Slashdot Mirror

Expensive intially can save untold cost later. I bought my Sony DDS3 drive back when they were twice what they are now, and it's been worth every penny. Come to think of it, that drive is the single most expensive computer component I've ever bought; it was money well spent. I've had to restore from tapes as much as 6 months old: no errors, no problems. The QIC80 drives I used to use were not that reliable. Travan etc. is pretty much QIC80 on steriods.

The only problem I have with it is that it's internal compression is about equivalent to gzip -1; I like better. A reasonably fast machine, the alternative compression program of your choice, and the buffer program can solve that problem.

Also, don't buy the "new" Adaptec 2940UWPro (the one that can drive all the connectors at once) SCSI controller to hook your new drive to, it doesn't speak to Sony tapes, and Adaptec's response is: "We know. Too bad." Excretus Est Ex Altitudine

I love Linux, but EPOC32 is already a very solid, responsive OS for a palmtop. I don't see the need for my Psion to run Linux.

here are a two definitions of what an operating system actually is:

whatis.com's definition of an operating system

foldoc's definition of an operating system

here are a two definitions of what an operating system actually is:

whatis.com's definition of an operating system

foldoc's definition of an operating system

Here are the first two chapters...

Chapters 1 & 2 can be read here.

Indeed the Sholes (qwerty) layout was not designed to slow down the typist. Instead, it was a clever way of arranging the keys (and corresponding hammers) to make sure the hammers on the typing machines would no longer jam because of being too close together. This ultimately allowed the typists to type faster.

Or even better, try this page for more info on the Dvorak layout.

Go to FOLDOC, and look up the term ``race condition''.

PS: I did an April Fools joke on Congress back in the early 90s involving a fake press release from "UIP" that droned on and on in technocrat-babble about "the national transportation vehicle initiative" whereby the Feds would build this enormous fusion powered truck for "national competitiveness". In it, I portrayed Craig Fields as having nothing but glowing praise for "the public private partnership" from his new position as President of gallium arsenide technology leader, Cray Computer Corporation where he had replaced Seymour Cray who, my fictitious story went, "died in a jeeping accident in the Rocky Mountains". This "joke" was sent to every congressional office years before Seymour Cray died in a jeeping accident in the Rocky Mountains. "Funny" how Cray died shortly after he violated his own historic avoidance of direct architectural service to the spookshops.

It's ok to run Microsoft out of buisiness, but not O'reilly. WHY?

Anyone who thinks the Open Source movement is going to run Our Favourite Software House out of business this side of Armageddon (oh, whoops, that was last month wasn't it?) is deluded. It has a part to play in software development, and is big enough to look after itself in the unlikely event of demand for Windows, Office and other software disappearing overnight - by which time the company would have sensed the change in fortunes and moved on to something else anyway.

And for all people may hate the company, this is a Good Thing, for the simple reason of competition. OFSH cannot use any corporate tactics to buy out or crush the Open Source community, because there is nothing tangible to attack. And marketing/PR tactics can only do so much - and isn't going to do _all_ that well in the face of companies who know they've been running servers on free software quite happily for years. So OFSH is forced to compete with an ephemeral, undefeatable rival.

So, OFSH will bust its guts to make sure that the next Windows release can outperform Linux. The Open Source developers will respond in kind. And so on, ad infinitum. Net result: accelerated growth and innovation. Not only does the Open Source community come out with a better operating system to keep up with what the latest non-free OS has to offer, but OFSH will *shock, horror!* come out with an OS which doesn't suck.

In the end, everybody wins. We get a better OS, they get a better OS - and ours is still free.

Linus and Richard Stallman (and ...) don't pay rent or buy food, etc.? Writing software isn't that much different than writing information in terms of workload: You sit in front of the computer and type.

Linus has a day job, I believe. I can't remember what RMS does. Talks at people, probably. Either way, they're now noted enough not to starve.

INFORMATION IS FREE! As in FREE BEER!

Information is power, and power always comes at a price. And beer's one-eighty a pint around here. ;)

[Sorry.]

I write articles and put them on the Internet for free because I believe that they might somehow help somebody, and because writing about a subject helps me understand it better.
I can't thank the LDP people enough for what they have given the world.

Good for you! However, I think there are a few points to note here.

Firstly it is up to the author how to make his work available. Just as when writing code you may licence it under any terms you wish, unless it is a derivation of someone else's work.

Secondly, I think the work involved in documentation is
(a) considerably less collaborative in terms of shared volume of input (there aren't all that many books with more than two or three authors);
(b) considerably less fun.
Thus, there has to be another incentive, and while altruism is all very well, it's in short supply.

Thirdly, as a customer I have the right to purchase non-free documentation if I so choose. This does not make the producer of said documentation evil. They are merely a company supplying a product in demand. There is, as you noted, a free alternative. Man pages, HOWTOs, FAQs, page upon page of HTML documentation scattered around a thousand websites - It's your call. Personally, I prefer to part with my forty quid and do it the easy way. =)

I don't care about the book, but we are arguing about the information.
You know.. the http and ftp protocol.

Erm, have you heard of RFCs?

The authors can easily put their documents online, and tell the publishers to print and distribute it!

True, but that's up to the author. They may choose not to, because (as previously stated) the basic information contained in the books is already available. What you're paying for is the author to make it easier for you to understand. Same way as you pay for a training course, in fact, only one Hell of a lot cheaper. :)

--
This isn't the post you're looking for. Move along.

The case is air-tight:

JavaScript is the fastest growing programming language (see table below).

The product of JavaScript's market share times its growth rate is the highest of any language. (see table below).

JavaScript source code is inherently "open" for "Everyone" browsing the web. "Everyone" can learn by example at will.

JavaScript is the most widely deployed interpreter. "Everyone" has a web browser.

JavaScript's dynamic features, including functions as first class objects with slots, put it in the same level of power as systems like CLOS. CLOS is, arguably, at least as powerful and general as Python. The fact that an appropriate, CLOS-like framework has not yet been widely deployed for JavaScript merely indicates developers are too distracted by immediate profit opportunities to think straight about JavaScript. This is a temporary stupor.

Back sometime in June, Randall J. Burns took Ted Shieh's figures, later reported by Larry Wall and computed growth rate in demand. We get an interesting picture:

C++.....Java......VB....Cobol....Perl...JavaScript ...Smalltalk...Python
-14%....+17%......-4%...-43%.....+18....+51%...... ...+1%.........+23%

When multiplied by the approximate market share:
C++.....Java......VB....Cobol....Perl...JavaScript ...Smalltalk...Python
40%.....25%.......10%...3 1/2%...+10....+10%.........1%..........1/2%

We get an interesting future dominance metric:
C++.....Java......VB....Cobol....Perl...JavaScript ...Smalltalk...Python
-560....340.......-40...-111.....180....510....... ...1...........12

I think it's carried on all Sunsite mirrors. In the UK it's located here. The UK magazine PC Plus has also previously distributed it on it's cover CD. PC Plus has been running a string of linux hands-on articles since the middle of last year mainly based on the SuSE distribution.

There's an entry in the FOLDOC for "slashdot effect"--see http://foldoc.doc.ic.ac.uk.

B-tree - as in Balanced Tree (although some people also say B-tree for Binary Tree).

B-tree - as in Balanced Tree (although some people also say B-tree for Binary Tree).

the brain's storage of "knowledge" is not a digital structure, so it cannot be measured in bits.

It most certainly is digital. What did you think neurons were? They store 'bits' of data. As well as working as gates and flow controls. The exact workings may not be exactly bit-wise (ie. it may have a few more settings than either on or off), but the general method is certainly broken down on a level that can be described in a digital manner.

The actual amount of data that would be found within the human mind is still quite controversial. The exact number isn't likely to be nailed down until we can mimic the human memory system on a computer, either via Artificial Neural Network or in an advanced form of biological based storage media. If we could mimic the human brain, we'd not only know the answer to this question, but we'd be able to lick the AI problem, too. Also noted... Better organized thoughts than mine. This has a few previous discussions of how large the brain's storage capacity might be.
~Anguirel (lit. Living Star-Iron)
"Veni; Vidi; Vi C++"

According to IC's web site, they have one Fujitsu VX/1 vector supercomputer and one Fujuitsu AP3000 massively parallel server, which is effectively a nest of 48 independent UltraSPARC systems linked together Beowulf style.

But when you look up Fujitsu on the Top-500 database, it turns out that only the vector supercomputer (VPP) series make the list, and none of their AP-xxxx systems.

For the VPP series, the entry-level for the top 500 is a twelve processor system rather than IC's single vector processor.

On the other hand, the AP-3000 would have enough total throughput at 45.6 Gflops to get on the list at number 172. But my guess is that it can only achieve that for problems that split into relatively big independent chunks.

That might be OK say for servers and big CFD models, but I suspect that the LINPACK test suite needs a much more fine-grained parallelisation, and would be much harder hit by communication latencies between nodes.

That's just a guess: perhaps any real supercomputer experts out there could say whether this sounds right ?

Dunno. But here's what I could come up with. FOLDOC says NB was the original name for C ("not B"?). The Jargon File (TNHD) doesn't have an entry. Neither does Que's dictionary. Perhaps it's something as simple as "notation block"? Or perhaps it a Latin abbreviation.
Christopher A. Bohn

STAR OFFICE 5.1 Download Sites:
FTP SITES:
Sun SITE UTK at University of Tennessee - Knoxville
TU Clausthal - Germany
Sun SITE Central Europe at RWTH-Aachen - Germany
Sun SITE Finland at the Tampere University of Technology
Sun SITE Switzerland - cnlab & SWITCH - Rapperswil & Zurich
Star Division - Germany
Star Division - Germany
AARNet Mirror Project - Brisbane, Queensland, Australia

HTTP SITES:
Sun SITE USA at University of North Carolina - UNC Chapel Hill
Sun SITE UTK at University of Tennessee - Knoxville
Sun SITE Central Europe at RWTH-Aachen - Germany
Sun SITE Finland at the Tampere University of Technology
Sun SITE Nordic at Kungliga Tekniska Högskolan - Stockholm
Sun SITE Belgium
Sun SITE Northern Europe at Imperial College - London
Sun SITE Switzerland - cnlab & SWITCH - Rapperswil & Zurich

Um, I recommend you ACTUALLY read some RFCs. Such as rfc 959, File Transfer Protocol.

Here's a link to it.

Nowhere in the text do the letters UDP even occur in order.

Furthermore, under section 3.3: Data connection management, we find the following:

Reuse of the Data Connection: When using the stream mode of data transfer the end of the file must be indicated by closing the connection. This causes a problem if multiple files are to be transfered in the session, due to need for TCP to hold the connection record for a time out period to guarantee the reliable communication. Thus the connection can not be reopened at once.

(Emphasis added)

Please make sure you know what you're talking about next time before you tell others they don't.

Um... while gas ppc is possible, it's not incredibly pleasant.

On the amiga, at least, there is a PPC assembler included in the StormC development environment for Haage&Partner's WarpOS system. It is a Macro Assembler, with syntax and functionality simlar to Amiga 68k Macro assembler ( i.e. C-like structures)

Note that Motorola-syntax assembler for PPC defines a large number of macros that are very useful, and I find PPC assembler programming far less trying than the icky, tacked-together x86.

NOT JUST FOR AMIGAS!!!
The VBCC package includes PASM, a motorola-syntax ppc macro assembler, and is open source and written to be portable, in theory. (although vbcc is unlikely to supplant gcc in the near future).
It can produce elf ppc binaries.

PASM:

http://devnull.owl.de/~frank/index_e.html

haage-Partner

www.haage-partner.com


go to amiga pages of
www.uk.aminet.org
or

www.cucug.org/amiga.html
and search vbcc and PPC too, if you're curious.

On Yahoo, I found a mirror locator in addition to mirrors at Ohio State, Switzerland, Internet FAQ Consortium, Japan, California, and the United Kingdom. Also noted in the discussion is a Slashdot reader mirror which could get Slashdotted.

FOLDOC says:

[...] The term comes from Swift's "Gulliver's Travels" via the famous paper "On Holy Wars and a Plea for Peace" by Danny Cohen, USC/ISI IEN 137, 1980-04-01. [...]

How about a form to perform a lookup on FOLDOC (the Free OnLine Dictionary of Computing)

[...]
grep: Get Regular Expression Pattern
[...]

Um, no. FOLDOC says that it comes from the ed (Ed is the Standard Editor!) command `"g/re/p", where re stands for a regular expression, to Globally search for the Regular Expression and Print the lines containing matches to it'. Some Unix commands are poorly named -- that's why most shells have support for aliasing./p>
--
W.A.S.T.E.

Hmmm...Centralise security server... I wonder what happen if it crash or being hacked.....

BTW, if you don't notice, UNIX have centralise security server, using Kerbose+YP/NIS. We here in Imperial College DoC we use this combo and everything is as safe as ever, even the Windows NT WKS we use which runs off SAMBA (one NT server here just to please NT WKS's profiles roaming/auto installation).

I wonder how long does it take for M$ to fix their OOB/Denial of service/etc attack and how long does it take for Linux.....

Enough said