Slashdot Mirror

http://wiki.debian.org/systemd
http://en.gentoo-wiki.com/wiki/Systemd

Strange, last I heard was that Debian added it as an *alternative*. Gentoo's had their own initsystem (they switched to openrc right about when I left 'em), but, to be honest, the average Gentoo user could probably boot his PC just by flicking the power randomly to clock the bits into RAM.

Maybe this is more of an issue with machines that have Windows pre-installed but I'm upgrading my motherboard and it has UEFI and the gentoo wiki doesn't make it seem so bad.

http://en.gentoo-wiki.com/wiki/UEFI

Laptops, of course are going to be an issue.

sorry the correct link was : http://en.wikipedia.org/wiki/Comparison_of_disk_encryption_software

I Usually use LUKS with DM-Crypt, but there are other tools more user-friendly that come with gnome.
Last day I discovered a gnome applet that manages crypted volumes written on the fly as you modify the mounted folder, that scale with the size of the content of the volume. (Dm-crypt has a defined volume size that you cannot outgrow, and the chiffered file used to mount the volume always has the maximum size it can reach --If I want a 15gB crypted volume, I get a 15gB file, no matter how empty the volume is.)

I know avoiding TrueCrypt sounds like tinfoil hat paranoia, but if you need te encrypt your data in the first place, maybe this is healthy paranoia.

4) Hosting https sites still require a unique IP for each site. If I, a a hosting provider, have 1000 sites on a server, I'd rather use one IP, than 1000 IPs.

Check out SNI. You have to have a browser that supports it, so IE6 users are hosed, but just about everything else most people are running will work (Chrome, Safari, FF, Opera).

Damn new layout...

I'm running it on Gentoo. It isn't in the official portage, so one has to get it from an overlay, though there are three to chose from. The official enlightenment one, and two others (Vapier's is recognised as 'enlightenment' by layman) with instructions on the wiki.

http://trac.enlightenment.org/e/wiki/Gentoo
http://en.gentoo-wiki.com/wiki/E17

It seems like the enlightenment one tries to keep ebuilds at stable builds, though Vapier's seems to get the latest from SVN, though I've never had problems getting from it.

That hasn't been true since SNI, which is supported in Apache 2.2, Firefox 2.0, IE 7, Opera 8, and Safari 3.2.1. Probably Chrome too, but I'm too lazy to look that up.

Anyway, here's a guide: http://en.gentoo-wiki.com/wiki/Apache2/SSL_and_Name_Based_Virtual_Hosts#Configuring_Name_Based_SSL_Virtual_Hosts

Except those other OS'es aren't running their apps in jails

FreeBSD has a jail, and Gentoo has a similar mechanism built around Linux chroot. But I understand your point: the difference is that the state-law owner of a device is the administrator and has the privilege to manage these jails. On the other hand, TiVo runs Linux, and it enforces restrictions against the owner. More Info...

LInux native "Ventrillo" http://www.mangler.org/ Works great for me, it now can output to ALSA not that PA crap*...

*Yes PA is crap, linux is a multi user environment, and i'd like all programs on my computer to make noise at the same time, that includes things like MPD, or any other service that i run. PA only works in a supported config with a single user. See http://www.pulseaudio.org/wiki/SystemWideInstance and http://www.pulseaudio.org/wiki/WhatIsWrongWithSystemMode and http://en.gentoo-wiki.com/wiki/PulseAudio#PulseAudio_Server

I haven't tried it yet, my music player is mpd as the mpd user, wine until recently did not support it, and the documentation for config is "if you can't figure it out tough shit, it's for distro packagers to figure out."

Some people may still have misconceptions about Gentoo. The negative stereotype has long passed, though. Gentoo is, really, a meta-distribution: a dist that lets you make your distribution based on what you want and need.

You could do what some folks have suggested and get a really ancient dist, and that may be fine .. but it will have all the limitations it had back in the day, and nothing new without a lot of manual compilation and work. (No newer shells, html renderers, etc.) Gentoo just automates the process, and since you're building for x86, you could easily build on another box as the parent suggests. (It's actually not trivial to truly cross-compile a dist between architectures last I checked, but I haven't really done a lot of research. However it is trivial to build for a different architecture which the build machine supports.)

This way you get all the stuff you want anyway, and all the work to do so is streamlined. Building a boot disk should be easy (as long as you can find a disk drive for your current box!). Check the wiki for details on how to do a lot of specialized things.

Mind you, I have to help out my Windows-using friends as well :(

You didn't hear? The best IT friend excuse was formed, it goes by: "I'm sorry man, I never learned Vista so I can't help you. :(". Seriously though it's the best excuse I've found for bailing out of those people-taking-advantage-of-your-juicy-brain situations.

Back to the topic though. I understand your argument about scattered and misleading information, but this is a self solving problem caused by the small magnitude of Linux. As Linux grows, so will this problem shrink. However there are still many detailed and very helpful resources for your distro, if you're using any major distro. My favourite being gentoo wiki. Apart from you have the regular Ubuntu resources, launchpad, ubuntuforums etc. In addition to this you have great independent resources such as linuxquestions.org

I like to use Ubuntu a lot, as you might notice, in my examples. Not because it's "the master distro", not at all actually. But it's unique in a way where "automagic" is a key word. This is what the common user wants, and we will see more ubuntu-like distros coming as time goes. For now I support Ubuntu as much as I support FOSS simply because it's a milestone in FOSS development that has already made it to history, any person whom doesn't recognize that fact is lying to himself. You don't have to promote it, you don't have to use it, you don't even have to like it, but you should never lie about what it is.

If you want to undervolt your CPU, you can already do it using this:

http://phc.athousandnights.de/

The latest patch is for 2.6.26, but it works cleanly for .27 as well, I'm posting with it right now. You probably also want some userspace stuff to automate undervolting at each boot:

http://gentoo-wiki.com/HOWTO_Undervolt_a_Pentium_M_CPU

It's not limited to Pentium M though. I've succesfully undervolted a P-M and a T2300.

The article is about Mandriva linux, don't you guys feel the difference?.. Gentoo users should read http://gentoo-wiki.com/TIP_Speed_up_your_boot_time and http://jolexa.wordpress.com/2008/09/22/gentoo-improve-boot-time/

How is Gentoo these days? Does it have a full install ISO "bootable" CD? Is it easier to install these days...or does one have to compile stuff overnight as it used to be? One thing I liked about it were the beautiful KDE implementations it showcased.

Meh. The Live cds for gentoo are very similar to Ubuntu now. They boot up to a Gnome environment where there is a double clickable icon labeled "Gentoo Linux Installer" on the desktop. The GUI install makes it easier, but in the end it really is just the old procedure. Another words you should be familiar with the gentoo handbook, but you don't have to reference it every 5 seconds like before. I believe the full install cd includes some precompiled binaries for at least gnome/kde. It's been at least a year since I have done an install, so don't quote me on it.

Compiles are still a pain in the butt. I personally renice emerge in make.conf so it doesn't hog the CPU and just hibernate when I shut the computer off. It's kind of like background updates that way. Gentoo still provides binaries for Firefox and OpenOffice.org which is nice. Firefox 3 hasn't been stabilized which is starting to get grating.

Gentoo has some very handy community sites nowadays, even if they are unofficial.
http://gentoo-wiki.com/
http://gentoo-portage.com/

I'll probably be switching linux distros eventually. I don't use gentoo so much because I like compiling things. The key is it's so customizable...which is exactly the reason I don't use Ubuntu. It peeved me how it loved to overwrite my configuration files. I wonder if Arch Linux would be most appropriate for me?

Linux jails YOU!

ozmanjusri's debunking is quite on-target to be honest. I've played WoW under Gentoo, and it's maximum frame rate was about 2/3rd that of windows (natively). In fact, Wine's average framerate before I added 2GiB more RAM was generally better than Windows. (Of course, this last tidbit might be due to the fact that Windows, with 1GiB RAM, was eating up far more than X--but I suspect a better VM, too.) If you plan on playing a game under Wine that is purported to do well, you should probably at least check the Gentoo World of Warcraft Howto and browse the performance tweaks section. Applying the registry tweak really does work, if you're playing an OpenGL game.

Of course, the game very likely has to be an OpenGL game in order to work in the first place. But, in the case of WoW, OpenGL mode isn't slow because it's running under Wine--it's slow because of Blizzard's implementation. (Seriously--try running WoW in OpenGL mode under Windows, you'll lose approximately the same FPS as you would under Wine.) Plus, in a rather odd twist of irony, Tribes' dedicated server (the original Tribes!) runs somewhat better under Wine!

I had nothing but problems with it when I was testing it a couple of months ago. I couldn't get the networking to work in NAT mode, and bridging mode on a laptop ain't always the best idea. Maybe I'll give it another shot.

Getting the networking system to work is a bit of a pain, but I've only had minor difficulties when using the host interface. NAT will work, but you won't be able to ping or access any resources in your own network (which is a bad thing if you have a fileserver at home and wish to access it on a VM). There are, however, a few tutorials that can help you get started with bridging your network for Windows hosts or a variety of Linux hosts.

FreeBSD is the only guest OS I've had difficulties with (even MSDOS will work, but it requires some additions to prevent it from eating up your cycles like crazy--FreeDOS plays nicely, though). I could only ever get the NAT-based networking to work and even then it would freeze whenever IO operations peaked.

Take a look at some of those articles, and you might be able to get networking up and running in VirtualBox! I have to say, for something of a FOSS offering, it's really nice.

I know its insanely non-standard, but I know gentoo has a article on how to map graphics memory as system ram from the kernel. http://gentoo-wiki.com/TIP_Use_memory_on_video_card_as_swap A potentially interesting idea since the stream proccessors could help offload some heavy compiling. Of course, I could be wrong and fully expect my flogging.

I use dm-crypt and cryptsetup at work for medical record stuff, and I have this to say: It hasn't fucked up yet (yes, I still take regular backups). The filesystem (ext3fs) is now on about a year, the system has survived at least one "whoops, that was the wrong power plug", and every time it comes up ready for me to enter the passphrase and go. It's not going to encrypt your boot partition, but a 50MB /boot (or a usb stick) with the appropriate devicemapper+dm-crypt kernel, bootloader and initramfs, and you'll be ready to mount the rest of the drives.

It also has handy flags for automatically mkfs'ing swap and temp partitions with random keys every boot.

My beef with it is that the cryptsetup "reference implementation" seems to have "become good enough" (last update in 2007) and nobody has done any work integrating it with TPM, fingerprint readers, and so on. On the enduser laptop side, someone should probably write an idle-detecting daemon to unmount and unmap the partition. It does look like suspend can work.

Unrelated to cryptsetup itself, there is a performance issue you'd need to be aware of in dm-crypt. The dm-crypt driver will spawn [kcryptd] process(es) at a nice level of -5 (default kernel level), putting its priority over most other user processes. When I first started testing this, I attempted to move a copy of our code tree into the partition and X stopped responding after a few minutes, then exited with a message to the effect that the X server was killing itself because it had decided it had locked up. You can renice the kcryptd process, but I suggest that you benchmark your expected load before setting anyone loose with it. (renice it to 0 while installing/setting up, then setting it back to negative niceness and test your applications) More positive niceness levels will improve interactivity but slow down filesystem access. I've found that on the core2duo test box, "normal desktop user" (web browsing, opening and closing one document-sized file at a time, etc) that default -5 does not cause any noticable glitches.

My disks would be pretty much spinning all the time even though for home usage i'd say I actually hit non-local disks maybe a few times a week at most

man hdparm

Its a fairly standard util included with most Linux distros. Specifically check the -Sx option (where x is an integer > 0), which sets idle-spindown time. Note, there is a warning regarding hdparm use with software/bios raid, and I havent tried using it myself on drives involved in RAID, so YMMV. Changing spindown time shouldnt adversely effect the RAID though, as it would just block IO until the drives spin up. I think the warning is more in regards to setting the DMA, 32 bit IO and other more IO/low level oriented options.

Since you only hit the drives occasionally, this would keep them turned "off" most of the time, so long as you dont keep swap or some active partition on the drives as well.

tm

Indeed, Screen is the way to go for text.

"man screen" should get you on your way.
I've found useful the following website:
http://gentoo-wiki.com/TIP_Using_screen
Near the bottom is a "live" session example.

This one is good too:
http://www.kuro5hin.org/story/2004/3/9/16838/14935

Gentoo Linux has a nice tutorial for that:

http://gentoo-wiki.com/HOWTO_Gentoo_Router_for_2_ISP,_load_balancing,_switch_traffic_if_link_is_down/up

By the way.. Google is your friend!

well, I guess this should work for you http://gentoo-wiki.com/TIP_Dual-Homed_Gentoo_Server

What is running on the box? With GNU/Linux it can range from not all that difficult to quite complex.

Then there's always the option of getting something like a WRT54GL and loading OpenWRT on it and setting that up - which would be even more complex since you're then getting into vlan configurations for the ports and such.

I'll google "Dual Internet connections" for you: http://www.google.com/search?btnG=Google+Search&q=dual+connections+to+the+internet

http://gentoo-wiki.com/Dual_internet_connections
http://community.spiceworks.com/topic/13347

Or just search for "How to build a linux router" otherwise there are some prebuilt linksys/netgear solutions that can handle dual ISP connections.

Not long ago, folks here happened to point me toward loki_compat_libs. Lo! LD_PRELOADing ye olde libstdc++, smpeg and SDL seems to make the game work perfectly, at least for me. And there were many nervestaplings once more. =)

I'm running:
$ uname -a
Linux death 2.6.25-gentoo-r4 #2 SMP Thu May 22 15:42:34 EDT 2008 x86_64 AMD Athlon(tm) 64 X2 Dual Core Processor 4400+ AuthenticAMD GNU/Linux

SMAC and SMACX work fine here if you download the libraries and follow the instructions at http://gentoo-wiki.com/HOWTO_Running_Old_Loki_Games

I run it via a slightly different command than what they give there though
LD_LIBRARY_PATH="/usr/lib/Loki_Compat/" /usr/lib/Loki_Compat/ld-linux.so.2 /usr/local/games/smac/smacx.dynamic

I had a similar problem with my wife being unable to send through Gmail being the problem. I had tried HTB and had some success, but it wasn't like an astonishing difference. My router, by the way, is an PII running Knoppix.

Anyway, I was getting quite desperate for a solution and I kept playing with wonder shaper from the Linux Documentation Project Advanced Routing Howto. It seemed like it helped a lot and I was sure I had the upstream queue issue under control but it wasn't all that I had hoped for.

Then finally I was taking heat from the wife again on the stupid Gmail upload thing and I started poking around for newer info. I came across this Gentoo tutorial that modified the original wonder shaper script. I modified it pretty heavily for my own situation and added some other stuff for port forwarding and kaboom, I got huge results. My downstram quadrupled and my latency dropped off like crazy. You can do shaping at home. It is a tempermental bitch though and I'm still constantly goofing with it.

Anyway, here's the link to the Gentoo script that basically changed my world. Works perfect with a copy of Knoppix 5.0. If I run it on my P2 266 with 384 megs of RAM it shows PPOE eating thirty percent of the CPU when I go over 4mbps downstream so it's doing something and the results are very obvious. It's like day and night. In fact, I had the server overheat last week and I had to get a new CPU. I may go with a beefier machine if I fry this replacement CPU. I had always assumed that a router didn't require much resources but this thing is clearly doing real work and producing real results.

http://gentoo-wiki.com/HOWTO_Packet_Shaping

http://gentoo-wiki.com/HOWTO_Packet_Shaping

As someone mentioned in a previous post, it's much easier to just whitelist priorized services such as ssh, telnet or gaming protocols than wasting too much CPU cycles on detecting obscure P2P protocols with layer 7 filters.

Personally, I use iptables & tc to setup a simple HTB (Hierarchical Token Bucket filter) system with 3 priority levels:

- Interactive: SSH (with Minimize-Delay TOS-Flag), Telnet, Jabber, ...
- Medium: HTTP, IMAP, SMTP, POP3, ...
- Low: All the rest

Shaping the upload speed is my only concern. All 3 classes may use the complete upload bandwidth. The interactive HTB class gets a guaranteed 90% of the bandwidth and a high burst value. The lowest HTB class has a burst of 0 and about 5% guaranteed upload speed.

While this is only primitive setup, it allows lag-free ssh with an unlimited upload in the background.

An in-depth how-to about the Linux Traffic Control system: http://www.tldp.org/HOWTO/Traffic-Control-HOWTO/index.html

A short pragmatic example using HTB & SFQ can be found here: http://gentoo-wiki.com/HOWTO_Packet_Shaping

"WINE doesn't cut it, nice try, but you either have to tweak WINE or used a hacked/cracked version of the commercial software that has been modified to work with WINE due to copyright protection."

Please read how to get Worlds of Warcraft working under WINE, the average person doesn't want to take that many steps to make it work, plus using a hacked/cracked version that might have a trojan inside of it stealing data.

What I am talking about is having a Worlds of Warcraft native Linux version, just insert the CD, install, then play. No messing around with command lines, using malware cracked versions to patch it with, tweaking system files, changing WINE configurations, using WINE addons, and even still having game freezes and lockups even if everything is configured correctly.

You are just too damn stupid to read my entire post, aren't you? Most Linux geeks are like that.

Windows users don't want to do half-a-million steps, and use malware hacked files, and then still have freezes and lockups just to get WINE working properly to run games inside of it. They don't even want to RTFM or RTFA about what they need to do to force WINE to work the way they want it to work or Linux for that matter. They want an OS that works the way they want it to work, and native software programs that don't need crappy things like WINE to work, and native driver support for their third party software that is reliable, and an easy to use and configure OS that doesn't need complex steps to use, it shouldn't force the average user to read books or web pages, or to even think, it should be intuitive enough to work that the user just clicks the mouse a few times and takes no less than a few minutes to get things working. This isn't that hard to figure out. But really stupid people don't understand that, and thus they fail to see why Linux hasn't caught on as well as Windows has, or even Mac OSX is catching on. Windows and Mac OSX don't need too much work to get things to work, unlike Linux.

Use something like this : http://gentoo-wiki.com/TIP_Bootable_Floppy_with_GRUB and keep this floppy somewhere they won't find it.

Your are referring to the time when it was decided that the linux kernel should have an 8K stack rather than a 4K stack, and while this didn't affect any applications, it broke any driver which used kernel stack memory?

> i'm not 100% sure on what you're asking,

There is a big kernel warning that pops up in 2.6.25, saying that SMBFS is obsolete and will be removed by 2.6.27. As someone who has never even heard of CIFS until I saw that, my first reaction was "OMG, those Linux guys broke Samba! I'm so screwed!". I think that there are quite a few other people who will feel the same way when they try to mount a windows share.

> http://www.gentoo-wiki.com/HOWTO_Setup_Samba

I'll copy the link, in case you don't get modded up. It seems that I'll only need to change the filesystem type in fstab, so it isn't a catastrophe.

i'm not 100% sure on what you're asking, but I'm guessing you are trying to change from smbfs to cifs, which isn't a big deal. Go in to the kernel config and select "CIFS", deselect "SMBFS" (you can have both selected, but there is no need), recompile, reboot. more details: http://www.gentoo-wiki.com/HOWTO_Setup_Samba

It would be nice, but it is probably not going to happen anytime soon... A few problems would be that some at KDE are still behind KHTML instead of WEBKIT, let alone be interested in a Gecko version. An other problem is that someone (Dirk Muller - http://wiki.kde.org/tiki-index.php?page=Firefox+KDE+Integration) already did something like that, but at mozilla you need to do a lot of difficulty to get your code checked-in. He probably wasn't interested in spending days, finding someone. They would give him an CSV account...

The easiest thing would probably websites like the following:
http://gentoo-wiki.com/HOWTO_Integrate_Firefox_with_KDE
http://ubuntuforums.org/showthread.php?t=110353

(there are a few more, could find the one I wanted...)

I'm not in front of the machine right now (and it's powered down), but I don't think that's it. I'm pretty sure I tired Ubuntu, at least once, and according to this: http://gentoo-wiki.com/HARDWARE_Epia_CN10000 the C7 is an i686. So does this: http://tinyurl.com/2pmpwy but it seems to indicate that a C3 (the previous VIA processor rev) wasn't an i686.

Phoenix were not involved in the 1.0 specifications.

Phoenix was involved in the 1.0 spec, regardless of what that PDF says. Most of the work for ACPI was done by Intel and Phoenix.

Your assertion is simply not backed up by available evidence. The best I can find is that Phoenix "contributed" to the 1.0 spec, and produced BIOSes that implemented the early specs. There is no evidence I can find that Phoenix was a leading player in the 1.0 spec. Perhaps you could supply some evidence?

Microsoft were not getting "broken DSDT tables" from manufacturers. Microsoft were shipping software that generated broken DSDT tables.

Do you even know what a DSDT table is? It's a compiled table that describes the system. It's written in a custom language called ASL (ACPI Source Language) and is compiled into AML (ACPI Machine Language) by the compiler. Back in 1998 Microsoft was sent broken ASL from the manufacturers that would not compile unless they tweaked the compiler. If they DIDN'T tweak the compiler ACPI features would not work in Windows (just like they don't work in Linux). What the fuck were they supposed to do?

Nowadays, manufacturers just compile the DSDT tables themselves using MS' compiler, instead of Intel's, because it's EASIER. Once again, how is this the fault of Microsoft? The manufacturers certainly CAN write proper ASL source and they CAN compile them with the more stringent Intel compiler. The fact they the choose not to do so is NOT MICROSOFT'S FAULT. It certainly CAN be done, presumably Apple did this for MacOS.

"Microsoft was sent broken ASL from the manufacturers" makes no sense at all. You are suggesting that manufacturers sent ASL tables to Microsoft for them to compile into binaries, which get sent back to the manufacturer. This picture is utterly wrong.

Microsoft and Intel produced compilers that the manufacturers download. Intel's compiler was written correctly, and gave error messages so that the manufacturers could correct the ASL source and recompile. Microsoft's compiler appears to have been designed to only flag faults that were significant to Microsoft's own OS, and I have supplied evidence that this may have been done on purpose, instigated at board level within Microsoft.

Intel's compiler is available Unix and Linux systems. Linux uses the Intel tools (to create fixed DSDTs), as well as its own code for implementing ACPI functionality.

I would imagine that Apple used Intel's compiler as it would probably work for OSX. Its also very apparent that installing OS-X on non-apple approved hardware will result in exactly the same issues as Linux finds, specifically a series of hardware problems as Apple's AppleACPIPlatform driver trips over buggy DSDT tables.

You will have to supply evidence that using Microsoft's compiler is easier, as both a trivial downloads, and I have used iasl on Linux with no hardships. You will also have to supply evidence that manufacturers are moving to Microsoft's compiler, as my feeling (I cant find evidence either way) is that the big players are moving to iasl as they start offering Linux. I know Compaq used the Microsoft ASL compiler, and clearly people stopped buying their kit for some reason.

I guess I'm saying that it's incumbent upon the Linux people to work around the problems in MS' compiler by re-implementing ACPI in Linux, make their own compiler, or convince the manufacturers to write proper ASL.

Linux has been working around the problem, by supplying corrected DSDT tables where appropriate, and blacklisting ACPI on unfixable hardware.

Bitching about MS doesn't accomplish any

I run Linux, but one of the things I'm least happy about is the horrible support for power management. None of the sleep, hibernate, etc., options work on my machine at all.

I once had a problem with this, and decided to investigate.

So I went through the forums and found that the problem was that the manufacturer of the laptop supplies a dsdt table that does not follow the published standards for dsdt tables.

So I found a corrected table for my laptop and suspend/resume now works. But I was interested as to why a manufacturer would supply a DSDT that didnt follow the specs. And heres what I found:

1. The ACPI standard is rather complicated, almost as if it was disigned to be hard to implement. Checking to find who the major players in defining the specification, I find my fist clue: "Conceived by Intel, Microsoft and Toshiba"
2. So why would they create such a complicated specification? My next clue was that Microsoft was the developer of one of two major 3DSDT compilers.
3. It appears that the DSDT compiler Microsoft created is very forgiving of errors that other compilers (such as from Intel) would flag.
4. I don't believe it is coincidence that the parts of the ACPI specification parsed strictly by the Microsoft compiler are those needed by Microsoft operating systems.

So Microsoft create a complicated specification, probably taking care to leave out important implementation details. Then they ship a compiler for the specification that only checks parts of the specification used by their own software. And thats why Linux has issues with suspend/resume on some hardware.

Does any of this sound familiar?

Although I *did* see a feature of the OSX that I liked -- the transparent terminal...

http://gentoo-wiki.com/Xorg_X11_and_Transparency

I'm sure this, with minor modifications, is applicable to your distribution of choice.

Which windows will detect by reading the number of letters in the OS name.

Google your question.
results: http://gentoo-wiki.com/HOWTO_Dual_Monitors
It's called TwinView. Mainly has to do with changing settings in xorg.conf, though you can run nvidia-xconfig and have that modify your xorg.conf to satisfy your needs. FYI, the fx5200 is fully supported.

But if those same guys are Linux OSF zealots then beware if you're closely touching or perhaps violating the GPL or any other open source license they favor. Because then everything is different and you should be made to comply no matter what. Why don't we leave these things as they are as well and only start making noise when someone actually complaints about it for reasons other than "Whaaa, you violated ...

I disagree with you there.
ACPI is a very precise and sensitive issue. So the workding of the news is misleading and yes, no flamfest is needed, but clarification around this issue is important.
If you have been through a few installations of linux on different laptops, you will know what I mean. Just do 3 install parties and you should begin to understand the problems in this area, and you should see that some laptops behave well and some behave strangely, you will start to be wary of certain brands and after a while and some documentation, you will learn that there is a spec, an intel compiler, a microsoft compiler, and that strangely the microsoft one is crappy to the point that some BIOSes are crappy because of it and not the other way around, and that some OSes have difficulties because of that situation.
Here is some doc:
http://gentoo-wiki.com/HOWTO_Fix_Common_ACPI_Problems
Here is some related info from a broken legal system that can't manage its misbehaving trusts anymore:
http://antitrust.slated.org/www.iowaconsumercase.org/011607/3000/PX03020.pdf

Even in terms of support to asus, it might be important. If this is just a temporary problem, then asus might come out as a reliable source of linux compatible laptops with working ACPI. Or not. Right now, there are more and more laptops working for linux out of the box and it's a no brainer to find one, but we have to stay alert so that the situation continues to improve.

using the g00g (see first hit):

http://www.google.com/search?num=50&hl=en&safe=off&q=PSP+JPEG+PDF+RSS&btnG=Search

working w/a linux-based system and the PSP:

http://gentoo-wiki.com/HOWTO_PSP

however, if you install an 'open' firmware on your PSP and use Bookr:

http://sourceforge.net/projects/bookr/

you'll find that you can fit many more books on your memory stick (and the Bookr app is superb)... the method of installing an open firmware on the PSP depends on what PSP and Sony firmware you have... the process basically involves:

1. collecting the firmware files
2. copying files to your PSP's memory stick
3. running an executable to 'downgrade' your PSP's firmware
4. then reflashing your PSP w/a 'hacked' PSP firmware that allows third-party executables, such as Bookr

Have you looked into Jack?

According to gentoo-wiki.com/MAN_hdparm -B is:
Set Advanced Power Management feature, if the drive supports it. A low value means aggressive power management and a high value means better performance. A value of 255 will disable apm on the drive.

I would say blame the hard drive vendor.

http://gentoo-wiki.com/HOWTO_X-forwarding

Read it and adapt to debian. All linux sytems can do this.

I know there is an X server for windows (xming) and cygwin will work as well. I'm sure there is something comparable for the mac as well.

If you're on Gentoo Linux, then you could try parallel startup to alleviate that step N+1 waiting for step N problem.
http://gentoo-wiki.com/TIP_Speed_up_your_boot_time#The_RC_File
I imagine that other distros probably support something similar.

I had DenyHosts set up to cover ssh brute force attempts.

Now I have even better protection: a dead motherboard; Noone will crack this baby /grumble

Atleast in Linux, you can use the memory as swap or a ramdisk.
Gentoo guide