Slashdot Mirror

At what point am I going to stop just reading about the culture and the people (else) and all the neat toys and actually become a geek myself? This sort of work depresses me in a big way - I have nothing to show for all the effort I put into learning the systems and languages. Sure, I'm only 2nd year university, and I never even thought much about writing software before last year (except in vague fantasies)!
At the same time, though, I think reading things like this extensively before even starting any of my planned projects (and I've been planning and planning!) will ultimately help me do the right thing in all the things I do.
One can hope, anyway!

*N

A Turing-machine is a very simple computer that has an infinite, linearly organize memory, has a finite number of states, and may, depending on the current state and the character at the current memory address, perform some simple actions like increase or decrease the memory pointer by one, or write a character at the current location.

There are a large number of different but equivalent specifications for Turing machines, and the so-called "Thesis of (Alonzo) Church" states that everything that can be computed at all can be computed by a Turing machine. So far, most computer scientists do agree.

Less strictly speaking, a programming language is Turing-complete if it can be used to compute all Turing-computable functions if we abstract from the limited memory in real computers. Any language that allows assignments, simple arithmetic and conditionals, and either goto or while, is Turing-complete.

There is a fairly good description at this web site.

EPGY is a way for gifted students in elementary through high school to take advanced classes when they aren't offered by local schools. They have math, physics, english, and apparently have added computer science since I took classes. They distribute lectures on CD-ROM, and all correspondence with the professor is via e-mail. They were also in the process of setting up some sort of chat room interface a year or two ago, don't know if that ever happened.

PS - I went to Case with Don, and well remember the IBM 650 mentioned in the dedication to Volume 1, along with the SOAP assembler. I don't intimidate easily now, let alone when I was a teenager, but.... Hell of a note when you're 60 years old and a highlight is that you once sat at the next keypunch to a genius.

Start at this link to his home page.

The reason I got a 3com card in the first place was because the PCMCIA pages mentioned that 3Com was one of the vendors that assisted "in the development of the Linux PCMCIA driver package by contributing hardware and/or technical documentation". They didn't know it at the time, but that won them a customer.

The Linux PCMCIA homepage is at:
http://hyper.stanford.edu/~dhi nds/pcmcia/pcmcia.html

The Crystal driver can be found here:
ftp://csb.stanford.edu/p ub/pcmcia/contrib/cs89x0_cs.tar.gz

The Linux PCMCIA homepage is at:
http://hyper.stanford.edu/~dhi nds/pcmcia/pcmcia.html

The Crystal driver can be found here:
ftp://csb.stanford.edu/p ub/pcmcia/contrib/cs89x0_cs.tar.gz

The PageRank system might work. Google uses this to rank web pages based on how frequently they are cited by other pages. If it can cope with millions of web pages it could work for slashdot. The downside is you get a global score rather than a "how much X likes Y" score.

Ok, so I might be poorly informed about the current trends in computer programming :-)

Is there anyone *really* using "The CWEB System of Structured Documentation" (http://www-cs-faculty.stanford.edu/~knuth/cweb.ht ml) to "Write programs of superior quality"?

I've never seen any programs written in CWEB (nor in FWEB or any other WEB) apart from those on Knuth's site and those that are distributed together with the CWEB distribution.

Combining Donald Knuth's MMIX, the Freedom CPU Project and Linux might transfer us into true Cyberspace.

The problem as I see it is the disconnect between the documentation and code. Both are descriptions of what the software does, one is readable by humans, the other by the target platform. Problems in both documentation and code arise when the two descriptions aren't equivalent.

One early attempt to remedy this was COBOL. This was to be a computer language that read like a natural language but could be compiled to a machine language. Little documentation should be necessary given a well-structured program source. Needless to say, COBOL didn't really succeed, the mapping between language and operation was too clumsy to be very effective or useful.

Knuth's made some good attempts in the direction of unifying human and machine code specificiations in his CWEB tool.

One day, who knows how long from now, programming may consist of having a conversation with a computer, and explaining a problem to it in natural language, while the details of implementation and correctness are mostly left to the computer. Nice documentation may be output, right along with the executable binary.

If you think that is far-fetched, you may be right. But just this morning I read a report that someone has written software capable of reading a Time magazine and answering questions about it. Very impressive, if true.

Jim

Many former geophysics grad students have been tempted away into pure computer science. A Stanford professor wrote some career advice for his students: "There are many good opportunities [in comp sci] because the computer world is always changing, and that puts young people on an even footing with older people... Are you planning to stay young forever? Math, Engineering, and Geophysics have their eternal verities: Fourier analysis, Maxwell equations, elasticity, finite differences, operators, eigenvectors, adjoints, conjugate-gradient solvers, expectation and covariance, moveout corrections, acoustic imaging, the list goes on and on. Learn these things and learn them well, because they can serve you for a lifetime."

Last I heard, John Koza (at Stanford) was in the process of building a 1,000-Pentium Beowulf-Style Cluster Computer for Genetic Programming.

Koza has been doing very cool things with GP for a long time, ever since he was at Michigan working with John Holland.

The goal of the OHS Project is to build an open source Open Hyperdocument System (OHS) - the critical missing piece of the technology for enabling dynamic, distributed collective knowledge work. OHS is designed to be:

Open- provides vendor-independent access to hyperdocuments within and across work groups,
platforms, and applications
Navigable- provides flexible, bi-directional linking to any object in any multimedia file
Customizable- allows users to select and create views of data best suited to task or preference
Dynamic- end-user functionality automatically changes over time to accommodate new multimedia
object formats and methods for viewing them
Collaborative- enables users to synchronously and asynchronously create, use, and modify
hyperdocuments and dialog about them.
Platform-neutral- user interaction components are written in Java
Interoperable- provides a unifying framework within which (future) multi-vendor applications can
interoperate
Reusable- knowledge stored in it can be used and shared across time, and across knowledge and
organizational domains
Extensible- new multimedia object types, means to interpret them, applications, views, and features
can be added easily
Standards-based- content formats are based on W3C XML standards

Actually, he started by offering $.01 for the first bug found in TeX. Then doubling it for every bug found after that... 7 bugs later the pot would be $2.56. Of course, it doesn't get interesting until the 14th bug or so.

As a standing offer, he pay $2.56 for each error in any of the books he's written. Why?... because it's a "hexadecimal dollar".

Hmmm. I guess there *is* a way to drain the M$ multi-billion dollar war chest...

In late July 1999, Genetic Programming Inc. started operating a new 1,000-node Beowulf-style parallel cluster computer consisting of 1,000 Pentium II 350 MHz processors and a host computer

In short, this is nothing new, research of this kind has been going on for a while, and all of them, I'm sure, came up with some new and interesting observations.

There are two explenations for why cnn noticed it and is presenting it as some kind of a breakthru:

a. (creationist) The Hand of Microsoft touched it.

b. (darwinian) a couple of researchers have evolved slightly better marketing skills than their colleagues

The whole debate about this being real life or not, real evolution or not etc. is kind of besides the point. If you write a program that evolves into something that plays tic-tac-toe, the result is interesting in itself, no need to make it behave like real life bacteria or whatever, anymore than there is to make a plane with flapping wings. That's why trying to find similarities with real life organisms is probably only interesting to biologists (as a plane with flapping wings would be). Quantitative mathematical formulas? Programming with equations sucks.

The tricky part of genetic programming is coming up with a programming language that would maximize the probability of meaningful mutations. If you were to assign a number to every element of a programming language, an array with 1k cells could contain an x^1000 possible programs, where x is the number of elements (plus variable identifiers, digits of a numbering system etc.).

Obviously if you did it in Visual C++, only a very small percentage of those programs would produce meaningful output, in Windows, most programs would crash the system, but then, that's a good thing.

If I was doing my taxes on Excell, I'd much rather the whole thing crash than start evolving.

ps. The best thing I learned back when I was doing some genetic programming was that sex is an unnecessary and inefficient way of creating good code :)

I ran into this...

http://sunburn.stanford.edu/~knuth/new s.html

...the other day. It's written by Don Knuth- you know, the guy who wrote "The Art Of Computer Programming" and TeX. (scroll down a page or two)

You can see his thoughts on the matter, but it boils down to- "Why not use KB for 1000 bytes, and KKB for 1024 bytes?"

It makes sense to me, plus his homepage is kindof neat. ;^)=

All in all, I think the new terms suck. I can see why the scientifical types want to have a clear term to use in order to clarify what they're speaking about, but why not then say: "ex-twenty" for two to the twenty exponent, or "ex-forty". If *you're* going to make an international standard that sounds silly, don't make *me* use it. (but maybe that's just the 'merican side of me ;^)=

--Robert (rames@utdallas.edu)

Firstly, Knuth has recently proposed a solution to this problem, where 'kilobyte' = 10 ** 3 bytes, and 'large kilobyte' = 2**10 bytes: see 'What is a kilobyte?' at his news page.

Secondly, the main reason hard disk manufacturers prefer the smaller decimal units is that it makes their disks sound bigger.

Thirdly, I believe that there is a difference in captialization for 'kilobyte' depending on which kind you mean. 1 kB = 1000 bytes, but 1 KB = 1024 bytes. (Don't forget also that the abbreviation for 'byte' is 'B'; lowercase 'b' is for 'bit'.)

There are some folks at stanford with an open source Secure Telnet and Secure FTP that they've built up around their Secure Remote Password protocol.. Give it a look http://srp.stanford.edu/srp/

That one was mentioned... "Stanford has one which is best known."

http://wearables.stanford.edu


..................................@ @

Just check bibliography in your latest publication. Most articles will be several years old. Important experiments are not done daily.
Quick communications are very important of course. That's what web was developed for. But IMO pre-print quality is not acceptable for the major trade publication. Different styles for different purposes.

This

There are many projects on synthesis of hardware from C/C++. A few years from now, you won't have to use VHDL or Verilog HDL to design hardware or program an FPGA.

The row architecture is not quite an FPGA anyway. It is more a distributed system with a bunch on processing units and their associated memory on a one chip. This makes it easier to program (worse comes to worse, you only use one processing element and one memory).

The main problem is to parallel memory accesses. This is especially tricky when you have refererences to mutiple locations in your code (e.g. because of pointers), dynamic memory allocation and recursions... There is still a lot of work to be done in that area (that's why we are talking 10/15 years from now).

Smallest webserver

Someone should run etrade's site through the demoronizer to get rid of all that MS-HTML. Yuck.

NSI Closes Top Level Domain Servers

NSI challenged over "obscene" domains

NSI Modifies "whois" agreement

Other related "alternative" DNS and related resources which I have seen mentioned here on /. or elsewhere: Not the European Union: eu.org (free domain names), The Internet Namespace Cooperative (provides alternative to mainstream root servers), The .us domain (an often overlooked alternative for those in the united states), Granite Canyon (free primary/secondary DNS). eu.org recently got very efficient and cleared a backlog of domains; Granite Canyon has had a lot of complaints about spotty service.

Suggested other readings: In whose domain, Exclusion and Coordination in Cyberspace, for the advanced user; Ask Mr. DNS and the FAQ for comp.protocols.tcp-ip.domains.

The website for Nexx Online does not show their contract. But from the FAQ, Nexx is not an ISP and only provides web hosting. There is no mention of spamming, probably because it's not easy to spam with the tools which they provide.

I think the spammer was spamming from ISPs, not from Nexx, and although spamming was not mentioned in the Nexx contract they cut off the spammer's response addresses.

So the significance of the ruling is actually that the Netiquette common sense was added to the contract. I think it is right, and wonder if "The Tragedy of the Commons" (game theory analysis, electronic commons) has been recognized by English law.

Oracle is gluing XML support onto 8io but it looks like a ploy to lock you into their tools. Fsck that.

My personal experiences with XML/XSL/XQL are that the whole wad of them currently make my life harder rather than easier. YES, I KNOW IT'S A MORE ELEGANT SOLUTION. But so was Common Lisp.

;-P

And we still have a ways to go in order to catch up. :( I wonder if anybody involved with Mozilla will try and bring in some of the Viola code so we can catch up with history?

And since we are on the topic of history and hypertext systems, have you seen this stuff on Englebart? http://unrev.stanford.edu/

There's a streaming video of his original 1968 demo, but it seems to require [yuk]Windows Media Player[/yuk].

-matt

I think we've conflated three distinct issues, and that we'd benefit from separating them:

1. The process by which information is generated and its quality is assessed - peer review, free-for-all, etc.
2. The mechanism and medium by which information is distributed - paper, www, etc.
3. The economic model by which information is distributed - for free, by subscription, per-use, etc.

Quality control becomes a problem with all (free-for-all, *, *) systems. (I disagree with (#41) that "it's about time to shake up the peer review system." Peer review is a great way to assure quality, addressing the questions raised eloquently in (#35, 54, etc.). "Non-elites" may clamor for "democratic" publishing, but Usenet illustrates its impact on quality.)

Similarly, publisher resistance may become a problem with all (*, *, free) systems. Archiving is a concern with (*, www, *). And so on.

By treating each of these three issues separately we can draw useful distinctions, e.g., there are at least two, very different Old Guards:

1. for-profit publishers (e.g. Reed Elsevier) who want to preserve (peer-reviewed, paper, subscription) because it's profitable
2. non-profit publishers (e.g. AAAS) who can accept (peer-reviewed, *, *) because they are driven by the professional demands of their members.

Robert

Stanford University has a "class A" 36.0.0.0/8, but this is (supposedly) being phased out and returned to IANA. Since we also have 171.64.0.0/18, there's no problem fitting all existing hosts back in. Are other universities that were granted "class A"s behaving similarly?

The Secure Remote Password protocol (SRP) provides a supposedly secure login session as well as an encrypted channel if you wish. The web site is well documented. Has anybody used this in the Real World ?

Some of the ideas and innovations you mentioned should rightly be credited to Douglas Englebart. They worked together, and Englebart wasn't the only one on the team, but the work at PARC came after the work done in the late 60's at the Stanford Research Institute.

(I know a web page where a supposedly knowledgeable person tells a reputable interviewer that the Gavilan was the first laptop computer. (Not even close.))

And contrary to popular opionion, the MITS machine was not the first PC. (Not even close.)

Furthermore, he left out all kinds of important milestones:

• Doug Englebart and co's work with the mouse, user interfaces, and more (1969)
• The Xerox Parc innovations, including GUI's, ethernet, laser printers, and more (mid-70's)
• Dynalogic, Kyocera, GRiD, Sharp, and more, who gave us portable computing as we know it (early 80's)

There are plenty of others, of course. Some of the names he left out -- Englebart, Metcalfe, Kay, Berkeley, Sutherland, and so on, are equally, if not more, important than the names on his list.

To find out more [plug:] check out the Vintage Computer Festival or my site.

This guy did a bad job of research resulting in another incomplete and misleading web page.

A friend of mine is working on the BABAR physics project at the SLAC facility at Stanford. He's been quite happy with Objectivity/DB. They deal with literally hundreds of terabytes of data, and have managed to store a terabyte of data into the database in a single day.

Try looking at the BABAR Computing System Home Page . I'm sure the people who work with the database would be more than happy to discuss their experiences.

Now, if you could "thumb" without any glove on, just an electronic eye strapped to your wrist, that would be cool. The image of that sort of reminds me of the "hand talking" that was described in Dune.

Still, based on the meagre amount of information I have absorbed on thumbing, I think the twiddler is better, if for no other reason that it has a built in pointer. Now if I could just convince the wife-unit that I need to $200 keyboard for my pilot...

There is no trap so deadly as the trap you set for yourself

HAL, or more formally HAL 9000, is the computer on board a spaceship in the movie "2001, A Space Odyssey". He had issues.

...Later on, I can customize whatever I want, but I don't want to have to deal with every feature at once.

This sounds like "progressive disclosure", an idea I first saw mentioned in articles about the Xerox Star. See, for example, this section from what I infer is an essay in the book Bringing Design to Software , which says:

The Open command was the basis for applying a technique of progressive disclosure--showing the user only the relevant information for a task at hand, and then providing a way to reveal more possibilities as they were needed.

Perhaps not exactly the same idea, but, if not, it still might be a somewhat related idea.

It's not too surprising in the case of VMWare,
considering its origins. It's important to distinguish between something like a word processor (which today is an engineering and design effort) and VMWare, which comes out of modern research at Stanford (
the Disco project).
Except where fed by research projects, it's fairly unlikely that the OSS community is going to engage in a lot of potentially fruitless work to develop new technologies. Most OSS tends towards the "build a better mousetrap" line, because it yields more predictable - and often more useful - results. For every project like Disco which results in something neat like VMWare, there are many which go *plop*.

In this model, publicly accessible hosts are considered "expendable", and their contents are recreated regularly through the firewall.

It doesn't address the issue of how to secure residential networks, unfortunately.

Not new, the idea has been around for a while, at least back to 1992 (see the references in this related paper on hyperbolic web visualization).

no what? You did not forget, or it is not case sensitive..
Man, you have too much free time I see, for all your bragging. (well, I have an excuse, my code is running in the background... third month of this debugging hell ;( )

BTW, my NT boxes do not crush as well. Besides that horrible memory leak in SP4, but it was fixed, thank you MS. Until users decide to go and add/change anything. Then there applications screw up system libraries all over the place - and there is no fucking way, short of a clean uninstall/reinstall to fix the mess. Then it works again tip top, indeed. Sheesh. Ease of use, mother fuckers. Yeah, once my thesis is over I can go and earn some money for what I am volunteered here to do...(NT desktop and application support for a research group) - if I can not find a respectable job of course.. People pay for this crap. It is still crap.

You asked for a database application example that NT cannot handle. Take a look at our stuff. (I am not with this directly, my friends are). If you say NT can handle that in any form - you are a bloody liar, who have no clue what he is talking about. Of course Linux is far from it as well, Solaris is used on mainframe level hardware. But Linux makes a good client.

no what? You did not forget, or it is not case sensitive..
Man, you have too much free time I see, for all your bragging. (well, I have an excuse, my code is running in the background... third month of this debugging hell ;( )

BTW, my NT boxes do not crush as well. Besides that horrible memory leak in SP4, but it was fixed, thank you MS. Until users decide to go and add/change anything. Then there applications screw up system libraries all over the place - and there is no fucking way, short of a clean uninstall/reinstall to fix the mess. Then it works again tip top, indeed. Sheesh. Ease of use, mother fuckers. Yeah, once my thesis is over I can go and earn some money for what I am volunteered here to do...(NT desktop and application support for a research group) - if I can not find a respectable job of course.. People pay for this crap. It is still crap.

You asked for a database application example that NT cannot handle. Take a look at our stuff. (I am not with this directly, my friends are). If you say NT can handle that in any form - you are a bloody liar, who have no clue what he is talking about. Of course Linux is far from it as well, Solaris is used on mainframe level hardware. But Linux makes a good client.

There was an event to mark the 30th anniversary of the invention. More info
here

Oddly, many old timers were there, but hardly any of today's major names in the tech industry. It was kinda sad. A forgotten invention that changed the world.

Actually, the first portable mp3 player I'll consider buying will be the gadget that reads from compact disks or from flash memory cards. Maybe a combo audio cd/pc cd super discman?

Too bad I'm all thumbs at hardware. At least we're getting there on the processor side. Remember the smallest webserver that was mentioned on slashdot last month?

comparing C++ and C speed under EGCS

was this

Seems you are wrong..

the best way to get mp3s is Hotline and Hotline 411 Search
there's even open source unix versions of Hotline here

Another project along those lines is at wearables.stanford.edu.