Slashdot Mirror

Shoot, I lied. Forget about a couple hundred K. If you buy that Java is in any way representative of the level of complexity this would require, you can likely do it in a couple dozen K. Quick Google search turned up a Java VM with a memory footprint of 10k.

Before I found Real Alternative and its necessary companion Media Player Classic I stayed far, far away from anything that used Real Player. I didn't want messages about Brittney Spears in my system tray, I didn't want to click 4 different links to bypass their premium player, and I certainly shied away from the massive load time.

I found out about it only after Click and Clack switched back to Real Player's format after having temporarily using Windows Media Player. Their reasoning was similar to mine; many older folks were having trouble locating the free Real Player. Despite the fact that Tom and Rau were able to make nice with Real Networks, I was never able to. But, thanks to my friend Sean, I shall never have to go through 4 different option menus to disable a message center again.

Besides, the Real Alternative codec seems better able to stream than Real's own player software. I assume the codec is just the "guts" of the player with no fluff...perhaps all of the extra system resources are being used by, oh, the message center checking on the latest dirt about TomKat or something.

If your job encouraged private projects, as Google does, do you think your job would be more fulfilling and productive?

Google doesn't encourage private projects. Google does encourage employees to spend 20% of their time on side projects. The difference seems like semantics, but is actually a big deal. Google owns the copyright and IP to every one of those side projects.

But yes, my job is much more fulfilling because it encourages side projects. I work on WiX (http://wix.sourceforge.net/) in my spare time, and it has definitely impacted my day job in a positive way. It's not a private project, though: the copyright for WiX code is owned by Microsoft, just like any other code I write during my day job.

I've never managed to find a MUD that had any other human players in them. I mean, I'm sure some humans are present in some of them, but divide that number by the number of rooms and it seems to explain why it's such a quiet experience.

mud.arctic.org port 2700

I'm pretty sure the mud I played on and off for the last 8 years had human players or at least some NPCs who were surprisingly fluent in Finnish and enjoyed running a 12 man group after you.

It's a dragonlance-based dikumud with a heavily-modified codebase (10 years worth) and an average of 60-80 people online, a few of whom may not kill you on sight.

There are still muds out there with 200+ people logged on regularly, though they tend to be roleplay-encouraged or enforced.

Also, all MUDS seem to require that you use telnet or some wretched dos box or whatever.

http://tintin.sourceforge.net/

Modern variant of the old tintin codebase, runs on Windows, Mac and Linux.

Great. Now we will not only have trouble making things go where we want them to on the screen, we will have to guess if the user is running in pixel doubled mode.

This is absolute bullsh*t. HTML should have gone to SVG a LONG LONG time ago. Working with actual pixel numbers is stupid for both web development and desktop development - I was amazed that MS decided to keep pixels when they unified the forms for desktop and web into one entity.

If we were working on a 1x1 canvas, we could even have anti-aliasing on our desktops.

Check out http://sourceforge.net/buzz-like for an example of an application that is both independent of resolution and allows the user to scale the interface.

"Question: which process would replicate sessions?"

There are expensive solution to do just that. Many L4-7 switches on the market does this. Of course, the whole point of the article is to home-craft the solution, but I think, SQL Relay does what application layer loadbalancer would but without the replicate session feature and with some what failover/loadbalance feature.

Of course, there is always Oracle... Believe me, I think, Postgresql is great open source product, but if I decide to go notch upward on db from MySQL, I rather go all the way to Oracle or DB2 just because of this very issue. Clustering in Oracle is a complete product, and does not require hacks and work around as much as MySQL and Postgresql.

if you try to interpolate you have serious problems when images are supposed to have hard lines

If you're concerned about destroying pixeled edges, run Scale2x, a non-linear interpolator, on the images before bilinear resampling them.

or there are a large number of images on the screen

Modern 3D video cards perform full-screen bilinear resampling with rotation and multitexturing 60 times a second.

those tiny icons used all over moden operating systems

This is a defect in your operating system's desktop environment for supporting non-scalable icons. Mac OS X uses 128x128 pixel icons that are scaled as needed, and many Free desktop environments can use scaled-down or even SVG icons.

Oh, and i upped the DPI, which makes a bunch of programs misbehave and look all gross because they expect a certain standard DPI.

If programs require 96dpi, then that's a defect. Don't use those programs, and make sure to report the defect to the publisher.

(But they cheated: the font definitions had extra information for hinting and scaffolding.)

So does TrueType, even if Microsoft's TrueType renderer is defective and most TrueType fonts in the wild have defective hints. But unfortunately, free software distributed in the United States cannot use the hints in TrueType until October 13, 2009 and must substitute its own hints.

(But they cheated: the font definitions had extra information for hinting and scaffolding.)

So does TrueType, even if Microsoft's TrueType renderer is defective and most TrueType fonts in the wild have defective hints. But unfortunately, free software distributed in the United States cannot use the hints in TrueType until October 13, 2009 and must substitute its own hints.

It is completely object-oriented and the event mechanism is even handled by the signal and slots approach, allowing the same programming elegance found in Qt-based software. It allows you to focus on the design and logic of your program in one place and one place only! Quite similar to how Qt hides the details of the underlying window system from the programmer.

See this overview

Note the familar Qt-like syntax in creating a tree widget.

Please check it out!

The first use that came to my mind is mpeg encoding. You could either encode a bunch of mpeg4 videos simultaneously, ...or, if you have a multithreaded encoder, you could encode 1 mpeg4 video really really fast.

Anything which requires lots of CPU cycles should benefit. Their are examples of such apps in the consumer space if you look around (some of them are even freeware/shareware). ...The real trick is finding the (preferably multi-threaded) apps which will work on your clustered OS.

There are a handful of linux distributions which support Open Mosix clustering and come with a variety of apps which work on the cluster. check out http://openmosix.sourceforge.net/instant_openmosix _clusters.html for some examples

• http://www.alice.org/
• http://www.ogre3d.org/
• http://www.yake.org/
• http://www.delta3d.org/
• http://www.panda3d.org/
• http://www.idsoftware.com/business/techdownloads/
• http://irrlicht.sourceforge.net/
• http://www.garagegames.com/products/1

World Foundry GDK
http://sourceforge.net/projects/wf-gdk/

You should try Irrlicht: [ http://irrlicht.sourceforge.net/ ]

It's relativily easy and yet versatile and powerfull

"If you're not interested in building a game from scratch, have you looked into implementing your game as a mod for some already existing game?"

Mod parent up for suggesting a mod.

Seriously.

Unless you have VERY specific requirements for your game, you should be able to get quite far by creating a mod for an existing game. Now, that could either be a close sourced game or an open sourced game, that's entirely up to you.

Since you stated that your game is never going to make any money anyway, going with a commercial, open source game seems viable. This opens up the possibilities even further. Depending on the type of game you had in mind, Quake3 and Descent2 are both mod'able and both have their source code freely available.

If you don't want to use a commercial open source game, you could use one of the many "free" open source game (feel free to insert the obligatory speech and beer comments here), e.g. Vegastrike ( http://vegastrike.sourceforge.net/ ).

All that said, you don't NEED the source if you can keep it 100% in the mod realm - and people have done some amazing things with mods!

The Swedish mathematician who proved a convergence theorem for Fourier series. without him there would be no IPOD. :p

Without Fourier transforms, we would have used time-domain methods for processing digital audio. Shorten, FLAC, Apple Lossless, and most other lossless audio codecs make use of an autoregressive analysis of a block of audio, followed by linear prediction with entropy coding of the residuals. The GSM Full Rate codec (implemented in Toast) and the Speex codec operate in much the same way, except they add pitch analysis (to filter out the periodicity of vowels and instrumental chords) and lossy quantization.

How does this compare to Keepass?

PasswordSafe pretty good, and has several linux ports

I'm an old Drive Image user from back at v3. Back before drive imaging caught on, the only players were Ghost and PQ Drive Image. Drive Image was smaller, DOS, and very scriptable; Ghost had the name recognition and wizards. I stuck with DI up to v2002. It had XP-NTFS support before Ghost did. It have no regrets. Symantec's buy out ruined the line, and they sucked up PQ's knowledge base never to let it see the light of day again. The next imaging suite I'll even try is Acronis. Ghost has grown too big, tries to do too much, and sucks a$$ on top. I'm still getting use out of our inital $400 edu lisence od DI after 5yrs.

After doing disk imaging for 6 yrs, I've begun moving to Unnattend (http://unattended.sourceforge.net/ ), and scripted or prepackaged app installs. Imaging works most times, but the few where it doesn't are a pain, and usually time sensitive.

I will add that there was a utility shipped in DI4 pro called "DeltaNow" that watched app installs and bundled them into an EXE. I still ue it today. I has saved my job many times. The only apps I have not been able to package with it are obviously MS Office ( killed windows install), Photoshop ( blue screen on startup ), and VirusScan ( blue screen ). Now I script those, or use thier own built in installers to install them unattended.

Even using a striped down XP inage, I've been able to bring up a completely usable workstation from blacnk drive in about 25mins using packaged apps. Much faster than the "magnum" DVD-sized image that had EVERY app preinstalled.

This thread is what I keep reading /. for. I really need to hit this thread with Acrobat for safe keeping so I can reference it later.

Heybiff

AI Minds For Robots As Rightful Persons are being developed as Open Source Artificial Intelligence for installation in PC-based robots in Korea, America, India, Europe and nearby parsecs of planetary space-time.

Mind is an artificial intelligence coded nitially in JavaScript for Web migration and in Forth for robots, evolving towards full civil rights on a par with human beings and towards superintelligence beyond any human IQ.

Mind.html in JavaScript has an installed user base of dozens of intelligent entities cached away on hard disks all over the world, with Update and News links for obtaining state-of-the-art AI software as the prior art blocking all future AI patent applications.

AGI Radar is an advisory "radar screen" of Artificial General Intelligence (AGI) projects advancing ineluctably towards a singularity and a cybernetic economy based on robotic persons outfitted with artificial intelligence.

Technological Singularity is now in a countdown for a Joint Stewardship of Earth between homo sapiens and intelligent robots who own the rights to their own AI software as their own intellectual property (IP).

Which is more than reasonable on the part of the service provider (my hat goes off to you for having the courage to stand firm on this point). Without true anonymity, free speech is an illusion. Of course, FUD will see to it that we'll never have anything remotely resembling anonymity - mention Freenet on Slashdot and watch the avalanche of "I'd never support Freenet - it can be used to spread CP!" Apple will win.

The key feature for me with Unattended is the ability to integrate drivers into the I386 folder so they are available during install. This is what enables such a wide variety of chipsets to be supported from a single installation point.

Then check out: http://driverpacks.net/Projects/DriverPacks/ (if you are using winxp).

It integrates (read the manual for howto) into a winxp cd install and a sysprep install. (needs DVD CD because all drivers together are more then one gigabytes).

Just to add to your list of resources, a few MSI's for opensource apps are located at:
  http://msi-repository.sourceforge.net/

NTFS has transparent compression.
File type implementation in shells available on Linux http://wyodesktop.sourceforge.net/mimetypes.html

Didn't IBM just buy this opensource project: http://geronimo.apache.org/ and make it http://sourceforge.net/powerbar/websphere/? 1) Take open source software 2) Brand it 3) Profit?

If you use IBM servers you get IBM director for free (you can also buy it, I dont know the price). This is a network management tool with a lot of surveilance and other nice features.
For a few bucks (I got an offer for approx 100) you can get a remote deployment option for IBM Director that lets you install packages, images and whatnot.
Unlike most other options IBM's RDM also supports linux installs.

Of course for a quick fix http://unattended.sourceforge.net/ is a very clean and appropriate solution.

I have used unattended with great success deploying several hundred XP installations. http://unattended.sourceforge.net/. It won't let you slipstream an install disc, but it will let you do complete, brainless unattended installs over network, hence the creative name. It has the added benefit of easy long-term maintenance and updates, which is a win over the install-disc or ghosting method.

And AdZap for squid was doing it before that. With roaming laptops and such, I found it easier to just install AdBlock which was much more effective.

With anti-adblocking code out there (along with javascript malware,) NoScript for FF is also a must. User Agent Switcher is also cool - make your browser look like a search engine such as googlebot... Can lead to interesting results on some sites.

The Inq. Article Right To Create blog Bob Jacobsen's site

And AdZap for squid was doing it before that. With roaming laptops and such, I found it easier to just install AdBlock which was much more effective.

With anti-adblocking code out there (along with javascript malware,) NoScript for FF is also a must. User Agent Switcher is also cool - make your browser look like a search engine such as googlebot... Can lead to interesting results on some sites.

Openomy is a free 1 Gb online file system, mountable under Linux using FUSE.

Besides... cross-site RSS is easy if you really -want- to implement it, either using a PHP/MYSQL push (their site tells you when something changes) or pull (your site polls them periodically) or, ideally, both. You could even make it so that arbitrary web servers can register with other arbitrary web servers for automatic push notification on changes. Maybe a dozen lines of PHP. I've done similar push/pull caching for other websites (not for RSS, but generating RSS based on suitably-structured database records is trivial). It isn't exactly rocket science....

Or you could just use an RSS aggregator like Lilina, but that's probably cheating somehow. :-D

Bottom line is that no, my web site can't do that, but it's because I don't care to have it do that stuff, not because I couldn't get it to do that by midnight tonight if I did. (Posted at 11:44.)

I own an ACER Aspire 1692WMLi. The wireless driver has firmware that is supplied by Intel, and in the license notice they explicitely say that they guarantee it meets government specifications only when you use their firmware (or approved by them, which is essentially the same).

You can read here:http://ipw2200.sourceforge.net/README.ipw2200

If you use your lappi on an airplane and it doesn't meet government specifications, you are in big legal trouble. Assuming you land;).

So ... I am sorry, but you cannot make the user responsible for a third party driver just because FSF feels it ain't right. It has to be a limit. Besides, making an up-to-date driver for ati cards is THEIR job.

I take exception to this.

The manuals in general are poorly organised and extremely verbose. There seems to be no good way for them to be updated. The man pages are notorious for not having good examples.

I consider myself to be somewhat of an expert because people actually pay me big bux to do things. It is extremely frustrating sometimes to get good information.

As an example - I posted a more or less correct road map on how to get sound working in a debian machine. Other than the fact that this is obsure - the woody configuration was borken. That was about 4 years ago as I recall.

The procedure is posted in sourceforge under GRIP support and people can find it here: http://sourceforge.net/tracker/?group_id=3714&atid =203714 ID 653979

While the post is not perfect - it is a start and it at least tells people what they should look for.

Over the last 4 years in the IRC linux help groups I have on several occations encountered people running into the same issues as I ran into in 2002. The manuals have NOT been updated. I have seen on a number of occasions A*holes telling people to RTFM. Like *WHAT* manual am I suppose to read? Why can't the software check for this? Why can't it put out a reasonable error message?

The situation is really bad in the Unix world and while it has gotten better over the years - there is a serious rift between what we need and what we have.

I have experiance on more than 13 operating sytems. Unix is the best by far. However the manuals are close to the worst with only the IBM mainframe manuals claiming the prize. Those are truely horrible - mostly because they are so thick that a person would have to spend a month to read even one of them.

Surely we can clean this mess up. A way to start is to open up the temple and let people actually correct and improve the manuals using a wiki style documentation system. I do know there are efforts in this area. The thing is these efforts need to be welded into the distros people commonly use.

People can disagree with me of course - but the flavour of the posts on this thread clearly indicate we have a major problem.

To be honest, one of the reasons I started Appleseed is because of all of the ads that people are bombarded with on sites like MySpace. The whole experience just seems crass.

Right now, social networking is being approached as if the users involved are merely demographics, potential markets, or advertising recipients. And that's really kind of sad for a technology which has so much sociological, political, and even economic potential for change.

I really honestly think that we won't see real social networking until we have an network of open source websites which all work together using some kind of standard commication protocol. Would the web itself have worked if there had only been six or 7 places to host a website? Where would email be if you had a dozen different proprietary methods for sending and recieving?

Why is social networking any different? MySpace, Friendster, Facebook, as far as I'm concerned, these are all the proof-of-concepts, but they're not the way the future will look.

Social networking, by definition, can not be monolithic and centrally controlled.

Jacl is a fantastic scripting language for Java:
http://tcljava.sourceforge.net/docs/website/index. html

I also enjoy just using Javascript for this kind of thing. See the Rhino project:
http://www.mozilla.org/rhino/

Indeed, and they're right. It really doesn't cost anything to just answer the question (if you're able to). I've been prowling on the Linux and BSD circuits for as long as they have been around (though rarely on IRC) and I have seen this kind of response all too often.

The good news is that there are still forums (fora?) around where good manners still seem to prevail. A good example is Dropline, devoted to Slackware and Gnome-ish issues, where newbies get treated with the same equanimity as the rest of us old fart^W hands.

I have been using LUKS to do this on my Gentoo box. It sees to it that the root partition is completely encrypted; only the boot partition is in the clear. (Gentoo itself has support for encrypted swap.) The instructions are a bit scattershot, but I was able to get it to work. YMMV.

Before doing this, use Darik's Boot and Nuke or something like that to fill the drive with random data; I do not think LUKS takes any steps to randomize the contents of a newly-created partition.

To plug a few projects where I have gotten a lot of help when my knowledge of the subject has been limited. By all means they could have given me a RTFM response but instead they answered my questions throughly and respectfully.

Zope http://www.zope.org/
OpenNMS http://www.opennms.org/index.php/Main_Page
MailScanner http://www.sng.ecs.soton.ac.uk/mailscanner/
MailWatch http://mailwatch.sourceforge.net/doku.php
SipX http://sipx-wiki.calivia.com/index.php/Main_Page

Before Suse for Sparc died:
http://lists.suse.com/archive/suse-sparc/
(Too bad that is gone, it was a nice Sparc based distro)

Unfortunately, I didn't use Nice so far, so I can't comment on speed. However, I can't say Scala code was too slow, except maybe for the compiler :-). Although Scala has no 'nullable' types as Nice, they can be emulated using the Option class (heavily used in the library). For the latter point, indeed, Scala can not inject methods into existing classes, but up to a point, they can be achieved using views.

An interesting comparison between Nice and Scala can be found here although it might be out of date (it considers Scala 1.0, while Scala 2.1 has been released recently).

Or, how about trying SISC?

Nice is another JVM-based language that could do with a mention. It puts more emphasis on programming correctness than Groovy, including features such as pre-conditions and post-conditions and safety from NullPointerExceptions. According to the Computer Language shootout, it compares favourably to Java in terms of speed and efficiency, whilst Groovy somehow manages to be several hundred times slower in places.

There is a really nice Scheme interpreter on the JVM called SISC:

http://sisc.sourceforge.net/

IMNSHO Scheme is so much nicer than Python et al that there's not even a decision to be made here as far as dynamic languages on the JVM.

There's also Scala, which is another statically typed language on the JVM. It is SORT of like Java, and can use Java classes natively, but is about 20 times more pleasant to use than Java (if you're into higher-order functions and pattern matching and that kind of thing... but then, everyone should be into that stuff). Anyway, Scala isn't a scripting language, but depending on what you're planning on doing, using it may persuade you that you didn't actually need a scripting language, you just needed a language that wasn't painful to use like Java is.

Scala:
http://scala.epfl.ch/

Fyi, I have also used Jpype with success, which allows bidirectional operation between Java and CPython.

There is JScheme, and Kawa.

for the ABM crowd cmusphinx might be a better toolkit. It has the open source goodness, is multi-platform (no wine shit) and has several versions in differnt programming languages including a pure java version (sphinx-4).

Command and control is a lot easier to do with voice recognition since the dictionary the engine has to choose from is so much smaller. Having voice recognition engines understand arbitrary words well is still a bit difficult.

Command and control is a lot easier to do with voice recognition since the dictionary the engine has to choose from is so much smaller. Having voice recognition engines understand arbitrary words well is still a bit difficult.

http://cmusphinx.sourceforge.net/html/cmusphinx.ph p

no idea how to use it, but you're such a techie, you'll surely figure out..