Slashdot Mirror

I hope the standard will include some kind of wavelet compression, like Dirac or Apple's Pixlet

With these you get better image quality at the same bitrate. Combine that with more bits and you have an awesome picture.

Lets hope they keep up with the current maths technology. Fourier transforms are just so last century!

I know if I saw somebody buying five 100 Ohm resistors at Radio Shack, I would assume they were building a JTAG.

But don't assume that building a JTAG (IEEE 1149.1) boundary-scan test harness => stealing TV programming. Digital logic and CPU design courses in university computer science often use FPGAs, and many FPGAs expose a 1149.1 port for uploading netlist data and debugging the circuit.

Here's the web site of JTAG Technologies B.V., and here's some free software for communicating with 1149.1 devices.

If you could rebuild everything from the ground up (and had tons of money to throw at it), you'd most likely want to build a system based on a very expensive vendor solution.

Assuming that you can't do that, your best bet would be to go with some sort of parallel filesystem, the likes of Lustre, GFS, Ibrix, GPFS or CxFS. The architectures of these vary, but the basic principle they share is performance scalability based on increasing the number of data paths to the disk. So if you have, say, 100 nodes on a high-speed network, you take 10 of them and attach them to your SAN. The parallel filesystem spans the entire SAN and so requests from the nodes can reach any bit on the SAN from any of the ten paths. If you need more performance, you add more paths: controllers, HBAs and storage nodes. I know GPFS scales linearly in performance based on the number of paths to the data, and I believe the others scale well also.

I haven't hit 50 TB on disk (I have on tape, but your post suggests that tape wouldn't give you the performance you need), but I have set up several 4-8 TB GPFS filesystems that could easily grow to 50 TB if I had the spindles.

Good luck finding a solution; symlink-based solutions on a convnentional *NIX filesystem are a nightmare; I sympathize.

you can still download your email using YahooPops from anywhere

Both the Racer binary (couldn't compile: "ar: q3ddlg.o: File format not recognized") and TORCS (compiled fine) freezed my X completely, and I had to hard-reboot (ctrl-alt-backspace does not kill X on my cheap mobo, the machine enters sleep mode instead). Ah, and it's a GeForce2 (nVidia driver 1.0-5336), kernel 2.6.6.

An other (unrelated) problem I have here is that many OpenGL apps seems to run in software mode (< 5fps, depending on the game). Note that, for instance, Quake3Arena runs with decent (> 50) fps.

The only racing game for linux runned ok for me was carworld, that suffers the same from other Free games: the guts of the game are working, but the game is far from complete, not even playable. They seem to have a tendency to stagnate at this (production/polishing) stage.

An excellent racing sim - still "in development" but with very accurate physics... Live For Speed. It's not OSS, and for Win32 only.

For OSS racing sims, there's Racer, or TORCS.

Realistic racing sims like LFS are a great tool for teaching a new driver, IMHO, because they'll teach respect for control of your vehicle. and that there are consequences to bad driving... unlike some of the "arcade-y" driving games where it's not even possible to get go off the track.

"absolute" positions a box with respect to its containing object; "fixed" positions a box with respect to the viewport. In other words, an object that has its position specified with "absolute" will move when the user scrolls the page, whereas a "fixed" object will always remain in the same spot in the browser window.

For an example, take a look at http://finkcommander.sourceforge.net/. The navigation bar on the right is fixed in browsers that support the element, and absolute otherwise. Load the page up in IE and Mozilla, make the windows small enough that a scrollbar appears, then scroll up and down and watch what happens.

there's a transcript of a talk given by Dr. Stephen Tweedie where he says


If you start a write operation to a disk, then even if the power fails in the middle of that sector write, the disk has enough power available, and it can actually steal power from the rotational energy of the spindle; it has enough power to complete the write of the sector that's being written right now. In all cases, the disks make that guarantee.


that impressed me no end.

This is probably a good time to mention EncPHP, a program I wrote that encapsulates a PHP script along with most of what it needs to run inside an EXE. Both console and GUI PHP apps are supported, though for GUI apps, if you choose to embed all of GTK inside the app, it's going to be huge.

It hasn't been updated in ages, but you can check out the latest version here: EncPHP on SourceForge

The sourcecode is included... it uses a free BASIC compiler to dynamically compile a dependance-free EXE with the PHP script, PHP interpreter, and required DLLs, inside an EXE. When the EXE is run, it extracts it all and runs the PHP script with the parameters passed on the command line. There are some simple options you can add to the start of your PHP script to tell it to include other files, and if it is a console or GUI app (to show or hide a console).

First, I must apologize - I've only quickly scanned maybe half of the comments above 2. So, hopefully this has been touched on...

I think the best thing to do is to get people using apps that run on linux. Promote OpenOffice, FireFox, gimp, et al (including GNU utilities for the real wannabe's). When they are accustomed to the apps, the change of OS isn't much worse than a Windows upgrade.

I hear lots of ripping on linux docs, but, let us be honest, Windows help ain't much to brag about. The average end user has very little interest in the details of the OS, and these people shouldn't have to fret about it. Those who are geeks shouldn't have a hard time learning linux. But if people have to learn the quirks of a new system AND new apps, well, that is just a lot to ask.

as newbie after newbie complain that linux needs a "clipy".

I guess you might need to teach them vi first. Fortunately, newbies are attracted to it's elegant simplicity!

Or you can pass them through this Shakspeare translator

You don't need to install realplayer if you use windows. Media player classic with real alternative works with this site, I just tried it.

Sites hosting real alternative seem to come and go. This link looks legit though I havn't tried it. I'm really cautious about exe's from the web these days.

Quicktime alternative is also worth getting. On one computer I had to
experiment with the directx settings to get it to play video properly.

I use this because any version of Real player just sucks. I would not pay for a service like this from Real. If it was something that was DivX I would if it was a good price.

It's worth pointing out though, that being user-friendly didn't stop Windows from actually being faster than X in a lot of tasks. E.g., repainting a Window works orders of magnitude faster under Windows, while in XFree86 you end up needing such silly tricks as processing only each n'th repaint when the user is resizing a window. Doubly so when the Linux equivalent reinvents the bloated wheel, e.g., by insisting to do its very own font rendering and themed widgets.

E.g., MS Visual C still optimizes a LOT better than GCC.

I know it will sound like blasphemy to a lot of the /. crowd, but MS really isn't a company of idiots who are just drooling over the prospect of coloured buttons. It's what you get when you cross (in more than one way;) a whole lot of hackers, with a whole lot of hard working usability experts.

Most of Microsofts's faults, such as never thinking twice about ignoring the standards if it can optimize better without them, or inventing its own formats, are the exact same things we admire in the archetipal idea of a hacker. (The one illustrated in the Jargon file, for example.)

And indeed it has committed more sins in the name of speed, than for all other reasons combined. (Anti-competitive behaviour included.) E.g., that's the reason why MSVC++ was always slightly deviating from the ANSI standard: they could optimize code better that way. E.g., that's the reason it let drivers run in kernel mode, and made Windows inherently unstable. E.g., deliberately pissing off Sun aside, all the changes they did to their implementation of Java were precisely aimed at making it very very fast. Etc.

So either way, what I'm trying to say is: "user-friendly" doesn't _have_ to mean "slower than a snail". Windows has managed to stay pretty fast (fast enough to play real time 3D games, for example) even while cattering to the newbies. I'm sure Linux will, too.

Now stability, that's another thing. No idea there, and indeed MS doesn't exactly come to mind as a good example there ;)

Plus, as was already said, it's not like anyone will stop you from running another desktop environment, if the newbie-inspired one gets too user-friendly for your taste. E.g., most distros ship with KDE, which is aimed at precisely that: looking like Windows to newbies, yet I happily run XFce 4 instead. A couple of co-workers run Ratpoison, and that's as far from Windows (or user friendly) as you can get in a graphics mode.

Not that I'd like to slap the world again but 10 years ago RISC OS users were using both Spatial and non-spatial windows, it's fantastic really using a 3 button mouse left mouse opened in a new window, right mouse button opened it in the same window, flexibility, controll and i've never navigated as fast since.

ROX Filer Implements this system but with double middle mouse click to open in a new window, Though I still maintain it's not as good as the classical RISC OS filer since it lacks the ability to right-click on the close icon to open the previous window and instead requires a toolbar to facilitate this (issues with X11 mostly), it's still a lot better than most file browsers for Linux :)

Crazy Terrorist ;)

Honestly though, if and when i've done video chat, it's been via Yahoo. Maybe we could somehow bribe gaim into making an iChat compatible version ;)

So, people in fact love when the machine works in a way resembling behaviour of real-life objects, but it seems that only when the "spatial" application is a web browser: they accept the book metaphor with web pages, but reject the drawer metaphor with folders and files. Sometimes they even abuse the physical metaphor of tabbed browsing by opening multiple pages - not subpages of the same web site! - in multiple tabs of a browser window. I even know few people who never open more than one browser window, viewing all pages in tabs; I hope they do not try to glue a daily set of newspapers together before reading them...

That's a tad unnecessary. Jabber is an open, extensible protocol which is relatively easy to handle. The standard exists to dissuade anyone from developing yet another IM protocol. From what I've read, the ideal is that Jabber proxies will be set up to translate AIMJabber, ICQJabber, and so on, so that Jabber only users can talk to everyone, and everyone can talk to Jabber users. Gaim, Trillian, and many other clients, on many platforms already support Jabber. I seriously thought everyone on Slashdot knew this sort of stuff.

That's a tad unnecessary. Jabber is an open, extensible protocol which is relatively easy to handle. The standard exists to dissuade anyone from developing yet another IM protocol. From what I've read, the ideal is that Jabber proxies will be set up to translate AIMJabber, ICQJabber, and so on, so that Jabber only users can talk to everyone, and everyone can talk to Jabber users. Gaim, Trillian, and many other clients, on many platforms already support Jabber. I seriously thought everyone on Slashdot knew this sort of stuff.

"I don't know how you missed that. Every single review of Gnome has trashed the whole thing for this one annoyance that many people have been begging for. The level of Gnome FUD has reached a climax in this thread and it disgusts me."

Conspiracy theory would say it's either KDE payback, or MS knocking down a competiter.

Anyway as I pointed out (lost in the noise). The Gnome people have made the change in CVS (arrogant people don't do that), and there is a way to make the change without a key. But then I bet some will say either "this is the way it should have always been. Those stupid GNOME developers", or "why should I have to dowload a piece of software to fix a broken UI? (sung to the tune of I can't get no satisfaction)

If AIM services begin to be charged, they will alienate their obscenely large user base, and fleeing users will start using ICQ/MSN/Yahoo/etc. instead.

This also might give the guys at Gaim-vv a boost via incentives and volunteer manpower.

"I *hated* the folder diarrhea that began with Mac OS. Some people love it. The option to turn it off and on should be an easily configured checkbox in the app, not something "hidden" in the gconf setup."

Then I suggest you either pull what's out of CVS (The GNOME developers do listen), or download gTweakUI and give the rest of us some peace and quiet.

So if only people migrating from Gnome 2.4 and below, KDE, Windows, and MacOS X (that is, a lot of people) would want to change an option, it's not really that important, so you should put a checkbox in a separate program that looks like regedit?

I'm happy you've found something you like, but it seems to me that this is an important sticking point for many users, so it deserves a more accessible toggle than digging through options in gconf.

First thing I tried was KDE on RedHat 9. What an abysmal failure that was. I upgraded the machines to 3.2.1 using the kde-redhat rpms available here

The problem we had with that setup was the file browser. It's way too complex for non-knowledgeable linux users. 800 tabs on the left side of the screen to get to different parts of the file system just simply doesn't work. Nobody could get to anything.

So I switched them to a custom compiled version of gnome 2.6 on redhat 9 (again, vendors restrict us to it). It's actually gone quite well. However, the change I've had to make across the board is getting rid of the spatial windows (a pretty easy option to change, and now part of our default user config). We use a very large file structure to get around our assets and shots, and navigating it with a spatial browser would have taken a ton of windows and the user would have spent way too much time closing windows. So, their browser window has actually been quite sucessful.

In short, the gnome browser view is a winner, but spatial navigation just doesn't work for very large directory structures.

and if you are interested in a *nix filemanager with a like capability you might try out evidence

most SACDs have a regular cd audio layer

Many titles do not have a Red Book layer.

besides, WHAT POINT IS SURROUND SOUND IN HEADPHONES?

It's for those SACD titles that come with neither a Red Book layer nor a stereo mix. Apply a variant of the earwax algorithm that Sox uses to move the perceived sound stage of stereo recordings in front of the listener, and surround encoded recordings will still sound OK.

freewrl might be though!!
freewrl

freewrl has been funded by the canadian government for a few years now

Things you shouldn't do with a paper notebook, or things that the paper notebook CAN'T do:

Store sensitive information

Backup data

Remind you of appointments or tasks (x days in advance, every year, every month, etc.)

Function as an alarm clock

Function as an MP3 player

Perform unit conversions

Search through entire contents instantly

Automate expense reports

Record voice memos

etc

One-letter class names? Is he nuts? That guy never had to maintain code I guess...

Pixel works for MandrakeSoft and responsible for many important modules, for example installation. IMHO it's one of the best Linux installs.

Anyway I agree with you and I'm sure Pixel too. He's pretty busy with many things and it seems this comparison was written "quickly" and mostly for himself to help designing his own language, merd in his free time.

Nevertheless the page is quite interesting, it would be nice to see more on this topic, especially from those people who are actually working on this area, not just as a free time hobby.

One-letter class names? Is he nuts? That guy never had to maintain code I guess...

Pixel works for MandrakeSoft and responsible for many important modules, for example installation. IMHO it's one of the best Linux installs.

Anyway I agree with you and I'm sure Pixel too. He's pretty busy with many things and it seems this comparison was written "quickly" and mostly for himself to help designing his own language, merd in his free time.

Nevertheless the page is quite interesting, it would be nice to see more on this topic, especially from those people who are actually working on this area, not just as a free time hobby.

There are projects that write more extensive general-purpose libraries than the one that comes with ocaml, for example OCaml ExtLib. I think the core OCaml group is plagued by an inclination towards using OCaml for research and academic purposes like proof tools etc, which takes away from the effort of writing large libraries for real-world use.

Last I looked at it, Io wasn't quite ready for the prime time. It looks like it has moved forward some since then, but in the way of being useful, libraries and the like, it still lags behind a lot of the medium- and big-languages.

Beyond Io, what would be the scripting language with a primarily prototype-based OOP system with the most usage and libraries? I know of a handful of similar languages, but not sure which one would make the most sense from the standpoint most potential scripters are coming from.

My favorite is Dialect. It is OSS/FS, but AFAIK, still only on Windows and WinCE. I use it mostly on WinCE, and it is great to use there. Completely source portable between desktop Windows OSes and PocketPC/WinCE. Compiles to an EXE, with everything you need in one file. Unlike some proto-OOP languages, you don't just have prototypes- you have both the structure of a class-based system and the flexibility and power of prototypes, cloning, changing/adding methods or data members on individual objects, etc etc.

Dialect kind of bills itself as a BASIC-like RAD language, but *do not* be scared away by that. It is a lot more like Python than Dialect in its syntax, and I like it a lot more than Python and especially VB myself. It has some really neet functions- one that comes to mind is like tie() in Perl, but even more powerful!- and is incredibly useful. Best way for writing GUI apps on WinCE hands down, and you can write, test, compile, run and deploy as EXE *everything* on the device itself, no need to use a desktop to write your app.

A shame no one has ported it to Linux, etc yet- the source is there. Having spoken with the devs some, it sounds like the Win32 GUI, ADO/ActiveX and other stuff will be bound to Windows, but the rest shouldn't be too hard to port. Would make a great VB-like language for Linux, though now with RealBasic, we finally have that.

Another language that is *really* overlooked is REBOL. It is often overlooked by the OSS community because it is closed source. A darn shame if you ask me, but as a pragmatic programmer (not one driven solely by ideology) I use REBOL and enjoy doing so. At first, it looked kind of gimmicky, a "network" scripting language. But after using it more, I am sometimes caught singing its praises. It is very poweful but the built-in libraries provide a lot of baseline power to programmers and even users. You can write web apps, text console apps or GUI apps- using its own cross-platform GUI toolkit. I have written apps on Mac OS 9 to distribute them to Linux, Windows and OS X and elsewhere to have them run perfectly. I think it is supported on 40-some platforms, though the GUI component- REBOL/View is on fewer, and still lacking on important platforms like WinCE/PPC and Mac OS X.

REBOL rules- it is a lot like Scheme/Lisp, but without parens- in other words, a great language. It has really nice network protocol support, writing an email client can take a handful of lines of script.

But... there is hope! A pretty new and slow going project, R-sharp (but nothing to do with .NET!) was released some time back, the start of an OSS REBOL implementation.

The funniest thing is that the author is probably fully aware of the word's meaning since he uses a fly for his icon.

I can't understand why this person took in Haskell and OCaml anyway* The comparison is being hosted under the pages for merd, a language which aims to be kind of a hybrid of Ruby and ML/Haskell. * I'm not even going to bother wondering about C# and Java. I think they are control cases.

• Netscape Communicator 4 for the web. This is what I used before upgrading my P100. If you use Linux, Dillo is a leightweight web browser (don't know how well it hacks the modern web).
• GAIM for all your IM needs. Supports AOL, Yahoo IM, etc. Available both for Windows and for Linux.
• If using Linux, don't use KDE nor GNOME. XFCE or FVWM are far more snappy and memory-efficient.

Perl is way too slow and it's not designed for quickly scripting web sites.

Clickable link for the lazy:
http://i860.sourceforge.net/gallery/

One more thing I forgot -- you might want to grab what you can from Massinova: Reborn with Streamripper. Never know when someone will take it down again due to bandwidth costs, RIAA threats, etc.

• FM sound quality with wide AM reach with digital reception quality
• Digital tuning support station text
• Supports multiple channels with digital audio/text/other services
• Can use existing transmitters

Useful links:

• Digital Radio Modiale organisation's homepage
• Open Source DRM software
• DRMRX - Commercial DRM software
• FRARS (where both myself and MOEYT are members)
• BBC: Research Department (Doing work on testing DRM)
• BBC Research Department papers (includes DRM papers
• MOEYT/Paul Marsh's homepage (lots of info on satcom and DRM info soon...)
• Introduction to the GNU software radio project with various links

I suggest you watch this presentation on the Psyco just-in-time compiler for Python and do some research on the Transmeta Crusoe processor to learn about run-time optimization.

Any limitations on name length are (I suspect) a "feature" of the ID3 spec rather than iRiver's fault, but I have no evidence for that - as I say, I haven't installed the index yet.

I actually did a very small bit of coding on the Rockbox open source Archos firmware replacement, and as it happens, it was on the ID3 tags, so I know a little about this.

ID3 version 1 is limited to thirty (30) characters per tag field (Artists, Album, etc.), but since the iRiver is limited to -- what is it -- 56 characters, this doesn't seem to be the same limit. ID3 v1, also, I think specifies a total size of 120 characters or something, so why not just set aside the 120 needed?

ID3 version 2 tag filed length isn't limited.

(Although Rockbox last I worked on it, was limited to ~300 characters (300 less null terminators) over all tags. I was careful to make sure that reading more than 300 characters was handled by (silently) truncating -- Rockbox doesn't use any dynamic memory allocation, so static structures and fixed sized were all I had.).

Incidentally, MS Windows users looking for a really good and full-featured tagging program (with automatic abbreviation if you want it, various other forms of smart tagging, and regular expressions for converting tags), should look no farther than the free and open source mp3bookhelper.

Actually, they've been using thin Citrix clients, and this is just a small test project to see if they can use PXES Universal Linux Thin Client instead. The users will still see a Windows desktop. One of the reasons they're doing this, is to be able to use really old hardware, like 90MHz Pentium PCs. I guess not all distributions are getting fatter and slower

As probably many of you will already know, there are legal ways to listen to quality, Free music (with no ads, of course). This Free (as in GPL), cross-platform program is a great example, but I'm sure there are others. Eventually we will drive RIIA (here in Spain: SGAE) off, and change IP laws, but we will start off a better position if "propietary music" is no longer perceived as sinonymous with "music", something we are already achieving with software.

In the meantime, don't be surprised by nasty tricks like this one.

I like nothing about the new window managers. The KDE/Gnome programs should run without them. I am trying to get my window manager flwm to work, but having no luck. It is too hard to change the window manager to your own and getting it to work with the new wm hints is difficult.

One must wonder why they chose a pen. I don't think I've ever owned a pen that picked stuff up and let me carry it across the room... I usually use my hands for that (pick and drop gloves ala Minority Report?)

The problem is, of course, how would this work with a desktop machine? Or something that isn't a tablet device?

I could be optimistic and say that one day in the future we'll all be walking around with regular ball-point (or some other fancy new kind of ink-delivery mechanism) pens with some flash memory and WiFi/Bluetooth/whateverthehellwe'reusing in them, allowing us to walk up to train station timetables and snip off the bits we need, or go to information kiosks and pick up the location of the computer store...

...but then it seems like even something supposedly as simple as drag & drop still isn't even consistent. Maybe instead of inventing new metaphors, we could concentrate on making the best use out of the ones we already have. If you want to take a look at something that uses drag & drop effectively, take a look at Rox. It's editor allows you to both open and save files by dragging and dropping them. Cute, if not rather isolated...

<a href="http://xu4.sourceforge.net/">remake</a&g t ;

Yes, serializing and caching to filesystem are easy (you can even use PHP serialization format with Ruby, although the built in Marshal does just as well), but it's still not as fast as simply keeping an object alive across requests, especially if they grow quite large. It's similar to the reason you don't want to recompile a script each request; loading a serialized cache involves IO, parsing and (simple) interpretation too.

Actually, turok-mmcache provides a way to do this using shared memory; it just says "$value", so presumably that means it'll work with arbitary PHP values.

We use Apache 2, btw; mod_fastcgi is stable there, and I believe thread-safe, so you can use a threaded Apache with forked PHP daemons safely. Even if not, it's nice to have the better designed Apache 2 without worries about mod_php's stability on it :)

Yes, serializing and caching to filesystem are easy (you can even use PHP serialization format with Ruby, although the built in Marshal does just as well), but it's still not as fast as simply keeping an object alive across requests, especially if they grow quite large. It's similar to the reason you don't want to recompile a script each request; loading a serialized cache involves IO, parsing and (simple) interpretation too.

Actually, turok-mmcache provides a way to do this using shared memory; it just says "$value", so presumably that means it'll work with arbitary PHP values.

We use Apache 2, btw; mod_fastcgi is stable there, and I believe thread-safe, so you can use a threaded Apache with forked PHP daemons safely. Even if not, it's nice to have the better designed Apache 2 without worries about mod_php's stability on it :)

The most recent version of e16 came out on 30 May 2004. It now uses imlib2 and freetype2 instead of fnlib and imlib.

enlightenment sf page.