Slashdot Mirror

you might be interested in this PXES. version 0.9 was released yesterday but never mind that.

this will not provide you with a thin client but with dummy clients. meaning that the client boots from the network via a pxe enabled card (most modern nics support this) and therefore you can disable (or remove) the floopy drive, the cdrom, the dvd, and the hard disk.

translation: the user cannot physically access the operating system except thru the network.

its pretty simple to setup and can even be used with windows terminal services but i do not recommend that legacy OS. in fact, i will recommend mandrake linux for several reasons:
1. draksec will sandbox your users very fast without much interaction from you (if required)
2. if you are not able to set up the server side on pxes. employ drakTermServ. mandrake's dummy terminal solution that comes standard with its distro.
3. it comes with kde which has a kiosk mode available (you will have to do your own research i m afraid)
4. it comes with openoffice.org meaning that your users can also read and write word documents, excel sheets, powerpoint presentations, adobe reader pdf (write!) and flash swf among other nice stuff.

did i mention the dummy clients wont have an operating system? meaning that you will only administer the operating system from the terminal server.

furthermore, (2 or 3 years ago) largo florida has a similar setup for 400 pc running of a 1gb ram dual p3-900 server or something like that

and if you are really stingy with money go get the hp d441 4 monitors, 4 keyboards and mice and 1 pc.

enjoy

Thinstation is a 'distro' that i'm currently using at work (a hospital). It can be used to connect to Citrix, RDP, VNC, Unix, Telnet/SSH, or (with the help of fluxbox/icewm) as a lightweight standalone linux workstation (with an optional FireFox package). The people on the mailinglist are VERY helpful as well, so you don't need to worry about support when you've a problem.

I can really recommend it as a thinclient solution.

It doesn't matter if Internet Explorer supports this. The great thing about XForms is that it can be implemented on the server-side. Take a look at chiba, which is real technology you can use today (and I am using), to develop with XForms. Servlets not good enough? Take a look at 10 different XForms implementations Who cares what IE does?

JACK uses a callback based API much like Core Audio.

Basically every high-end (e.g. ardour, JAMin, Rosegarden, Hydrogen, etc.) uses it.

You can get really low latency using it if you have good sound hardware (e.g. RME Hammerfall for extremely low latency or even an M-Audio Delta 1010). Something like an SBLive! (what I have) will need a period size of 2048 bytes with two periods to avoid underrunning (I have a Dual AthlonMP 2800+ so I'm pretty sure it's the sound card...). Stuff like QJackCtl and Jack-Rack make controlling Jack easy.

Getting realtime mode working for a normal user can be tricky, but Debian makes it really easy. Just install the realtime-lsm package and build the realtime-lsm-source package for your kernel and all users in the audio group gain the ability to run applications realtime (at least with the default config). It could be made easier (mainly by prebuilding the realtime-lsm modules for the stock kernels) but GNU/Linux pro-audio is still mostly for hackers and adventurous people right now. Stuff like PlanetCCRMA and AGNULA are aiming to make everything work out of the box. I have yet to try either (I use Debian so PlanetCCRMA is useless for me) but it looks like DeMuDi has everything set up for recording out of the box.

JACK uses a callback based API much like Core Audio.

Basically every high-end (e.g. ardour, JAMin, Rosegarden, Hydrogen, etc.) uses it.

You can get really low latency using it if you have good sound hardware (e.g. RME Hammerfall for extremely low latency or even an M-Audio Delta 1010). Something like an SBLive! (what I have) will need a period size of 2048 bytes with two periods to avoid underrunning (I have a Dual AthlonMP 2800+ so I'm pretty sure it's the sound card...). Stuff like QJackCtl and Jack-Rack make controlling Jack easy.

Getting realtime mode working for a normal user can be tricky, but Debian makes it really easy. Just install the realtime-lsm package and build the realtime-lsm-source package for your kernel and all users in the audio group gain the ability to run applications realtime (at least with the default config). It could be made easier (mainly by prebuilding the realtime-lsm modules for the stock kernels) but GNU/Linux pro-audio is still mostly for hackers and adventurous people right now. Stuff like PlanetCCRMA and AGNULA are aiming to make everything work out of the box. I have yet to try either (I use Debian so PlanetCCRMA is useless for me) but it looks like DeMuDi has everything set up for recording out of the box.

JACK uses a callback based API much like Core Audio.

Basically every high-end (e.g. ardour, JAMin, Rosegarden, Hydrogen, etc.) uses it.

You can get really low latency using it if you have good sound hardware (e.g. RME Hammerfall for extremely low latency or even an M-Audio Delta 1010). Something like an SBLive! (what I have) will need a period size of 2048 bytes with two periods to avoid underrunning (I have a Dual AthlonMP 2800+ so I'm pretty sure it's the sound card...). Stuff like QJackCtl and Jack-Rack make controlling Jack easy.

Getting realtime mode working for a normal user can be tricky, but Debian makes it really easy. Just install the realtime-lsm package and build the realtime-lsm-source package for your kernel and all users in the audio group gain the ability to run applications realtime (at least with the default config). It could be made easier (mainly by prebuilding the realtime-lsm modules for the stock kernels) but GNU/Linux pro-audio is still mostly for hackers and adventurous people right now. Stuff like PlanetCCRMA and AGNULA are aiming to make everything work out of the box. I have yet to try either (I use Debian so PlanetCCRMA is useless for me) but it looks like DeMuDi has everything set up for recording out of the box.

gaim's sf project page

So does my project.

As I mentioned on the linked page: "Also, the FSF makes all contributors attribute copyrights to the FSF. They do this for legal reasons. Mozilla did not, and when they decided to re-license, they had to contact every contributor. Because of this, we too require that any contributors attribute the copyrights to the jasabe project. Of course, you are still free to fork the project and keep your changes under your copyright, but we cannot accept your changes into the main jasabe tree."

Don't believe me? More info can be found on the FSF page as well as on their FAQ.

Not that it matters much for our project. It's only important if you have contributors. ;-)

The lack of a geneology program was the one thing holding me back from recommending Linux to her. But I recently ran across Gramps. The interface is different than Reunion, which is what she uses now, but it imports Reunion files just fine.

A Quicken replacement is my other concern. GnuCash is definitely powerful enough, but the interface is significantly different from Quicken. Plus, I like sticking with KDE apps if I can because of the better integration (when running KDE). I'll probably show her GnuCash and KMyMoney and she what she thinks of them. Hopefully the next release of KMyMoney will support enough reports that it'll work for her.

The next time she visits, I'm going to have her try out my machine, set up with a user account like hers would be. I'm interested to see what she thinks.

Install an SSH Server on the windows machine, use SHFS to mount the remote filesystem to a directory, then use rsync to copy it.

Figure out how to install ipw2200, and you've got that down to the DVD+RW.

I do use XFCE too, plus the Rox pinboard - makes the perfect combo, and still keeps the environment lean and fast.

However, I beg to disagree with your last sentence. The "bloat" in gnome is something relative - it may be heavier on the system, but Gnome and its apps feels far more integrated than XFCE. XFCE is pretty much only the panel, an eye-candied window manager, and a taskbar, and while it comes with easy to use configuration tools, they are very limited in the sense that there aren't not much room for customizing - something that gnome surely wins. The taskbar, for instance, have no real meaningful configuration, and always lives separated from the panel - IMHO, it should be a plugin, so you could attach the taskbar to the panel, thus freeing desktop space.

I can live with that tough. My main beef with the state of desktop on Linux is the fragmented situation of the GUI Tookits (mainly QT vs. GTK, though there are lesser ones). The problem is not having many toolkits per se, but the fact that this leaves the desktop with an unconsistent appearance. I'm all for having toolkit choices, but I wish they'd unite to create a standard themeing format, so a theme could be used on both toolkits, thus leaving a more or less consistent appearance to the desktop (there's still the GUI guidelines).

Well, not gonna happen anytime soon, tough... :-(

While I haven't used Cloudscape in a very long time, I imagine this is more competition to other Java open source databases like HSQL, Axion, or McKoi.

Most of these databases are used by "embedding" them into an application (something not uncommon in Java programming), not as a standalone database server like Oracle or Postgres. Of course, like I said, it's been a long time since I looked at Cloudscape so it could have changed to be more of a standlone server.

I'm also surprised I haven't heard more about this in Apache, but I imagine it will first go through the Apache Incubator to sort out any legal issues and then end up somewhere in the Apache Database project. If anyone has more info, I'm interested to know.

My way around this is to simply admit I am never going to complete a game.

I'm tackling the content side instead, sorta. What I have written basically is a terrain tile generator. I noticed a lot of the 'craft like games suffer from lack of good graphics, and figured that terrain was part of the graphics that might be easier to generate than to render individually.

If you are morbidly curious my stuff is at gasnippets but I don't consider it finished yet. Working on DirectFB and OpenGL variants of the same thing, to learn about the platforms more than anything else.

Doing it this way I get to work on something I find interesting (e.g. how to make good looking terrain in the range of 100000's of tile variations, without having to sit at gimp for 3 years), but not have to worry about the rest of the trappings for a game. When its done to a level I am happy with, I will post it to the 'craft like games I think could benefit from it.

Yes. You can find more about it here.

Our application is just about all Java, written in a J2EE application container, and almost all of our tests are written in some form of Java as well. We use Test Driven Design to drive our API design, and end up with unit tests written with JUnit. We run these unit tests continuously, after every batch of checkins, using Cruise Control which performs an automated build.

To perform functional tests, we use JWebUnit which sits on top of HTTPUnit. The combination emulates a web browser, performing clicks and form sets against the actual server. We automate the running of these tests against an installed system, using Tinderbox, and we do this on every platform that we support (combinations of application servers, databases, and operating systems). The tinderboxes run 4 times a day, using the Cruise Control results.

It takes a lot of hardware (the tinderboxes for each platform) but the infrastructure is easy to set up with free software. The HttpUnit tests are tedious to write, mostly because they take a long time to run. Because of this, we don't consider a feature (user story in XP) complete until it has an automated HttpUnit test.

As for process, we have a small QA team that design test documents for each user story. We use those test documents to design the HttpUnit tests. Since we do this during the development stage, we involve the QA team on a daily basis.

When a user story is considered finished by the development team, the tests are reviewed by QA and the functionality is reviewed by the product manager that designed the feature. This way, we have no big surprises when everything is done, everyone can see the product progress day by day.

I won't say we ship a 100% bug free product, but our quality is through the roof compared to what it used to be, and it's actually much higher than we were even expecting.

Our application is just about all Java, written in a J2EE application container, and almost all of our tests are written in some form of Java as well. We use Test Driven Design to drive our API design, and end up with unit tests written with JUnit. We run these unit tests continuously, after every batch of checkins, using Cruise Control which performs an automated build.

To perform functional tests, we use JWebUnit which sits on top of HTTPUnit. The combination emulates a web browser, performing clicks and form sets against the actual server. We automate the running of these tests against an installed system, using Tinderbox, and we do this on every platform that we support (combinations of application servers, databases, and operating systems). The tinderboxes run 4 times a day, using the Cruise Control results.

It takes a lot of hardware (the tinderboxes for each platform) but the infrastructure is easy to set up with free software. The HttpUnit tests are tedious to write, mostly because they take a long time to run. Because of this, we don't consider a feature (user story in XP) complete until it has an automated HttpUnit test.

As for process, we have a small QA team that design test documents for each user story. We use those test documents to design the HttpUnit tests. Since we do this during the development stage, we involve the QA team on a daily basis.

When a user story is considered finished by the development team, the tests are reviewed by QA and the functionality is reviewed by the product manager that designed the feature. This way, we have no big surprises when everything is done, everyone can see the product progress day by day.

I won't say we ship a 100% bug free product, but our quality is through the roof compared to what it used to be, and it's actually much higher than we were even expecting.

Our application is just about all Java, written in a J2EE application container, and almost all of our tests are written in some form of Java as well. We use Test Driven Design to drive our API design, and end up with unit tests written with JUnit. We run these unit tests continuously, after every batch of checkins, using Cruise Control which performs an automated build.

To perform functional tests, we use JWebUnit which sits on top of HTTPUnit. The combination emulates a web browser, performing clicks and form sets against the actual server. We automate the running of these tests against an installed system, using Tinderbox, and we do this on every platform that we support (combinations of application servers, databases, and operating systems). The tinderboxes run 4 times a day, using the Cruise Control results.

It takes a lot of hardware (the tinderboxes for each platform) but the infrastructure is easy to set up with free software. The HttpUnit tests are tedious to write, mostly because they take a long time to run. Because of this, we don't consider a feature (user story in XP) complete until it has an automated HttpUnit test.

As for process, we have a small QA team that design test documents for each user story. We use those test documents to design the HttpUnit tests. Since we do this during the development stage, we involve the QA team on a daily basis.

When a user story is considered finished by the development team, the tests are reviewed by QA and the functionality is reviewed by the product manager that designed the feature. This way, we have no big surprises when everything is done, everyone can see the product progress day by day.

I won't say we ship a 100% bug free product, but our quality is through the roof compared to what it used to be, and it's actually much higher than we were even expecting.

Wow, using sloccount on the full POPFile source shows that developing it would have cost around $500K in a regular software company. That seems about right given the length of time we've been working on it and the number of people involved. Cool tool.

Now if only I could push the donations up above $5,000 :-)

John.

My compliments for the quick post :) For everybody wondering what the post meant, see http://zoidberg.sf.net

What does it take to be a hacker?

I've been spending some of my recent spare time writing the dynamic recompilation core of a Z80 emulator. I'll probably shoot for the 68000 next. http://cottage.consolemul.com/

The project before that, I wrote a bytecode weaving tool that transforms Java 1.5 bytecode to Java 1.4 bytecode http://retroweaver.sf.net/

Java has an overwhelming share of mediocre developers, but that doesn't mean Java developers can't be "hackers".

I may be wrong, but my understanding of cable TV here in Australia has been that our population is simply too sparse to support wide-spread roll outs. Foxtel and a few other operators have cable around the big three cities (Sydney, Melbourne, Brisbane... what about Perth and Adelaide?) but everywhere else requires satellite. And satellite is an extra hassle, and I'm guessing is more expensive for the operator.

But I think you're also right about us not being "real heavy tv watchers". I used to get Foxtel when I lived in Sydney and while it had some good things from time to time, it never really engrossed me to a great deal. With the internet, computer games, Triple J radio, a good collection of DVD's, and several gigs of ripped MP3/Vorbis music, the five FTA channels are entertaining enough for me. I never find myself thinking "geez I'm bored and what I'd really like to do right now is flick through 500 channels of crap and vegetate".

Besides, there's always the D1 Home Media Centre as well as DIY MythTV and Freevo boxen.

I agree that 3D is over-used. It doesn't *have* to be used to create fun games -- all of these play on Linux, and most (maybe all?) play on Windows and other platforms as well.

SuperTux is a Mario Brothers "clone" that's really quite nice. Apparently it was the Linux Game Tome Game of the Month for March 2004, which is like Extreme Makeover for Linux games. OpenGL acceleration is nice (gives me about 30fps extra on my AMD64 3200+ with nVidia FX5900XT card), but it runs pretty decently without it. You should note that to make Tux run or shoot fireballs in the game, you need to hold down the left Control key... I didn't know that and ended up cheating with the level editor. I finally discovered the key-mappings, and I'm currently slightly addicted to it... :)

Pingus is a cool, open-source, penguin-themed (of course :) Lemmings-like game that doesn't require any kind of acceleration, just a decent system. There aren't many levels available yet, but it's a good-looking game and fun to play around with.

Project: Starfighter is also pretty good, although I find the levels a little too hard for my liking. Hardware acceleration is pretty much required, unfortunately.

And there's always the venerable and *highly* addictive Frozen Bubble... that game has wasted so much of my time... and I've enjoyed every minute!

I've been really impressed -- there seems to be an explosion of good, free games for Linux lately, though it could just be that I'm finally finding out about them. Although I don't have too much of a problem with 3D, I knew Neverball was getting to me when I started dreaming about it...

http://stabilizer.sf.net

The idea is to quickly make buggy GUI applications more useful, without requiring any involvement from the developers.

> How hard would it be to build in a small
> tiny bit of error that would only be
> useful in cases of calibration of
> high-tech weapons?

I think it'd be tricky, because it would break other high-precision things as well. And the other folks using the open source project would say "hey, this fellow Fred just submitted a patch. something looks odd about it. Fred, why does line 314 do a bit shift without checking the foobar?" And then the patch would be rejected.

> If 3000 lines of dense mathematically
> rich C were checked in

I doubt any maintainer would accept such a patch. I don't accept patches for PMD without reading them, and if I got a 3K line patch I'd reject it out of hand.

I think the OP missed some of the best talks:

XenoLinux - a virtual machine layer to support linux and other free OSes at almost native speeds.
Alot faster than UML!

CKRM - not new but I didn't know about it. From their sf site:
The Class-based Kernel Resource Management (CKRM) project seeks to develop Linux kernel mechanisms providing differentiated service to resources such as CPU time, memory pages, I/O and incoming network bandwith based on user defined groups of tasks called classes

TIPC - Transparent Inter Process Communication protocol is specially designed for intra cluster communication but definately not for the internet.

I think the OP missed some of the best talks:

XenoLinux - a virtual machine layer to support linux and other free OSes at almost native speeds.
Alot faster than UML!

CKRM - not new but I didn't know about it. From their sf site:
The Class-based Kernel Resource Management (CKRM) project seeks to develop Linux kernel mechanisms providing differentiated service to resources such as CPU time, memory pages, I/O and incoming network bandwith based on user defined groups of tasks called classes

TIPC - Transparent Inter Process Communication protocol is specially designed for intra cluster communication but definately not for the internet.

> Earlier they were dumping anything matching *mail*.
> We discovered this, this morning by searching for qmail sites.

Funnily enough, I was searching for `libgmail` (my Python binding for Gmail access) when it stopped working. But it only forbid it when I tried searching for a specific version number "0.0.6", which I eventually generalised to "..." being banned.

Initially I was like "Crap, they could've just asked me to stop working on it..!". :-)

Does OGG/FLAC come with licensing cost? I dunno for sure myself, but I don't think so, and the argument seems kind of weak.

My take (as an iPod owner - Linux powered :) is that Apple want control more than anything. As long as you're subscribing to Apple's view on what "insanely great" and "thinking differently" constitutes, you'll have a very good time with your Apple products and services.

Step off that trail though, and in Apple's eyes you change from being a revenue stream to a threat (and/or diluter of their vision). That's what licenses, patents, marketing (yes, they have invested a lot in getting thousands of foot soldiers^W^Wcustomers protecting and enforcing the company gospel) and a fierce legal department are to protect.

Not that they seem to be any different than other companies, maybe save for their zealous customers.

About time they added it. RoX had it for a while now.r

> Is there a linux P2P alternative?

I recommend giFT. You can find it at gift.sf.net, and modern distros should have it in their repositories (try something like "urpmi gift" or "apt-get install gift" to see if it automatically installs from the internet). You can use different front-end clients with it, which is a nice feature. Apollon is the popular gui front-end, and giFTcurs is an excellent ncurses-based (works in text terminals, but has tabs and layouts like a gui app) front-end. It can be attached to FastTrack (that's the Kazaa network) as well as Gnutella and gift's own OpenFT network. It's possible that somebody will eventually add support for other P2P networks, so I'm crossing my fingers and hoping. :)

--
-JC
http://www.jc-news.com/coding/freedom/

Is there a linux P2P alternative?

• Download and build pngcrush
• example usage:

  ./pngcrush --reduce --bit_depth 8 --brute file1 file2

Definetly the way to go.

I volunteer for a school in the Phoenix area that I've set up on a K12LTSP network.
The money that we've saved on licensing over the last two years has been enough for us to buy bigger better server hardware and replace most of our CRT monitors with flat panels. Also the time saving of centralized administration and not cleaning windows machines of spyware has been immense.

<shamless_plug>
And if you haven't seen it yet, check out k12wincd it may help you with getting students and parents familiar with open source software. I plan on sending every student home with a copy of the CD at the start of next year.
</shameless_plug>

Remember, K-12 computer classes should be teaching basic computing concepts (hardware,internet, word processing, spreadsheets, etc.) NOT teaching brand loyalty.

I'm happy to see an SWT-based GUI application other than the Eclipse IDE itself.

Maybe you should try Azureus, for your BitTorrent needs (I know, many are happy with btdownloadcurses :P this is just an example of another SWT-based app)

All except for PearPC...

no, but you can install Linux on your iPod...

This one I don't think is a very big deal

To me it is. To switch from FLAC to Apple Lossless would be a major pain in the ass. I have a lot of these files; hundreds of gigabytes worth. I keep them in FLAC format because they are supported more or less natively by:

• WinAMP and Xmms (with plugins)
• Nero (with a plugin)
• My Squeezebox
• Tag&Rename and MediaTagger, the two best audio file taggers I've encountered (Windows)
• Formerly, my Rio Karma. But it was a POS and I sold it on ebay.

FLAC is audio format that best suits my usage patterns, which is why I'd like to be able to use it with my iPod. Which, of course, is the best portable audio player around.

Converting everything over to Apple Lossless would be a large time investment, not to mention the fact that there don't seem to be any standalone encoders for it (URL me if you've got one, I'm all ears). I have a number of scripts I use for converting other lossless audio files to FLAC and tagging FLAC files I download, so cross-platform, command line tools are my preference.

boring...

Whats next? portage for OpenBSD? Well at least OpenBSD could make use of a 2nd port system... instead of Darwin getting the 3rd one.
With DarwinPorts and Fink there is already a source based and a binary based system for 3rd party applications.
Wonder what portage could brings us that isn't already there...

I've had a 10gb 2nd generation (Windows) iPod for two years. It works fantastic with Linux. All I needed was a firewire adapter and a newer kernel. (2.4 to start with, now 2.6). I've recently moved to gtkpod and it works great. With the right kernel options, you can have it installed and working in no time.

mLan is a superset of normal firewire audio (IEC 6something), which is starting to get some attention in JACK via the firewire generic driver. I'd do it myself with my fa-101, but the standard is kind of expensive.

Well, they're both audio and MIDI sequencers for Linux. They both support outboard MIDI gear, native Linux ALSA soft synths, and audio and transport synchronisation through JACK. Both look and feel somewhat like the big sequencers for other platforms. Both use Qt for their GUIs, though Rosegarden also uses the KDE libraries. I happen to think Rosegarden has the more polished GUI, but I'm biased (see disclaimer at bottom).

Both support arbitrary numbers of MIDI and audio tracks, with audio mixing and routing capabilities. Both support LADSPA effects plugins, of which there are hundreds available free. Both can (with some work!) be made to use VST effects through vstserver. MusE can host VST instruments with libfst using Wine; either of them can drive VST instruments hosted separately using jack_fst. None of these VST solutions is currently at all easy to configure and build, but see here for more information. Rosegarden is implementing the DSSI synth API and will probably ultimately use a DSSI wrapper for VST instruments. Rosegarden can also be run without audio support if you only want MIDI or score, whereas MusE always requires JACK.

My rather fuzzy impression of the difference in "feel" between them is that MusE feels like it came more from a studio/MIDI gearhead background, Rosegarden more from a composition background. MusE is a bit further ahead with things like instrument plugins, audio routing (send/returns etc) and automation. Rosegarden places a lot more emphasis on the score editor, whereas the one that used to be in MusE was actually removed completely for the 0.7 release.

Rosegarden is a somewhat bigger and more complicated program than MusE (c 200K lines of code vs 130K LOC), which you may see as good or bad depending on whether the extra code happens to do stuff you want or not. They're both written in C++, should you happen to care.

On the subject of soundcard support: the usual quick answer is "it's probably supported unless it's FireWire or made by MOTU". In particular the current M-Audio, Terratec and RME semi-pro gear mostly works fine, as well as most consumer cards. The lack of FireWire audio device support currently is a pain though. Anyway, see the ALSA soundcard matrix for detailed information.

(Disclaimer: I am a Rosegarden developer and, although I track MusE CVS, I have never actually managed to get MusE 0.7 to build because I don't have the right libraries and autotools. So do take all this with a pinch of salt.)

link correction: http://mlipod.sf.net

gaim-vv supports voice and video msn and yahoo protocols. The msn one is based on sip.

I was a trillian user for two years and have now been a Gaim user for a year. The only thing that would change for me is the number of sockets that my computer would maintain.

I wonder if this movement would also spark a movement toward disabling 3rd-party clients. That would NOT be good.

Trillian
Gaim

This prediction by this man will only come to be correct if M$ makes sure their next console can't run linux (I highly doubt that nobody's going to be able to make a simple way to do it again).

Still in development, but keep your eyes on Planeshift a very ambitious and so far very well done online role-playing game. (It's playable now but not complete and a very large download). I hear that a new preview version (with a MacOSX port) is on it's way.

There are a few other games in progress based on CrystalSpace as well (which works on Mac, Linux, Windows, and other platforms as well with varying degrees of support).

Oh, is this where I put the shameless plug for my game too? It isn't? Oh well, here it is anyway:

OpenMortal - geek fighting for all! :)

No site should require compiled PHP code. If you need speed, get a cache extension like APC, Turck MMCache which are free, or Zend Accelerator or ionCube's accelerator if you require a commercial product with support.

If you need even faster code, write a PHP extension (in C, compiled) to handle your resource-intensive functions.

Both of those solutions are much more versatile than a compiler that doesn't support the full language.

Some folks were suspicious of PHP5, and being a longtime PHP programmer, I am very pleased with the changes and additions in PHP5. Can't wait to test it out. Personally, I'm not sure if I'll use *all* of the new stuff, yet I'm sure I'll have to play with the coolest additions for the hell of it, and sort out what I'll be using and what will remain vestigial in my scripts. I will add that some of the previous PHP version quirkiness seems to be fixed.

I am certain this is not the last we'll hear about PHP5 on Slashdot, yet I am only hoping that it's creative/cool stuff, and not security problem/exploit stuff.

I can't wait to see what kinds of changes I can make to my content management system that PHP5 will bring.