Slashdot Mirror

Of course not, this is, and it hasn't been updated in four years. I guess nothing of note has happened in the Linux world sine 2011.

http://linuxgazette.net/186/an...

Yes, there must be a Nigerian conspiracy to keep CapsLock.

You will find the original Borland Pascal 7.01 lurking somewhere on the internet, on some "VINTAGE" something named url. Easily to be found via "google".

Yes, there is freepascal, yes there is delphi, ..

But Pascal really - and only - rocks on a rock bottom DOS-WFW3.11 machine with ISA slots, Creative Labs SB16 , I have one machine prepared and in operation.

And I do code in Pascal, Assembler(tasm) and interface with the real world[1] what do I need a raspberry Y for .. ;)

Ed Nisely was so kind to open source his book [2]

And actually there is really great literature out in the wilde like Buchanon's applied PC interfacing .. and Interfacing Sensors to the IBM PC (it's more like an excelent sensors handbook)

The ISA-Bus has some advantages over GPIO, with bus drivers you can do GPIO, however you can also DMA and PIO based data transfer.

Also to interface ISA-cards(Sound, NE2000) to other projects is mostly fairly simple when starting with port based i/o.

No big devellopment environment just Borland Pascal, TASM and Turbo C++

Be aware of the delay loop error on machines with more than ~200Mhz you need to apply a patch (also found on the internet try searching for xfdisk - which was written in borland pascal)

[1] http://linuxgazette.net/124/du...
[2] http://softsolder.com/2011/10/...

http://forums.linuxmint.com/viewtopic.php?f=47&t=99890

http://fixunix.com/redhat/489285-rhel-5-1-rhel-5-2-dependency-hell.html

http://www.freelists.org/post/oracle-l/Linux-X86-26-Oracle-9204-and-RPM-Dependency-Hell

http://unix.stackexchange.com/questions/53077/firefox-circular-dependency-hell-on-linux-mint-13

http://linuxgazette.net/issue71/tag/3.html

http://lwn.net/Articles/198455/

http://forums.opensuse.org/english/get-technical-help-here/applications/406017-does-dependency-hell-still-exist-2.html
* "The best (worst) way to get dependency hell is to add too many repositories. Eventually one will find the right combination of conflicting repositories to create a situation of dependency hell."
* "The problem is the more repositories that are added, the less and less likely that applications are built against the same set. "
* "Dependency hell ALWAYS exists. ALWAYS (Yes, Debian users, I'm talking to YOU). The question is whether or not if somebody tries to NOT go to dependency hell, is the system going to take them there anyway."

http://unix.stackexchange.com/questions/32594/fedora-16-dependency-hell

http://www.eonlinegratis.com/2013/dependency-hell-trying-to-install-gcc-on-centos4/

Oh, dependency hell is real alright!

getting the software to play nice with such a setup is not currently viable ...

It has been viable for over a decade. Plenty of groups have done it:

http://linuxgazette.net/124/smith.html

And Microsoft is jumping on the bandwagon:

http://www.microsoft.com/windows/multipoint/

See the Wiki for more:
https://en.wikipedia.org/wiki/Multiseat_configuration

Most interesting is Fedora 17 automatically enabling it when appropriate hardware is connected, which should mean RHEL7/CentOS7 will, too.

your $50 tablet is actually a fully-fledged computing device.

Yes, but as a "thick-client" it's a brutally low-end and very limited device. As a thin client, it's a high-end workstation, with huge amounts of memory, unlimited storage, etc.

And if a big market ever developed for thin-clients, you can bet these same tablet manufacturers would come out with even cheaper, stripped-down devices that are only good enough to be used as a thin-client. At what price point would you say it's a good idea? $25? $15? $10?

Fiddly, but on the right track for getting the most out of the 12 PCs. *NIX is your friend here - what you want is what X.org refers to as a "multiseat" system - your only limit is how many discrete graphics cards you can cram into your PC(s). Here's a walkthough of setting one up with six seats. If you can get them, you could also hook up some VT102 or similar dumb terminals to the same box and maybe rig up some UPSs so you can have power available when it's needed as opposed to when it's available.

Typewriters are definitely the way to go for the initial lessons. Manual ones for starters, then maybe electronic ones if they are available for the next stage - just make sure that any ribbons that are needed are cloth so that they can be re-inked though.

Some more thoughts for the next stage - moving them onto actual "computers". You might want to looking into re-purpose one or more of the 12 PCs as a *NIX box using your favourite distro (via LiveCD if need be), then running some dumb terminals off it. With an RS-232 port board and a bunch of Wyse terminals you could easily run a whole classroom from a single decently specified server box, or even some more modern terminals that have Ethernet. Dumb terminals are still quite common in India and other countries I've been too in the region, so I'd imagine they wouldn't be too hard or expensive to obtain across the border in Bangladesh. Ask around too - maybe a local company that has upgraded to PCs might have some in a store room they'd be willing to donate for some favourable press... Alternatively, if you can get hold of some extra graphics adapters, monitors and keyboards, you could take a look at this tutorial that walks you through building a six-headed Linux box - one server, six keyboards & monitors and (more importantly) concurrent users. That's all going to need some power, but a heck of a lot less than an equivalent number of stand-alone PCs.

...that whole "digital magazine" thing has been done already.

It's called the Linux Gazette.

linuxjournal is squatting linuxgazette.com. You may want to try the net domain for the real thing.

I stopped buying linux journal when the articles became too predictable - rehashes of the same articles from the same month the previous year. Boring!

Don't all POSIX compliant OS support hard linking?

I don't think you can regularly hard link directories. See this for example: http://linuxgazette.net/93/tag/2.html

It is possible to DIY (do it for yourself) when creating Live CDs or Live USBs. That way you can use whatever Linux distribution you prefer, or are most comfortable with. As you note, some skills are required, but the tricks are fairly well known. The fine points of Linux startup is somewhat distribution dependent. For my own purposes, I created a rescue environment based on Slackware Linux that boots on a wide variety of hardware "out of the box." I have written a couple of blog posts about it here and here.

One thing that helps a lot is that the Linux device stack is layered appropriately with logical devices before reaching the physical layer of devices. The logical device layers are provided by the device mapper facility. Device mapper provides the toolset that works with a myriad of possibilities (non-encrypted simple snapshots, encrypted snapshots, encrypted physical devices, etc.). These modes can be mixed and matched, and the possibility of using loopback devices further complicates possible designs. In fact, there are so many "correct solutions," that it can be tricky to select which solution most directly meets the users needs. The basic technique is described here.

By "distinct", I meant in a sense that application can tell from which device a keypress, click, or movement event came. For example, in a shooting range game, did player 1 click her primary mouse button to fire her weapon, or did player 2 click his? DirectInput on Windows combines all connected keyboards' keypresses into one virtual keyboard device and all connected mice's movements and button presses into one virtual mouse device.

On linux, however, you can get distinct inputs from distinct input devices; else things like multi-head wouldn't work.

The latter is not what I disagreed with. Yes, there is no room for the NX bit in the plain 32bit page tables, however that does *not* mean that PAE is an absolute requirement for NX support.

http://linuxgazette.net/107/pramode.html
http://en.wikipedia.org/wiki/NX_bit
http://www.wordiq.com/definition/NX_bit

May I point you to: The Thin Line Between 'Victim' and 'Idiot' and The isolate utility

And also a shoutout to lwn.net from whom I've been stealing much information for my Slashdot submissions and not giving them appropriate credit.

Go with clientless thinclients depending on how large your install base is.

Multiseat linux setups are rather easy to setup. There are even a few commercial applications. (My local library has one).

If they don't need DVI, there are some solutions which you just run a single USB cable and it carries sound, USB & VGA. If your work environment is setup so people are relatively close together, give each desk cluster a quad or octocore CPU and everyone their own X session.

I have a friend with a few kids (3-4) and the oldests are now starting to both want to use the computer. You can either 1) Buy 2 separate computers 2) Get 1 decent computer (dual core should be plenty for a 5 year old) and just setup 2 monitors side by side.

http://linuxgazette.net/124/smith.html
http://www.oreillynet.com/onlamp/blog/2005/12/multiseat_x_under_linux.html

You should maybe point out the problems with it to the author: Rick Moen. He's an editor over at http://linuxgazette.net/

The essay displays fine for me is all I can say... and I did rather like it as an essay --that's my entire involvement with it.

Thanks for the links, anyway.

Cables. Duh... Search for MultiSeat.

http://netpatia.blogspot.com/2009/06/multiseat-in-ubuntu-904.html
http://linuxgazette.net/124/smith.html

Great if you have kids or a larger family. One decently powered machine can power multiple "computers".

The technology they used was sending VGA & Audio over USB.

tc basically allows you to activate netem (a network emulator in linux). I dont know about now, but when I had used it for a project a year ago, you had to compile your kernel with netem enabled. tc then allowed you to modify your link properties to emulate wan links. Had used this with tcpprobe to analyze the performance of an Inverse Increase Additive Decrease congestion control algorithm that we had written for academic purposes (adapted from http://nms.lcs.mit.edu/papers/binomial-infocom01.pdf) and compare its performance with newreno. Fun stuff. This was a helpful reference: http://linuxgazette.net/135/pfeiffer.html

Ah, there is, of course, iPodLinux, which will function on the Touch in surely no time. I've been occupying myself heavily lately with Linux games - due to the high degree of generativity inherent to OSS, even one game can offer a seemingly endless supply of entertainment. There are, I should mention, flight sims, and here's a good list of CAD programs for Linux: http://linuxgazette.net/issue54/frost.html. Most games with the Quake/Doom/Marathon/Cube etc. engines will also run on Linux, albeit with some small tweaking at times.\ and finally we finish with a quote from David Lynch about video on the iPhone: http://www.youtube.com/watch?v=wKiIroiCvZ0

Here is a HOWTO:

http://linuxgazette.net/151/weiner.html

j.

I still miss the directory history (Ctrl-PageUp/Dn) from 4NT. I see one person does too and came up with solution

Those features may be new to ext3, but not to the real competitors. I see nothing that might grant an edge over JFS or XFS. The real justifications will come from performance tests.

This reminds me of the recent NTFS article here, which actually suggested that since Hans Reiser is in jail and reiser4 is dead, we should consider NTFS. WTF? The ludicrousness of using NTFS as the primary filesystem is further justified in this article by its similar performance to ZFS, but both run in user-space (and are thus horrible in performance), so neither is really an option. What the heck is wrong with JFS and XFS?

Here are some real comparisons: First, Wikipedia's Comparison of file systems gets you started with a nice mapping of features. Second, a benchmarking of filesystems from 2006 which is still quite applicable (though it doesn't yet cover ext4). What we need is a comparison of EXT4 to XFS and JFS (et al), with EXT2/3 in there for reference.

Recall that the biggest reason for using ext3 is that it is supported best of all the filesystems. If all hell breaks loose, even Tomsrtbt (an ancient rescue floppy pre-dating knoppix) can fix it. Ext4 breaks this backwards-compatibility to ext2. Therefore, I see no reason to use it. One might as well use something more stable and proven, especially while we lack numbers suggesting it performs as well or better.

Comparisons between ReiserFS, JFS, XFS, ext2/3: http://linuxgazette.net/102/piszcz.html

Direct link to benchmark images: http://linuxgazette.net/102/misc/piszcz/lg-102-piszcz-images.tar.gz

It's basically a tie between ReiserFS, JFS, and XFS depending on how your disk is being used most often.

Comparisons between ReiserFS, JFS, XFS, ext2/3: http://linuxgazette.net/102/piszcz.html

Direct link to benchmark images: http://linuxgazette.net/102/misc/piszcz/lg-102-piszcz-images.tar.gz

It's basically a tie between ReiserFS, JFS, and XFS depending on how your disk is being used most often.

Yea, look at linux...No way would it be possible to reset the root password if you had physical access to the machine.

I can't believe all the people who are freaking out about this. This isn't a remote exploit. This isn't a massive security hole. This is trivial stuff that anyone who is reasonably computer savvy should be able to do.

That's all good. Now, try doing that, but without losing all the evidence currently stored in RAM (a ramdisk, a buffer not yet sync'ed to the disk, cache, application data, etc.). The Microsoft device lets you do that : It also eliminates the need to seize a computer itself, which typically involves disconnecting from a network, turning off the power and potentially losing data. Instead, the investigator can scan for evidence on site.

Yes, this is a massive security hole. The investigator only needs physical access because he wants to plug in the USB drive (probably because of procedure and need to do that to get admisible evidence). What is on the USB drive is software though, and software can be made remotely. Evil computer haxxors don't care about procedure, and they don't need physical access to use that hole.

Yea, look at linux...No way would it be possible to reset the root password if you had physical access to the machine.

I can't believe all the people who are freaking out about this. This isn't a remote exploit. This isn't a massive security hole. This is trivial stuff that anyone who is reasonably computer savvy should be able to do.

"not enough multiplayer games are designed for use with SDTV or EDTV output and four USB game controllers. Developers of multiplayer games for PC seem to be of the general mindset that each player has his own computer and his own copy of the game."

That is because you can build a multi-head system, with each seat getting its own keyboard, mouse, display, and x11 session. You can have 4 people playing Quake Wars on one linux box. Or, if you want, you can have 2 seats playing quake wars, while the other 2 play Unreal Tournament 3. Try that on a console.

http://linuxgazette.net/124/smith.html

Dude, linux can be set up to use as many keyboards/ mice / screens as you have. And you can get a usb/ game controller adapters for all your favorite console controllers.

The only way to get six players on one pc playing Quake Wars each with their own keyboard / mouse / screen is with linux.

http://linuxgazette.net/124/smith.html

4 people can play games on the same PC.

Just not with Windows.

http://linuxgazette.net/124/smith.html

split screen is for lusers. I have 2 lcds, 2 keyboards, and 2 mice attached to my box. As such, I can have 2 people logged in at once, both running quake 3. It is nice using a console to usb controller adapter. Then each player can choose to use whatever controller they want. Try plugging 6 playstation controllers into your xbox and playing with 5 friends. Then you can talk about something you simply cannot do in the drool-proof console world.

http://linuxgazette.net/124/smith.html

I have argued against having TWO desktop environments in Linux since 2000: http://linuxgazette.net/issue56/giraldo.html

I totally agree with you on this issue. I want a single GUI API to code for that will work in every Linux out there, I want a single installer that will work in every Linux out there and I want standard system libraries that don't break between upgrades and the very common DLL hell. I want to upgrade mi Linux distro without having to wait forever to compile, or without having to reconfigure everything that I had changed manually again or having to install ten thousand dependencies for a 5kb thingie.

What we have instead is ugly ./configure scripts.

The argument about 'but we have choice' is at first sight, logical, but when tested at large it is a nightmare. End users don't want choice, they want 'it just works' together with 'it is beautiful', because having to make choices feels like work. For people who love to thinker with their devices that's good, but most people is not like that at all.

Having things been different the 'Linux desktop year' would have been around 2002. Now it is too late to change things. OSX has also made me lose interest in Linux.

Windows XP is different, I like it because to me every piece of hardware and software works, including my favorite programmer editor, cell phone software, video editor, video import card, games, hardware for games like steering wheels, etc. No badmouthing Windows from me.

How about one PC with multiple graphic cards, kbd and mouse all playing along at the same time. Wait, that's another old thing, just look at this six headed six usersolution they created with linux.

Wow, this like what my 1993 SparcStation 10 can do. You have a single CPU unit with multiple framebuffers, thus allowing for up to four sets of monitors, keyboards, and mice.

You mean, something like a six-headed, six-user linux system?

No it hasn't. That was a sensible idea and not too difficult to set up when using Linux - see here for details and an example of a six-headed configuration. What Microsoft is proposing is sharing a single screen which is a fucking stupid idea, because you get left with very little screen real-estate to work with unless you have a nice widescreen monitor and if you can afford that you'll probably be better off just getting two cheap monitors and a cheap graphics card.

Sure, you have virtual terminals on Linux but that only lets one user use the console at a time. More on par with this is the Six-headed configuration. Add virtualization and I don't see why something very similar can't be archived. However, splitting of the display seems absolutely stupid. Even with a wide screens, the viewing configuration wouldn't be ergonomic so a multi-headed configuration with multiple console inputs seems smarter and more economical (since you don't have to start widening cubicles and paying for neck and back strain injuries) to me. The only reason to do something like what they're trying to do is simply to figure how to do it and then hopefully apply that knowledge to something that's actually useful.

> Linux has no way in the standard kernel (when I last looked at this)
> to differentiate input devices so all mice and keyboards end up
> globbed as one input device.

You must have looked at it a long time ago. No patching is needed. It is just a simple matter of configuring Xorg to use different input/video devices for different screens. I am talking about X here. Regarding linux console (text one) you are right.

http://linuxgazette.net/124/smith.html

http://linuxgazette.net/124/smith.html

Not all good news though, seems it was a bit unstable. Still, it's a start. :)

http://linuxgazette.net/124/smith.html
http://www.google.com/search?q=multiseat+x
http://cs.senecac.on.ca/~ctyler/ruby/
http://blog.chris.tylers.info/index.php?/archives/ 14-Multiseat-X-Under-X11R6.97.0.html
http://www.linuxtoday.com/infrastructure/200512070 0126OSDTHL

"Protected processes" are a reasonable idea. They're certainly better than putting video and audio processing in the kernel as part of the DRM system. But apparently Microsoft botched the implementation.

Microsoft has for some years allowed processes to do too much to other processes. Things like "injecting" a DLL or thread into a running process from the outside, or "hooking" system calls, are inherently security problems. In the Windows world, normal processes can do that to each other. This tends to be overdone, with too much "hooking" of system calls and such, a tradition from the DOS era. The UNIX/Linux world doesn't have that tradition. Fortunately.

In the Linux world, the things you can't do to a Microsoft "protected process" are roughly equivalent to the functions of the PTRACE call. In SElinux, the mandatory security system controls which processes can use PTRACE on which other processes. So SELinux already has "protected processes", but with a better security model.

If we have to have DRM, protected processes aren't a bad idea. But what you want is for them to be compartmented, not privileged. They should be running in a compartment which prevents other processes from attaching to them, but they don't need the privilege of attaching to other processes. So the video decoder can be protected, but doesn't have enough privileges to act as an aimbot for some game. The security system for a game should be able to lock the game processes into a compartment which other processes cannot enter, preventing cheats. Enforce separation, not privilege.

I frigging love TextMate. I love the directory browsing "project" system, I love the fact that it's dead simple to create custom macros/commands (using bash/python/ruby/perl!), and I love its snippets features.

However. It's not cheap, not cross-platform, and the "Emacs-like keybindings" are just bastardized enough to drive me up the wall. So I'm sticking with Emacs for now.

For anyone interested, here are some quick ways to modify the One True Editor to behave a bit more like TextMate:

Directory browsing of projects: Try Emacs Code Browser.
Snippets: Check out msf-abbrev. You'll be able to specify cursor location, fields, etc. similar to TextMate. I've also heard good things about Skeleton Mode).
Macros: Try the Power Macros package.
Quick(er) buffer-switching: The ido.el package works wonders for me. Note: If you're used to running dired from find-file, you'll want to set ido-show-dot-for-dired to t in your .emacs.

/etc is et cetera. And dsw, the predecessor to rm -i, has a more amusing etymology. I've been hacking UNIX since v6. If I needed a source of reliable UNIX history, I would not turn to the Gnome project, and I would not turn to Norway. If you want an authoritative answer, ask Dennis Ritchie. If you want a reliable answer, try an old USENIX hacker, or UNIX historian Peter Salus.

How about hard links?

Granted, it isn't perfect; you'd likely use extra tools to manage it (e.g. http://linuxgazette.net/issue44/oneill.html), you couldn't label across filesystems, and disk space reporting would have a field day.

I wonder which filesystems have this sort of functionality built in? Omitting Reiser4...

three comments to offer:

1) Another option is to reduce the 3 PCs to 1 PC running multi-head multi-user:
ex. http://linuxgazette.net/124/smith.html if you NEED windows then run inside of free vmware server.
If you don't want a DIY solution, there's a commercial version at http://userful.com/

2) I'm guessing your fans blow in since you said the cable hole is at the top. The cable hole should be more than just and inch or so in diameter and probably atleast the size of one of the fan holes.

3) to reduce the fan noise, mount the fans inside the box and run curved(noise baffle) ducting to the holes.

LoB

That is not advertising. That's the EULA. You will certainly recognize it as a virtually identical sibling of the EULA that you find with any commercial software. These EULAs also disclaim everything possible except where prohibited by law, including the warranties of merchantability and fitness for a particular purpose.

This is the advertising:

"...if we truly desire world domination, we've got to get our LSD into the corporate elite's conceptual water supply and alter the beast's consciousness. That means we need to co-opt the media that shape decision-making at the highest corporate levels of the Fortune 500. Personally, all the press interviews and stuff I've done have been aimed towards the one goal of becoming visible enough to those guys that they would come to us wanting to know the open source community's story. This has begun to happen (besides the Forbes interview, I was a background source for the Economist coverage) -- but it's nowhere near finished. It won't be finished until they have all gotten and spread the message, and the superior reliability/quality/cost advantages of open source have become diffused common knowledge among the CEOs, CTOs, and CIOs who read them."

Eric Raymond, August 1998

Go and have a look here if you haven't already. There's some great stuff.

XFS is an awesome filesystem, and has been ranked the overall best in at least two fs benchmarks:- here, and here. Given what I've read here, I'm possibly considering making it my own default fs...at least for some things.

There's also some OpenGL related projects, as well as some kernel work. What this could also mean for them is that even if they do have to sell SUSE clusters, they can still have some individuality in the offering. Sure, anyone can burn xfsprogs to a CD...but SGI can still market themselves as the people who invented the fs, and thus the people who are most intimate with the code, and thus who can possibly most quickly/easily extend it, or fix it if something breaks.

XFS doesn't really use less space than ext3. Try setting up two 512MB filesystem, 512 byte block size -- one xfs, one ext3, then fill it with tiny files. Guess which can hold the most files? Guess which creates the most files faster.

I was going to move my /usr/portage (gentoo) from ext3 to xfs to get some performance improvements, but I was in for a pleasant surprise when xfs actually ran slower and used more space. (This was on Linux 2.6.18)

So please, quit spreading FUD.

Ext3 isn't the best filesystem around, but it certainly isn't crap, and it's *very* stable, which makes it an excellent choice.

As for resierfs, I have no idea. Every time I've used it, it has crashed on me. And you know what? That makes it a piss poor filesystem.

http://linuxgazette.net/122/TWDT.html#piszcz

It sounds like you want to do something like this. That way, you have a separate root account for each person who needs it, and you just delete the account of the person who left. No one else ever knows or cares that anything changed.

It seems like he had a lot of emacs responsibility at one time. Wouldn't that have to be earned?

JWZ mentions him here.

Linuxgazette mentions his work on HTML extensions for emacs here.

I still run SuSE 9.2. You already know about LinuxQuestions.org. If your ready try these links:

http://www.justlinux.com/
http://www.yolinux.com/
http://www.pclinuxonline.com/

Some online magazines (I suggest you read the past issues):

http://www.linuxmagazine.com/
http://linuxgazette.net/

Enjoy,

Though I've read up on the subject, as far as anecdotal support goes I've only seen a small number of thin client sites in recent years. However, of the thin client sites I've seen, I have to say that the ones running LTSP or a variant, have been really pleasant to use and the ones running (or more correctly trying to run) Windows-based thin clients have been poor to unusable.

A third option to consider would be to buy extra screens, mice and keyboards and go multi-headed aka "multi-seat". It'll more than do for stations where students are just using the web (to include mail) or working on papers, spreadsheets, or presentations.

Yeah it's handy because you don't need to su and remember TWO really strong passwords, but that's the great part of needing to su. It really makes you think about what you're doing before you do it.

So say you're at your buddies house who happens to run Ubuntu. He goes to the bathroom or something... you quickly reboot, boot up in single user mode (very easy to do with GRUB), set a password on the root account, reboot back into multi user mode.