Slashdot Mirror

Greets!

The reason that 3D isn't popular or practical - paper. Our current metaphor for information derives from Xerox Parc, a PAPER company. A faithful emulation of an office desk is NOT the best way to represent the complex infoverse we live in.

And the current web is not the best way to represent it either. Go back to hypertextual research before the web - look at Guide, look at Microcosm, before the brain damage of HTML and Mosaic set in.

Even better, go and look at Xanadu and ZigZag - representing information and the relationships between individual pieces of it is a complex task, perhaps made harder by our current metaphors. See ANYTHING by Ted Nelson, such as his technical briefing at the latest Hypertext conference.Read Vannevar Bush's "As we may think"

I would argue that we don't need 3D browsers, but MULTIDIMENSIONAL infoviewers, that can let us define the relationships and properties that we are interested at any moment, AND LET US CHANGE THEM easily and intuitively - I still remember the only good part of Johnny Mnemonic - zooming around cyberspace - also, to a lesser degree Lawnmower Man.

This is the way forward, and we need to learn from the games industry - Look at Homeworld, Q3D, even Elite - these are the kind of intuitive navigational and representational metaphors we ned to adopt to allow people to create, browse, populate and interact with their information.

Let us be imaginative, and move forwards to a representation of information as something we can use, rather than something we write down.

Links:
Microcosm:(Home) http://www.iam.ecs.soton.ac.uk/
(Review)http://www.man.ac.uk/MVC/SIMA/mcosm/toc.ht ml
Guide: http://www.iath.virginia.edu/elab/hfl0142.html
HyperText Conference: http://www.ht01.org/
GZigZag - http://gzigzag.sf.net
Xanadu: http://www.udanax.com
http://www.xanadu.com
As We May Think: http://www.theatlantic.com/unbound/flashbks/comput er/bushf.htm
The electronic labyrinth - a good intro to hypertext, slanted toward literature http://www.iath.virginia.edu/elab/elab.html

It's nice to see that so far more projects have been hilighted for being open than for not. I'd like to add Galeon to the list. The devel list happily accepts patches, even my poor attempts to produce graphics for it.

Umm just use Fin or Null scanning and you'll be fine. Nmap is very proficient(sp) with these scans, if Syn is logged Fin, Xmas, Null will keep you under the radar and out of sight.

NMap

correct me if i'm wrong :}
CAEthaver2

--mikeeusa--

>> Any properly administrated linux box has a >>decent iptables / ipchains script. If not, it's >>about time to read the docs.
>>From what I've read in the article, tripwire >>should be able to detect an infection. Not so >>much to worry about, I guess.
>>... and of course nmap to scan for open 5503 >>ports (damn, it's now illegal to do so here at >>our university).

In your case I'd probably choose docbook, but if you're just looking for a way to automatically setup the links between several XML documents you might take a look at w3make (which I wrote) or a similar project called XWeb which was written by Peter Becker.

... that the entire economy could revolve around a single corporation. The downfall of Microsoft could only help the economy. To suggest that killing (to use the violent terminology geeks are so famous for in the software design world) Microsoft is killing the US economy is downright preposterous. Take, for example, Sony corporation. While based in Japan, its US assets siginificantly exceed those of Microsoft. Microsoft only looks gigantic because Bill Gates happens to be the man who holds the world records for gaining, losing, and posessing the most money. That's only because he owns 30% of Microsoft's stock.

--
euphoria

I'm going to answer a slightly different question, why do I do it.

There are a few distinct reasons I do it:

1. Sometimes it the same reason I take photographs, I just want to do something creative. The goal isn't to solve a problem for myself, or anyone else, but just to have some fun. Once I've done that it seems like a good idea to share the code, in case others want to have fun. Sourceforge is nice here because I can share the code with minimal effort, and no commitment of my resources. To some extent both w3juke and xtank (anyone have a link? I stopped screwing with it almost a decade ago!) are explained by this.
2. Some times I'm a little more focused. I have a method I want to test out, and either no problem at work can benefit, or I'm a little leery of experimenting with that method at work. I learned how to use the STL on my own dime for example. This covers a lot of ground (w3juke was my STL playground for a while which is why it has odd uses of things like priority queues, and O(logN) algos used for datasets not likely to have more the 50 or so elements!). Again, if the code is sound, there is no reason not to share it.
3. Some times it really is to solve a problem I'm having (either not at work, or not one that justifies work time at least). My mkavi program (not currently up -- it was at my last employers anon ftp site, maybe I'll get it off to sourceforge soon) is an example of that. I really wanted to convert some raytracings I had made into movies, and didn't have a Windows box or anything. So I wrote a program that did it. I didn't have a 100% free license (it was a "free for noncommercial use" deal), and an aerospace firm actually ended up paying a little money for it, enough to buy me a windows machine if I wanted :-). However it counts for some definitions of Open Source, and more importantly it describes a motive I have for doing Open Source work sometimes.
4. Sometimes I do Open Source work on other people's code because it almost solves a problem I have (frequently I thought it did, but it turns out not to due to a bug). That is not quite the same as above, but pretty close.

That said, there are some reasons I don't do Open Source:

1. To get recognition. True, I got my first full time job as a result of writing xtank, but by and large few people get any real recognition for writing Open Source code. A few get to be Rock Stars (Linus for example), more get known only in "the field" (Keith Bostic for example), but very few people get much recognition. Esp. with the advertising clause of the BSDL being on the outs these days :-)
2. To change the world. The world ain't so bad, and my code seems unlikely to change it anyway.
3. To make money. It's too haphazard. Consulting work makes money, making money off Open Source seems too hard. I do try to get my (rare) consulting clients to let me Open Source code I produce for them, but it rarely pans out (it did for a Real Estate firm, anyone want some crappy DB code? It's free!).
4. To destroy Microsoft. Don't get me wrong, I would like to destroy them, I just don't think my software will do it.

I'm going to answer a slightly different question, why do I do it.

There are a few distinct reasons I do it:

1. Sometimes it the same reason I take photographs, I just want to do something creative. The goal isn't to solve a problem for myself, or anyone else, but just to have some fun. Once I've done that it seems like a good idea to share the code, in case others want to have fun. Sourceforge is nice here because I can share the code with minimal effort, and no commitment of my resources. To some extent both w3juke and xtank (anyone have a link? I stopped screwing with it almost a decade ago!) are explained by this.
2. Some times I'm a little more focused. I have a method I want to test out, and either no problem at work can benefit, or I'm a little leery of experimenting with that method at work. I learned how to use the STL on my own dime for example. This covers a lot of ground (w3juke was my STL playground for a while which is why it has odd uses of things like priority queues, and O(logN) algos used for datasets not likely to have more the 50 or so elements!). Again, if the code is sound, there is no reason not to share it.
3. Some times it really is to solve a problem I'm having (either not at work, or not one that justifies work time at least). My mkavi program (not currently up -- it was at my last employers anon ftp site, maybe I'll get it off to sourceforge soon) is an example of that. I really wanted to convert some raytracings I had made into movies, and didn't have a Windows box or anything. So I wrote a program that did it. I didn't have a 100% free license (it was a "free for noncommercial use" deal), and an aerospace firm actually ended up paying a little money for it, enough to buy me a windows machine if I wanted :-). However it counts for some definitions of Open Source, and more importantly it describes a motive I have for doing Open Source work sometimes.
4. Sometimes I do Open Source work on other people's code because it almost solves a problem I have (frequently I thought it did, but it turns out not to due to a bug). That is not quite the same as above, but pretty close.

That said, there are some reasons I don't do Open Source:

1. To get recognition. True, I got my first full time job as a result of writing xtank, but by and large few people get any real recognition for writing Open Source code. A few get to be Rock Stars (Linus for example), more get known only in "the field" (Keith Bostic for example), but very few people get much recognition. Esp. with the advertising clause of the BSDL being on the outs these days :-)
2. To change the world. The world ain't so bad, and my code seems unlikely to change it anyway.
3. To make money. It's too haphazard. Consulting work makes money, making money off Open Source seems too hard. I do try to get my (rare) consulting clients to let me Open Source code I produce for them, but it rarely pans out (it did for a Real Estate firm, anyone want some crappy DB code? It's free!).
4. To destroy Microsoft. Don't get me wrong, I would like to destroy them, I just don't think my software will do it.

Damn, first I thought he wanted Freenet :-)

Warning: shameless plug, but very helpful.

We're working on it: http://sf.net/projects/trovesendtwo/ And we could use some help if you're JAPH...

All this needless effort can be avoided. On the author/maintainer side a client app should be able to update these product directories, check for diffs in listings, and save the author/maintainers a lot of work and track the history of the product changes. If a directory itself disappears the data is NOT lost. Just upload it all to the next one. And the next one. And the next one. The directory people can now go about promoting their site (and your products).

The problem with sites like this one are the amount of work created for everyone, code authors/project maintainers & the directory people. And thus they die or burnout. And now all the work is lost?? The site is now just a press release. It's a shame.

One side benefit, to the people you are trying to reach with your products, is that you can disseminate your product info to as many directories/sites with no more effort than updating once and letting the client-app do all the work of updating those sites you choose to list your product at.

Warning: shameless plug, but very helpful.

We're working on it: http://sf.net/projects/trovesendtwo/ And we could use some help if you're JAPH...

All this needless effort can be avoided. On the author/maintainer side a client app should be able to update these product directories, check for diffs in listings, and save the author/maintainers a lot of work and track the history of the product changes. If a directory itself disappears the data is NOT lost. Just upload it all to the next one. And the next one. And the next one. The directory people can now go about promoting their site (and your products).

The problem with sites like this one are the amount of work created for everyone, code authors/project maintainers & the directory people. And thus they die or burnout. And now all the work is lost?? The site is now just a press release. It's a shame.

One side benefit, to the people you are trying to reach with your products, is that you can disseminate your product info to as many directories/sites with no more effort than updating once and letting the client-app do all the work of updating those sites you choose to list your product at.

Just for kicks I tried to compare ImLib2 to libAfterImage, and guess what - its not really all that fast, but it is definately bloated at the same time.

here and here

Hell, there are so many people on these DSL networks running compromised servers, it's like a shotgun wound. You don't care where the bullets went in or where they came out. You wrap the thing until it stops bleeding.

Or, guess what, you're dead. Right now they have a few limited resources, 1) Bandwidth and 2) Time. If they want to fix the problem fast they cut port 80. They let the bleeding stop and then they open up.

If you want to write a program that helps them signature each freaking IP on their network and then filter which one is okay or not, go start your project: SF

These guys want to keep most of their customers. When 100% users are having bandwidth problems because of a virus they drop like flies. When 1% that are running Linux get port 80 blocked drop, they don't give a @#$!.

Quit your whining. And go get webhosting for 10$/month at an offsite provider instead of trying to create a web server from crappy components. You're not losing money are you?!!? I'd have to laugh in your face if your trying to run a useful webserver over DSL.

-out-

You want both?

Check out Falcon's Eye. It is a visually pleasing version of Nethack that I've been totally addicted to. It even has a big intro and a soundtrack. How can you go wrong?

"they will deny the charge and, if that doesn't work, deny that the GPL is enforceable."

As for it being "spamware", I would be shocked if it was ever used for spam. First, companies using real CRM software (not ACT!) aren't likely to be spamming in the first place. CRM software is partly intended to help target marketing campaigns. The software I wrote interfaced with the query and report packages in the CRM system to generate the distribution list for a document. It would then mail merge the document to be sent, based on the queries and reports.

Here's a typical usage scenario: You're company makes widgets. You have recently introduced the "Green Widget Extender", for use with green widgets. You want to send a press release, but only to the clients using green widgets already. So you create the press release, complete with mail merge fields. You then enter a query for all clients using green widgets. The fax/email server then mail merges the press release, and sends it onto the receivers fax or email box. The receiver gets a document:

Ms. Purchases Everything
Head of Purchasing
SomeCompany
Dear Ms. Everything:
Increase the productivity of your green widgets with the new green widget extender! ...

Furthermore, the software had a field in the database specifying the preferred delivery method for a contact: fax, email, or snail mail. If the preferred method was fax or email, the message server would distribute it electronically. But if the preference was snail mail, then the message server would mail merge the document, and spool it for printing along a mailing label so it could be snail mailed. So the contact only received faxes or emails if that's what they preferred. Assuming of course that the correct information was in the database!

Finally, an inexpensive install of the software, including the base packages, message server, and all the configuration, was usually over $50,000 usd. Most of the clients who purchased the message server had installs that cost in excess of $200,000 usd. People that have this kind of money to blow on CRM software aren't wasting their time spamming people.

---
Geoffrey Wossum
Project AKO - http://ako.sf.net

So we rigged the program so that when it should have faxed, it instead secretly emailed the document along with the destination fax number to a fixed email address -- the email address of the guy we had sitting in the next room. So when the client wanted to fax something themselves to see it work, it really got emailed to our guy, who then faxed it manually. The client went away happy, thinking the program was fully functional. I got the fax functionality working using Symantec WinFax the next week, so no major harm was done. But I got a great story out of it.

---
Geoffrey Wossum
Project AKO - http://ako.sf.net

Spread it to the windows world too?
The "official" jabber client (official meaning the one by the people at jabber.com is released for windows only. Actually, the only reason I prefer Gabber is that it looks prettier (IMO), but I wouldn't count that as a usability issue :)

freenet:SSK@sUOkGXJDjktWahCNZmvg0sDkEKgQAgE/foldr. org/linux-2.4.7.tar.bz2
freenet:SSK@sUOkGXJDjktWahCNZmvg0sDkEKgQAgE/fold r. org/linux-2.4.7.tar.bz2.sign

What is Freenet?

I still use FreeBSD, though :)

It's functional, but there is no server actually hosting it, so you are limited to browsing a static copy of it, and reading its specifications.

It't not really secure either, it's a tool designed for communities, not for whole countries (not yet anyway). It could even replace a tool like Slashdot or K5H (conceptually at least)!

As far as I'm concerned, it's the best way to let a community express itself through writing any sort of texts: constitution, book of laws, newspaper, weblog, novel, poem... Those texts being assembled in a library which can be browsed using a filter (to filter the elements which have a lower acceptation level, thus to have a high signal/noise ratio).

It's called VeniVidiVoti, and is about 80% complete. I'm looking for a community willing to test drive it, or for server space where to install the project's library.

The gnome-python package is here:
www.gnome.org.

The package includes lots of simple example programs which are easy to read, and handles calling the init functions for you.

There are more tutorials on using python to create Gtk apps here.

Retrieve with:

http://localhost:8081/KSK@galeon.png
or
freenet_request KSK@galeon.png galeon.png

Freenet: http://freenet.sf.net

The CHK for this key, for the paranoid, is:

CHK@iE7SmyIIP8rYKqT77jhdJjDcgB8OAwE,OHOBWuZQ703Mw9 YpjUxFpA

"The Slashdot Effect is good for Freenet" - Gill Bates

If you're running a Freenet node, you can grab the kernel source using the following key:

freenet:SSK@sUOkGXJDjktWahCNZmvg0sDkEKgQAgE/foldr. org/linux-2.4.6.tar.bz2

It sure makes monitoring my Snort logs a lot easier.

Check out my HTTPMail project which is a client for the HTTPMail protocol used by Outlook Express for Hotmail and MSN mail. There's even a POP3 proxy.
--

ixlib features ECMAScript 4 (=JavaScript 2) with classes and inheritance, it has a focus on embeddability, and it offers an easy C++ interface. Why would you need CSS? :^)

Another collection of Win32 binaries of typical Unix tools can be found at SourceForge: the
GnuWin32 project.

The RISC OS scheme has security problems on multiuser systems (running !Boot when you look at a directory is not good!) but ROX doesn't use boot files.

Incidentally, ROX uses Gtk and XDND, so it should play nice with GNOME apps.

You can.

Archive, System and several other applications install like this. You just need a filer that supports it.

You can.

Archive, System and several other applications install like this. You just need a filer that supports it.

You can.

Archive, System and several other applications install like this. You just need a filer that supports it.

So instead of one mega OSS site, we'd have a large network of smaller machines hosting only as many projects as they could.

Once you get Freenet.sf.net hosted somewhere else, you can throw the contents of SourceForge's web space onto Freenet. Do mailing lists through Yahoo! Groups. But you'll still need some sort of replacement for the CVS services.

I think you must be talking about VirtualDub at virtualdub.sf.net? This is an awesome win32 app for linear editing and video capture (allows you to compress with DivX, for instance). It's one of the most professional pieces of software I've seen...

Mac OS X OTOH hides much of the filesystem and brings Start menu-like abstractions. A nice touch is the package system, where an app (or lib, or whatever) behaves like a single file, yet actually is a whole hierarchy of files and dirs for the OS. Makes packaging complex apps really nice, and allows drag-and-drop installs just like the Old Days.

Absolutely! The problem with the Windows and Unix filesystems isn't that they're big, it's that they're organised is a system-friendly, rather than user-friendly, way.

A user-oriented filesystem would put a program's binary and its help files together, for example, rather than having a 'bin' directory and a 'doc' directory.

As you say, if you lay things out sensibly then you don't even need an installer, just cp / tar will do (or graphical equivalents).

<Plug>Anyway, check out the ROX Destop which supports this already!</Plug>