourself is absurd. As a programmer, I see that this severely limits the ability to make good implementations. Suddenly, our laws would have turned against us, limiting our freedom.

Luckily, this doesn't seem to be the case now. Although we shouldn't loose track of what we're dealing with. This is only a tentative judgement on three issues and while it's good to focus, I think the other issues might be just as important.

There's two places I'd like to ask you to go to, which doesn't relate precisely to the topic at hand, but which talks about programming freedom.

• Patent Reform Is Not Enough by RMS which talks about why it isn't enough to simply patch the current patent laws.
• League for Programming Freedom which is an organization that opposes software patents and other issues that prevents programmers to work freely.

I know that these guys filed in April 1996, but we presented "Creole" in October 1996. Although the SIGDOC documentation says "Latte," we had discovered that Borland had a product in the works named that, so we changed the name to "Creole."


Creole was an SGML/HTML parser and compiler with a built-in scripting language and virtual-machine -- the CVM (Creole Virtual Machine). Requests to any number of web servers would be passed by a webserver plug-in or CGI script to the Creole server for processing. Creole itself could also run as a CGI program.


It seems to fit the bill.


NC State University was using the first version of Latte on their website in late 1995 and 1996.


We might have prior art... but it'll be hard to prove. I'd love to prove it, of course, for the sheer pleasure of invalidating their patent and returning to the public domain the idea of distributed processing for the web.



The SICDOG proceedings abstract is here.

see sja-play.el for an elisp player. i call it "sja" for "simon jansen asciimation". this file is part of project ADHOC, under GNU GPL.

see sja-play.el for an elisp player. i call it "sja" for "simon jansen asciimation". this file is part of project ADHOC, under GNU GPL.

_The Journal of the Institute for Hacks Tomfoolery & Pranks at MIT_, Brian M. Leibowitz, 1990, published by the MIT Museum, ISBN 0-917027-03-5

_"Is This the Way to Baker House?"_, edited by Ira Haverson and Tiffany Fulton-Pearson, 1996, published by the MIT Museum, ISBN 0-917027-04-3

Both are B&W illustrated, the first is a history and the second is a collection of essays.

They aren't availible from amazon.com or bn.com, but the MIT Museum shop online has them for $20.95 each here. Ordering Instructions are here.

_The Journal of the Institute for Hacks Tomfoolery & Pranks at MIT_, Brian M. Leibowitz, 1990, published by the MIT Museum, ISBN 0-917027-03-5

_"Is This the Way to Baker House?"_, edited by Ira Haverson and Tiffany Fulton-Pearson, 1996, published by the MIT Museum, ISBN 0-917027-04-3

Both are B&W illustrated, the first is a history and the second is a collection of essays.

They aren't availible from amazon.com or bn.com, but the MIT Museum shop online has them for $20.95 each here. Ordering Instructions are here.

If you're interested in some of MIT's other hacks, go here.

MIT issued an official press release praising the hack. You can read it here.

you need to go look at the hacks page. The hack in question is here

you need to go look at the hacks page. The hack in question is here

... is that the official MIT web page has a link to this on their front page. I don't know about other schools, but Duke has a pole shoved so far up it's a** that if we pulled anything like that, not only would we be arrested, the school would do everything in their power to suppress the news. No way it would stay up for a day- much less make it to the school's front page. Argh... if only I could tolerate cold :)
~luge

There is a whole page at MIT about their hacks -- hacks.mit.edu. Very nifty. Other "modifications" to the dome have been a police cruiser at the top and the whole thing done up as a jack-o-lantern.

Of course there's dust on the NT boxes. Cambridge, MA is the neighborhood where X and Kerberos were born. And I hope Cambridge stays that way!

Has anyone seen the logo that Debian distros put on the default Apache top-level page? It's a very beautiful silhouette of a gnu and a penguin, with "Debian GNU/Linux" across the top.

In case no one has seen it, I have it here.

As good as this design is, I came across it only by chance, and have never seen it anywhere else. If it could be cannibalized (remove "Debian," maybe touch up here and there) it would make a splendid logo for the purpose at hand . . .

(P.S.: And if "GNU Inside" is too Intel-ish-- and I suspect it is-- why not use, say, "GNU System" or "The GNU System?")

Not quite as good as yuo had in mind in some respects, but allows you to ``remember'' things you never new. Check out the Remembrance Agent.

If you download MIT Kerberos 5, it includes GSSFTP which is a Kerberized FTP service. Unless you have a Kerberos infrastructure at your location, however, this may be an excessively complicated solution for you.

Something very similar has been done with some interesting results. This link http://foner.www. media.mit.edu/people/foner/Julia/section3_3.html takes you to a description of a bot named "Julia" who did a good job of impersonating a human on a Mud. It took one person 13 days to figure out that it really wasn't a human he was trying to chat up.

Why People Think Computers Can't

Just as a side note..Linux DOES run kerberos (case in point CMU and MIT) ... So it can be almost as secure (if not equally secure)as the SP.

>could they make gloves that would read sign language...?

Yes. They have. In fact this hat does it without gloves.

>could they make gloves that would read sign language...?

Yes. They have. In fact this hat does it without gloves.

Did anyone notice this?:

snip
---
What I really want is a machine that unifies my communications at a high level. That pushes us back to something that's more like a small portable or wearable PC. These appliances tend to grow functions and grow extensions over time, and eventually they end up being full-fledged computers even if they don't look like them on the outside. The appliances in the future are going to be like very small, very lightweight, and very carryable PCs that just happen to have a simple interface wrapped around them. And yes, I think that Linux will dominate them.
---
snip