Slashdot Mirror

I know this sounds like flame bait, but I'm sick of the FreeBSD advocates claiming that they are faster than Linux when it comes to network servers. If that's true, where are the numbers to prove it? I want to see SpecWeb results!

I don't have specweb results. I don't care about specweb. Further, I do not agree with you that it is an adequate test of network serving performance.

Here is a real-life, well set-up and functioning test - posted on slashdot weeks ago. Please note that the "primitave" FreeBSD 4.2 SMP _still_ outperforms linux 2.4 - and it's going to be disgusting about 3 months after 5.0-RELEASE, when the thing matures.

benchmark here

If you don't believe the results, set up a test like that on your own hardware, and experiance the fact that linux is playing "follow the leader" first hand.

http://www.minidisc.org/part_Sony_MDH-10.html

I found this comment at Byte's web site from one of their stories about CeBit.

"Yes, and there was quite a large Bluetooth piconet work (at least 50 devices, 10 base stations) in the American pavilion in Building One, and it all Just Worked. No muss, no fuss, just operational equipment, interoperating when wanted, staying out of the way of other systems when not."

Very interesting considering Bluetooth was supposed to be a total flop!!!

They recently changed their terms of service mandating that Juno can use its customers' computers for that very purpose. They can terminate your service if you don't leave your computer on 24x7 so that it can do the processing and dial in to Juno at whim for more data. The fun details are at www.byte.com/column/BYT20010222S0004.

Naturally they portray this as a benevolent thing and a chance to be part of their "Virtual Supercomputing Project," which claims to be completely voluntary, despite the fact that their Terms of Service directly contradict this:

2.5. You expressly permit and authorize Juno to (i) download to your computer one or more pieces of software (the "Computational Software") designed to perform computations, which may be unrelated to the operation of the Service, on behalf of Juno (or on behalf of such third parties as may be authorized by Juno, subject to the Privacy Statement), (ii) run the Computational Software on your computer to perform and store the results of such computations, and (iii) upload such results to Juno's central computers during a subsequent connection, whether initiated by you in the course of using the Service or by the Computational Software as further described below ... you agree not to take any action to disable or interfere with the operation of ... any component of the Computational Software.

[snip]

You acknowledge that your compliance with the requirements of this Section 2.5 may be considered by Juno to be an inseparable part of the Service, and that any interference with the operation of the Computational Software (including, but not limited to, any failure to leave your computer turned on at all times) may result in termination or limitation of your use of the Service.

Happy computing! :)

Cheers,

Apoptosis is a biological term for "cellular death". A few years ago Byte magazine published this article about a product called Apoptosis that unleashes "cellular death" on cellular phones. It supposedly disconnects cellular phone calls within a short distance. There were certainly times when people talking loudly in public places on their cellphone made me wish I had one of those.

I suspect this product was a hoad, but back in the days of analog cellphones it was actually very easy to build a device that does this: the frequency spacing between the transmit and receive frequency was exactly 40MHz. You could build a cheap device that receives the strongest signal around, mixes it with a 40MHz oscillator and transmits the result back. Multipying two sines results in the sum and difference of their frequencies - one of them will jam the receiver side of the phone very efficiently.


-

First, two references:

Scott Muller's Upgrading and Repairing PCs Its first chapter is 'Personal Computer Background' which i find quite interesting. I think it provides good background to non-technical people in your audience.

Byte Magazine : My favourite geek magazine when i was younger... They have this 20th aniversary edition that talks about "modern" history of PC era.

You could safely ignore what i am going to say...

A. You forgot to mention Alan Turing! How dare you... well i am sure you know about him, just forget to mention it on your post. But now you witness first hand some slashdot steam...

B. What's your audience? As one slashdotter already said, different audiences would require different kinds of materials and presention.

C. What would be different if computer literacy programs are being taught to average North American Highschoolers?

D. If i were one of your students, i would be interested in:

History of Computer Cultures (IBM, PDP, LISP, AI, MIT Vs Stanford Vs Oxford, Atari, Commodore, Mac vs MS,...)

History of impact on society (WWII, business -- COBOL, scientists, PC in offices, Internet...)

History of computation (Hilbert, Russell, Godel, Church, Turing...)

History of Computer Hareware (mainframe,minicomputer,PCs,networked computer)

History of Electronics (Vacuum,transitters,ICs,microprocessor...) (Is it highschool stuff?)

And, impact/restrains of Von Neumann machines -- sequential vs parallel, program ?= data...

All in 16 weeks!

Good luck...
Ricky

(2) Even many of those individuals with talent have insufficient knowledge of (and/or, apparently, desire to learn about) the art and science of software engineering and so persist in making the same stupid mistakes that have been well-documented for 30+ years.

(3) As a result, anyone who has had to recruit software developers can tell you how much muck you have to sift through to find the gems.

(4) I can't speak for the relevance of most CS departments; I know that my undergraduate CS program (BSCS, BYU, 1978) helped me tremendously when I went out into the real world. But that may have been an anomoly; I had some brilliant teachers with real-world experience (one had worked at Bell Labs; another went on to co-author and co-found Word Perfect).

(5) After some years in the workforce, many of those with talent and skills find they can double or triple their salary by becoming a consultant. This leads to a talent-flight from organizations.

In short, you're trying to find someone with talent, training, inclination to your topics and circumstances, and a lack of awareness of how much s/he could be making elsewhere. :-)

Best of luck. ..bruce..

Perl Is Bigger, But Python Is Growing Faster.

Python Is More Deeply Object-Oriented.

Perl Is more Powerful And More Mature In Some Ways.

Perl Lacks A Killer App, Zope Is Python's Killer App.

Python Is Designed To Be A Good First Language For A Beginning Programmer, Whereas Perl Is Most Useful To Programmers Familiar With C, Sed Or Awk, And UNIX Command Idioms.

His complete write up is here. (warning the article is a bit long... 4 pages)

And finally Python Humor

This news is reaaly five months old, this just confirms what all the Tech press was saying back in October when MS invested $135Mus in Corel.

BYTE:Analyzing Microsoft's Corel Investment Strange Bedfellows: Curiouser And Curiouser

ZDNet: Microsoft .NET for Linux?

WIRED: Corel, Microsoft form alliance

Data Drive

MD Still Camera

Video Camera

My response to byte.com's article:

In your January 22 article about the DataPlay storage device, the author writes: "How long this claim, and its copyright-protection features, survive contact with the anti-intellectual-property-rights types remains to be seen". I believe this is misunderstanding the philosophy behind the opposition to SDMI. The big media corporations have consistently and repeatedly abused the rights of both the artists and their consumers, both by lobbying for new laws such as the DMCA and the "Sony Bono" Copyright Extension Act, and by twisting existing copyright law and ignoring international copyright treaties with such abuses as region coding of DVDs (which has been carried over to DVD Audio, making a mockery of their reasons for using it on DVD movies). Fair Use and the first sale principle are being eroded or bypassed entirely, with the introduction of the "You're buying a licence, not a copy" model, which, if effective, will remove the need for the recording companies to respect the consumer side of copyright law.

Interesting point. I'm not sure whether I agree about readability or not, but just wanted to add a bit of trivia: in the empire's premiere Word processor, there's actually an option to get a white-on-blue color scheme. That was put there, way back when, at the request of a certain Jerry Pournelle (of Byte fame, among lots of other things). If I remember correctly, Jerry required this to switch, because that was the color scheme of the word processor he was used to. Can't remember the name of that (probably ancient) program, though... OK, that's all from the meaningless-trivia dept, go back to whatever. ;^)

I can mention at least 2 other failures.

• MSX - MS designed an 8 bit home computer that was not very popular
• AT Work - MS tried to add "intelligence" to Fax, Printers and Copiers by putting an OS in them

"Or how about the hundred times I've asked where I could start from to be able to tweak the virtual memory management of Linux because I have certain needs to take care of. No one can tell me."

Well, I'll tell you. Or rather, I'll give you this link. It's a good start.

For more information about BeOS (including Screenshots) click here.

A couple of quick links from a google search:
A Byte Magazine article from 1996.
An article that was published in Scientific American.
The latter link says that storing "trillions of bytes" in the size of a sugar cube seems possible.
It is a "3d type storage"- with layers of data that can be read by varying the angle at which a laser is used to read it. (See the first article)

I am sure we have all experienced the horror stories of pointy haired managers in the real world. Maybe one of these days, Stephen King will even do a story on it. Those needing examples can inspect this site, and also check out this column. Although there are many other examples easy to find around the net.

Sadly, the thing that worries me is that it takes more than a haircut cut to change a pointy haired manager.

The art of managing your managers is an arcane art indeed.

here
here
here

"Everything that can be invented has been invented."

"Everything that can be invented has been invented."

Everything an Internet console does can be duplicated by more feature-rich systems

Can you carry your PC into the bathroom with you, and browse websites? Carry it into bed or out on the porch to read an e-book (and if you don't have the book, buy it online and begin reading it). Interact with television programs, live sporting events with cool stats applets from ESPN, all from a reclined position in your easy chair? Can you carry your PC to the dining room table and read NYTimes.com while you eat breakfast in the morning?

And this is just the start.

Look, I can play DVD's on my computer, but it's more satisfying to use a DVD player, and even my grandma can do it.

I can record video on my computer from my TV, and play it back. But it's easier to use a VCR.

I can send digital audio to my stereo system from my computer, and have it play over the speakers, but it's easier to use a CD player (and would be even easier to use an entertainment appliance, like Be's Aura platform).

The long history of failures that "Internet appliances" have met indicates that there's not much interest in these kind of products.

What internet appliances? All I've seen are wired, network computers. Those have a small niche, but bring on the broadband, wireless connected devices. The REAL internet appliances!

Here they come...

-thomas

Actually, we have seen it. It's called "AMD". They take the CISC instructions and translate them into RISC instructions that can be more highly optimized, especially in parallel. Read about it at this Byte article on the AMD K6 (nee Nexgen Nx586). For those who need to read about the latest and greatest, try this Althon architecture overview (about a third of the way down). Without RISC, AMD would never have been able to efficiently make use of all those "extra" logic units.

Sure, we'll probably never get to see how a mature RISC chip will perform, but even "CISC" chips are getting more RISCy. And maybe Compaq will really put some oomph behind the Alpha one of these days.

Walt

http://www.byte.com/column/BYT20000925S0005

Jon Udell's December 18th column addresses the issue you raised, questioning whether it is in Sun's best interest to keep the language and the JVM tightly coupled. The column includes a survey of some of the people who have implemented languages for the JVM.

--
Brian Jepson

Some of the things we found about Soviet technology were astonishing: For example, in 1990, most US computer chip leads were spaced 1/10 of an inch apart. The Soviet Ministry in charge of cloning western chips had mandated metric spacing, but one-tenth of an inch works out to be about 2.54 mm.; an odd metric size.

The Soviet solution? A "metric inch" with 2.5 mm spacing. This means that Soviet clone chips could be an exact electrical and functional equivalent of their Western counterparts, and even look exactly the same--- until you tried to plug them into a western socket. The Soviet chips would almost fit--- but not quite.

It's a real issue, because Microsoft uses MFC to control developers. They're scared of a good, portable, widely used GUI toolkit. That's why it's important to have one.

There's antitrust history on this, related to Borland, which had the first C++ wrapper for Windows. Symantec decided years ago that fighting Microsoft with a closed-source cross-platform SDK was hopeless, which is why they dumped Bedrock.

Open source, though, has a big advantage here. Open source doesn't go away if the vendor does. And this is something marketed to programmers, who can fix the thing. So if a free GUI toolkit for Windows gets a reasonable amount of use, it can live for a long time. An open source version may have a bigger potential market than a closed source version. The vendor may have to fund the project by selling support, but then, that's the Red Hat business model.

Theres a reasonable Byte article on these things here. These things sounds like DVD killers, assuming they're practical. They hold more, they're just as speedy, and they could be made at a good price.

But.. they're 5 inches across still. When are we going to get something smaller? Why not stick 30Gb on a 2 inch disc? That'd be a killer for portables.

http://www.byte.com/column/B e_V iew/BYT20001031S0001

http://www.benews.com/c gi- bin/mwf/topic_show.pl?id=13857

. .

Since I know nothing about this subject since yesterday when I came across

this excellent review of SensAble feedback device in Byte.com

the link's all I have to say. But the review of the SensAble kit seems to have some good insights.

==Idle Random Thoughts. Usual Disclaimers Apply==

Knowing that /. is a Linux haven, I still disagree that some of these items are myths. Particularly in the "Installation" portion of the Linux myth dispeller.

I contend that the Linux experts have forgotten how difficult Linux can be when you're new to it. In the case I present below, this was my very first fresh installation of a UNIX/Linux system, although I'm familiar with UNIX and have been a system admin for several years now (small network).

I have set up a single Linux machine as my firewall/masqerade and mail server, and I can say the following:

1. While installing Linux as a workstation may be possible for the average computer user (one who could install Windows), I do not believe it to be easy. It would certainly take more than the hour or so mentioned in the myth dispeller. And unless you have significant experience as a UNIX developer and/or system admin, I am absolutely sure that you will not be able to install Linux as a server in an hour or two.
2. I had more difficulty installing and configuring the utilities than I did installing and configuring the OS. But without the utilities, the system was pretty much useless to me. By "utilities", I mean ipchains, qpopper (a POP3 server), mail, and ftpd. And, of course, once I got into this stuff I had to do kernel recompiles.
3. Without the linuxconf utility (provided with Red Hat 6.2) I would have been completely lost. Even so, linuxconf contained an error (or a confusion) concerning the configuration of multiple NE2000-compatible network cards. Fortunately, I found on-line help (which I believe to be one of the TRUE advantages of Linux).
4. Even Jerry Pournelle, computer god extraordinaire, had to call an expert (see this link). And even he admits that it's easier to configure WinNT as a server than Linux (see this link).
5. The first thing the Red Hat installation program did was ask me for a driver disk. Huh? I had no idea what it wanted. After some experimentation, I realized I could just hit cancel and continue with the installation.

Here are my counters to the myths:

1. Linux is difficult to install if you have not done it before, especially if you wish to use it as anything other than a workstation. It does, however, come with some utilities that make it easier to install than other UNIX flavors.
2. If you are installing a Linux system for the first time, make sure you have access to a UNIX/Linux expert. It will save you a lot of time and frustration.
3. Even though Linux is open-source, there is a higher likelihood of errors (or confusions) in the installation/configuration utilities (I ran into some, as I mentioned above). The reason for this is that the open source community doesn't put as much money and effort into testing (or into ease of use) as is done by a company like Microsoft or Apple. Also while the open source community may be a massive code review resource, I don't see the installation/configuration utilities getting as much scrutiny as they would at a large company.
4. Installation/configuration program errors (or confusing instruction) have a smaller chance of being fixed than similar errors would on Windows/MacOS, because:
   1. Someone has to notice the errors, find out who to report them to, and actually report them.
   2. Some programmer, somewhere, has to decide that he/she wants to fix this in his/her spare time. And we all know how exciting it is to work on installation/configuration programs.

Knowing that /. is a Linux haven, I still disagree that some of these items are myths. Particularly in the "Installation" portion of the Linux myth dispeller.

I contend that the Linux experts have forgotten how difficult Linux can be when you're new to it. In the case I present below, this was my very first fresh installation of a UNIX/Linux system, although I'm familiar with UNIX and have been a system admin for several years now (small network).

I have set up a single Linux machine as my firewall/masqerade and mail server, and I can say the following:

1. While installing Linux as a workstation may be possible for the average computer user (one who could install Windows), I do not believe it to be easy. It would certainly take more than the hour or so mentioned in the myth dispeller. And unless you have significant experience as a UNIX developer and/or system admin, I am absolutely sure that you will not be able to install Linux as a server in an hour or two.
2. I had more difficulty installing and configuring the utilities than I did installing and configuring the OS. But without the utilities, the system was pretty much useless to me. By "utilities", I mean ipchains, qpopper (a POP3 server), mail, and ftpd. And, of course, once I got into this stuff I had to do kernel recompiles.
3. Without the linuxconf utility (provided with Red Hat 6.2) I would have been completely lost. Even so, linuxconf contained an error (or a confusion) concerning the configuration of multiple NE2000-compatible network cards. Fortunately, I found on-line help (which I believe to be one of the TRUE advantages of Linux).
4. Even Jerry Pournelle, computer god extraordinaire, had to call an expert (see this link). And even he admits that it's easier to configure WinNT as a server than Linux (see this link).
5. The first thing the Red Hat installation program did was ask me for a driver disk. Huh? I had no idea what it wanted. After some experimentation, I realized I could just hit cancel and continue with the installation.

Here are my counters to the myths:

1. Linux is difficult to install if you have not done it before, especially if you wish to use it as anything other than a workstation. It does, however, come with some utilities that make it easier to install than other UNIX flavors.
2. If you are installing a Linux system for the first time, make sure you have access to a UNIX/Linux expert. It will save you a lot of time and frustration.
3. Even though Linux is open-source, there is a higher likelihood of errors (or confusions) in the installation/configuration utilities (I ran into some, as I mentioned above). The reason for this is that the open source community doesn't put as much money and effort into testing (or into ease of use) as is done by a company like Microsoft or Apple. Also while the open source community may be a massive code review resource, I don't see the installation/configuration utilities getting as much scrutiny as they would at a large company.
4. Installation/configuration program errors (or confusing instruction) have a smaller chance of being fixed than similar errors would on Windows/MacOS, because:
   1. Someone has to notice the errors, find out who to report them to, and actually report them.
   2. Some programmer, somewhere, has to decide that he/she wants to fix this in his/her spare time. And we all know how exciting it is to work on installation/configuration programs.

Jon Udell wrote this column at byte: Selling Ice Cubes To Eskimos

excerpt: "Contentville isn't selling ease of access or convenience, so much as an editorial process that selects and (one would hope) contextualizes items of interest. Is that valuable? Hugely so. In an era of crushing information overload, we depend on editorial services to draw important matters to our attention, and frame them in a useful context. What bothers me about Contentville is the way it misrepresents the nature of its service. If there's real editorial effort involved (and you can, of course, judge for yourself the quality of Contentville's editorial process), then isn't that the product I should be asked to pay for? Or as in the case of byte.com, that advertisers should be asked to pay for?"

• As soon as you start talking about sending money over the Net, you come up against the same issues of authentication and certification that have us all paying Verisign et al to vouch for the fact that we are who we say we are.

Alex Bischoff
---

Well said...

Joe Brancatelli says much the same in Apple's Driven Me To Wintel.

The Amiga Foundation Layer

1994 Byte intro to Taos

IBM was showing its new ThinkPad lines, including those that offer pre-loaded Caldera Linux.

Promising, isn't it?

Seriously though I see to recall an article back in the old, dead tree edition. Er, just a sec

...Man, I wish all the magazines had archives as complete as the old Byte ones. Anyhoo, they have a couple links about holographic storage there too:

1996 and 1998

Not bad, but essentially saying the same thing as the tweak3d article. Some of the other, non-holographic versions of storage sund like they may actually see the light of day first.

Seriously though I see to recall an article back in the old, dead tree edition. Er, just a sec

...Man, I wish all the magazines had archives as complete as the old Byte ones. Anyhoo, they have a couple links about holographic storage there too:

1996 and 1998

Not bad, but essentially saying the same thing as the tweak3d article. Some of the other, non-holographic versions of storage sund like they may actually see the light of day first.

Seriously though I see to recall an article back in the old, dead tree edition. Er, just a sec

...Man, I wish all the magazines had archives as complete as the old Byte ones. Anyhoo, they have a couple links about holographic storage there too:

1996 and 1998

Not bad, but essentially saying the same thing as the tweak3d article. Some of the other, non-holographic versions of storage sund like they may actually see the light of day first.

Amiga/Tao`s JVM technology is according to ARM "Because of the patented techniques, the intent JTE runs Java applications extremely quickly, more than 30 times faster than competitor's products"!

President of Amiga said at an Amiga show in Germany that it was around 22x faster with multimedia than ANY other JVM including Sun`s efforts. He also said Sun was introducing this new technology to its clients and it will become the new JVM standard. He also demonstrated the new SDK with blazing Java speeds.

An Amiga assembler coder founded a company in 1992 which is today known as the Tao Group.Read this interesting article. Its platform independence goes much further than solely application, but kernel, device drivers etc as well!

Well, there's always the ever-trusty Linux Router Project single-floppy distribution. That's exactly what it's designed for: a single-floppy that can do NAT/IP masq/routing etc. Unfortunately, Dave Cinege, the maintainer of the official distribution recently suffered a major systems failure, so the website might not be up. Might want to try the catch-all info site, lrp.c0wz.com for mirrors and better information, as the main site is outdated, anyway.

Also, there's a spinoff distribution using 2.2.x, and named after mountains. Previously there was Materhorn, and now it's Eiger, I think. It's maintained by Matthew Grant and is located at lrp.plain.co.nz.

There's also a commercial LRP spinoff called Coyote Linux. Looks pretty easy to use, but it costs money if you want a Windows-based disk creator (the free one is Linux based).

Trevor Marshall at Byte did a series of articles on using LRP as a home router. You can find them starting here to see how to have just a modem and your 10bT NICs set up. They continue here to add in DHCPd and 100bT cards, which teaches you all about LRP modules. Not sure there are any more in the series, but you can look around Byte's site.

Well, there's always the ever-trusty Linux Router Project single-floppy distribution. That's exactly what it's designed for: a single-floppy that can do NAT/IP masq/routing etc. Unfortunately, Dave Cinege, the maintainer of the official distribution recently suffered a major systems failure, so the website might not be up. Might want to try the catch-all info site, lrp.c0wz.com for mirrors and better information, as the main site is outdated, anyway.

Also, there's a spinoff distribution using 2.2.x, and named after mountains. Previously there was Materhorn, and now it's Eiger, I think. It's maintained by Matthew Grant and is located at lrp.plain.co.nz.

There's also a commercial LRP spinoff called Coyote Linux. Looks pretty easy to use, but it costs money if you want a Windows-based disk creator (the free one is Linux based).

Trevor Marshall at Byte did a series of articles on using LRP as a home router. You can find them starting here to see how to have just a modem and your 10bT NICs set up. They continue here to add in DHCPd and 100bT cards, which teaches you all about LRP modules. Not sure there are any more in the series, but you can look around Byte's site.

> Don't believe everything they tell you.

He! I saw it with my own eyes! I tell you it is capable of running 3 versions of Quake and 2 versions of Doom simultaniously at full speed while hosted on Linux! Without an host (read Elate RTOS only) it will be even faster (up to two times as fast!)

> A faster interpreter does not help very much.

It is not an interpreter.(The JVM was totally rewritten in VP code) It uses some new translating technique. VP code is very compact and the translation happens in loadtime. The code is mostly smaller than the native code into which it will be translated! The translation is so fast that it can translate the code faster than it can be read for an harddrive.

> Java is ahead of its time in a way, since it is not dependent on specific instruction sets.

Sun has done a good job. But Tao is the real expert at this. In 1992 an Amiga assemble Games writter erected a new company upon his great new idea of an hardware independent OS. 8 years later his company is ready to revolutionize the computing industry.

Read this interesting article from 1994.

http://www.netsaint.org

I just set up NetSaint this morning to monitor some servers, and I am really impressed with it. It took me less than 2 hours to set up to monitor an entire network, and that included reading the (very good) documentation.

You can monitor ping results, system loads, disk space, users, zombine processes, HTTP, PostgreSQL, etc. etc. etc. on Linux/other UNIX boxes, Windows boxes, and printers. The results can be viewed from the command line, or through CGI scripts. The CGI scripts show network status maps (all your servers at once) and can even show them using VRML! This thing is awesome! Alerts can be sent via e-mail or pagers.

There's also an article about setting up NetSaint in Byte:

http://www.byte.com/column/BYT19990728S0008

The article talks about what's coming out in ms Office 2001 and how ms is using their desktop monopoly to force their way into PDA market (something their OS software hasn't allowed them to do).

From the article:
Ah, but what if you have a Palm organizer? Isn't this new-fangled PIM redundant, since you're already working with a (portable, don't forget) calendar, to-do list and contact list? It seems that way to me, but Microsoft says the new application will synchronize with Palm's address book, tasks, and notes. E-mail remains a problem. Clearly, the mail integration offered by Office 2001 is its main selling point when compared to the Palm solution. Still, if you have a Palm, it seems that Office 2001 will complicate your life rather than simplify it.

[sigh].

This is an omission of choice for purely marketing reasons. ms knows how important the PDA and wireless market will be. They know they have a poor product in their OS. They know they need at least a 40 share to be relevant to the platform in the future. So they ignore the overwhelming demands of their users to "just make it easy" and choose to intentianly place roadblocks between Office and Palm. Forcing their users to stay locked into the "ms way" and punish the users that don't.
___

I think piranha does just this. Byte just ran a look at Linux HA clusters: http://www.byte.com/column/BYT20000510S 0001

Go to Byte magazine and check out John Udel's columns dealing with NNTP groupware. You may think you want something less complex than /. , but even slashdot isnt mail or newsreader based (sadly). If you want pple to post to a mailserver or use a client like that, definately read John Udel or his books. Sorry no link but i'll put one here from a search on Byte

Well the option is called save as HTML, not XML. I'm not complaining about the format not being well-formed XML.

HTML 4.0, on the other hand...

BTW, someone pointed out this byte article, which I pretty much agree with.

Basically, it's already a great step to have a format which is more or less human readable. This will makes it alot easier to create third party parsers/conversion tools.

Also, if you are looking for a word processor that does use XML -- genuine, correct, pure XML -- as it's native format, you should check out AbiWord.

It's all I use now, ever since I got sick of giant, bloated, slow StarOffice, the office suite that thinks it's an operating system. It's nice to finally see a word processor aimed at people who need want a word processor, not a spreadsheets not an email program, and not sure as hell not an "integrated desktop". Stupid StarOffice. Oh yeah, and AbiWord is GPL'd, unlike StarOffice. Kick ass.

The bus came by and I got on
That's when it all began
There was cowboy Neal
At the wheel
Of a bus to never-ever land

It's not really XML based... More like HTML with some XML-like stuff sprinkled in.

Byte has a short description of the Word format that you might want to look at.

I've looked a little at the Excel format. Once thing that seems clear it that the O2000 formats are almost human readable. It shouldn't be that difficult for someone to whip a converter -- well, it should be easier than parsing the binary formats.
--