They've had that for a while now. If you look in the Makefile, you'll see that `uname -r`=2.3 gets mapped to the 2.4 defines, so for a while there I was compiling a module for a 2.4 kernel! I went back to the stable because I had some very serious filesystem corruption (like stuff in/lib... ick!), giving me a good excuse to change distros to Debian.
On the subject of sound, I noticed that Debian does not have a save/load of sound settings (volume in particular) like Red Hat and company do. So I wrote my own little script to do that. If you want it, e-mail me, but it's really very simple so you could probably write it yourself. Another Debian quirk -- ahem, feature -- is that the sound module is not loaded automatically like in RH, but if you put it in/etc/modules with all other modules you want loaded at boot time, there it will be. If you have to give the module any parameters, you can do it there or put them in/etc/modutils/$modulename (I have in./ne, 'options ne io=0x200', to match my isapnp setting it to io 0x200, for example).
There's a UDMA patch available for the 2.2 kernels. It's linked to in the UDMA mini-howto (see ldp.org). It's basically the same as what's going into 2.4 (aka, in 2.3 right now). However, I haven't had any luck getting it to work on my ABIT BE6.
Good point... MS hardware does tend to be good. I have a MS Sidewinder joystick and two MS Intellimice. I like that wheel. And of course, check out 'imwheel' for X (Window, not Box... big difference).
btw, it's a "modified Windows". Most likely Win98 w/ some junk removed and a few little things changed.
> I have this sneaky feeling that as part of the settlement deal on the US v. Microsoft case, Microsoft will provide the specifications necessary to run gaming applications written completely in Linux on XBox.
Like we (?) need specs... How different can it be from a standard (whatever that means) x86 machine? Windows, especially Win98 (which IMHO is what the "modified" Windows will be based on), depends on all sorts of subtleties about the x86 platform -- PnP, for example (I know that isn't the best example, but I'm not a architecture lawyer). The only thing that I could possibly imagine being different is the BIOS. (Which one are they going to use anyway?) In that case, modify LILO as necessary, possibly modify isapnp to handle different defaults, and recompile kernel. The only possible problem might be getting it to boot the kernel to start with... but then there's always LOADLIN (they will keep MS-DOS in Windows-XBox). Tada, awesome Linux box. If it wasn't Microsoft, I would actually consider buying one.
If you think that I am wrong, I probably am. Okay, Kenneth, stop being so modest...
Debian's package manager is even more complete in this regard. It provides a complete configuration system in which you can choose the configuration interface, the kinds of configuration questions you get asked, and get asked all questions before the packages start installing so you can leave it alone while it configures.
It also has a well-defined notion of what is a configuration file and how to deal with these when (un)installing or upgrading a package. There is never any question about where a file goes. It hides lots of messiness from the user, and an inexperienced user can easily set up the configuration program to only ask high-priority question. This allows the inexperienced user to be hidden from much of the complexity that is going on, but if needed it is trivial to access this complexity. Now that is what I call a good user interface.
As you can probably tell, I like Debian. Even upgrading from RH6 to Debian 2.2 frozen.
From screenshots on the site, the interface looks a lot like Bochs's VGA-in-X interface. Though in the Release Notes it said something about removing a UI.
Right now it doesn't do much except run kernels that say "Hello World."
Oh man, that ruined a few years of my early life... I remember one of my first non-pointless programming projects was an extensive modification of Nibbles. I figured out how its speed-checking routine worked (it used For...Next loops for delay in the main subroutines to avoid the floating point ops and to get better precision, but timed a similar loop to determine a multiplier for the iterations of the For (excuse me, FOR) loops later on. This was before I took any algebra or even pre-algebra... a good guess is probably age 7 or 8... and I'm sure that understanding the little formula that it used and some other things was the root of my interest in mathematics from that point on). I discovered its key-getting-and-parsing section, and once I had read over it at least 50 times, I understood it and added my own keys. I tweaked some other routines too so that you could turn off walls (i.e., go right through them), start at any level (that was easy), return the snake to the middle of the screen, change how much it increased after hitting each number, and lots of other stuff that I dreaded leaving.
Of course, now that I use Linux, I very quickly found gnibbles (Debian maintainers please put/usr/games in the default path!!!) (the Debian package is gnome-gnibbles, fyi). It's even cooler! Pass-through walls (I never thought of that!), bonuses that make you shorter, totally new levels, up to four snakes (once me and one of my friends played four-player GNibbles, each one doing two snakes... that was fun!), and unmatched speed!!! And more overjoyed was I to find a Gorillas clone for the TI-89 over at ticalc.org! Unfourtunately, it doesn't do the little sun-smiles-when-you-throw-a-banana-at-it thing (screen's too small) that I just loved (hacked it once to do that at random...). Anyone know of a Linux port?
You missed money.bas. I didn't hack that one very much (too boring), but I did see that it had a small machine code section for something like fast screen scrolling. It even had menus. High-tech BASIC coding!
Before leaving BASIC and moving on to more advanced languages, I undertook a very large project, specifically to fix up and enhance my friend's program called "Vocab Quizzer" (for English). I cleaned it up a lot, sped it up a lot, made the code a lot neater, fixed bugs, and (this took up most of the time) added lots of awesome graphics. I found probably the fastest and least memory-intensive way (in BASIC, that is) to draw and maintain a starfield bitmap in the background of everything in the program. I added in a grade-keeping section that was password-protected, and even devised a checksum algorithm to allow grades of home or lab-based quizzes to be verified by the teacher. Way-cool for a 6th grader like me. Okay, so it was pretty pathetic. But that wasn't important. It's what it led to that was important. Scheme, C. And of course I always liked CLIs, so Linux came naturally. Did I mention I discovered Slackware at about that time? Now I'm using Debian, and very very happy with it. Look how far we've come!
And to bring it all full-circle, I can run DOS from Linux. Vocab Quizzer -> QBASIC -> DOSEMU -> Linux -> P3-450.
I was looking around in CompUSA a few weeks ago and saw a LinkSys NIC box that mentioned that TurboLinux was bundled... I didn't get it so I can't say whether it was correct or not, so either they messed up and put the sticker on boxes too early or the announcement is a little late.
I'm posting this from Debian 2.2 frozen, Mozilla M13. I WANT CRYPTO! Okay, okay, not that badly. But this shows that Debian quite surely includes Mozilla. Actually, I had not installed it in my initial install (which, btw, was downloaded over a 38.4 modem), so:
apt-get mozilla
Then wait... and wait... and start X, bring up a terminal, and type 'mozilla'. Simple. Then it crashes. Not that often, fourtunately.
HDTV (err, DTV) is not at all dead. Why? Corporations are spending big bucks to upgrade their transmitters, facilities, etc. for high-bandwidth difital broadcast. They are not going to simply throuw away all of this. The problem was probably that the FCC was too aggressive with its timetable and thought that we could transition a lot quicker than we have been. But it'll come, I guarantee that. There will be some rocky spots with the transmission format (we are going to have to switch over to COFDM), but the US will have a digital broadcast system operating in the near future. Now whether or not the broadcasters choose to carry digital video over this network is a different story, but TV is something that can't be replaced anytime soon. There is the prospect of using only a few megabits of the 19.2 MBit/sec data stream as a standard-definition TV signal (which will still be better than today's signals) and the reast for push services and possibly even net downloading (of course it couldn't be 2-way).
Good. Other people have said that [insert your registar here] has tons of problems if you want to do your own e-mail but they also provide e-mail services.
The list of accredited ICAAN registars is at InterNic.net. This is just to let you know the alternatives, and if some registration company is really allowed to register domains.
So in their present state, you can't really read anything outside the largest circular region on the disk. Darn.
Okay, so I don't really know that much about CD technology (what a shame), but I do know that you can record data in a number of sessions, and that newer drives (MultiRead drives, including most all CD-R[W]s) can read more than 1 session. Anybody know if you could somehow use that unused space by creating some more sessions to fill it in? You would need a lot of sessions to fill in all of it (how many sessions can you have on a CD anyway?) if this were indeed possible. I am assuming that the only reason that you couldn't just structure your iso9660 fs just right to avoid read/write on the clipped-off parts of the disk is because the laser is just going to wander over there anyway (read, Bad Thing).
This is a big problem with lots of binary-only Linux software: Linux runs on many different kinds of processors. Wine constrains you to x86 (at least for emulation; I have never used Wine as a porting library.). On the other hand, that might not be that bad for MS -- "We released a port! (it only runs under Intel processors, so tough luck)"
That means Unix, and once that happens, how hard is it to make a Linux/BSD port?
Really hard. Why? 'cuz Gates has to swallow it.
And remember that OS X has Aqua. I'm willing to bet that MS is going to take very full advantage of Aqua's prettiness to hide some mess. So what if it doesn't work? It looks nice!
And remember that MS likes to take advantage of lots and lots of undocumented features released to them under NDAs (okay, that's just what I observe). Linux probably doesn't have 95% of those "features".
Nope. Didn't work. I booted the kernel off of a disk; it detected the controller fine and even got most of the way through reading my partition table and then suddenly stopped, then after a short time said "Timeout waiting for DMA". I sort of have an idea what that is, but on the contrary I have no idea how to fix it. This is starting to get really annoying. I bought a board for its UDMA, and its UDMA isn't working!!! Is there anyone with Gentus who has gotten it to work on the BE6? If so, can you please send me the diffs between your kernel (source, of course) and the standard 2.2.14 (or is it.14? whatever it is.)!?!? I tried all the different UDMA modes on the HPT366 controller, and some of them work even worse than UDMA Mode 4 (which was what I tried first), but none any better. (btw, the HPT version for the BE6 latest BIOS is 1.21 (or some mix of decimal points among that)). This says nothing about the patch linked to in the parent comment working on the BP6.
Anyway, you probably deduced from the above that I am not happy.
Don't like Red Hat? I'm pretty sure the significant modification to the kernel can be found here.
I hope it works with the HPT366 okay.
An HPT366 driver is built into the 2.3.x kernels (at least 2.3.45, which I have tried and then dumped because of nasty fs corruption). The docs for that say that the HPT366 chipset is in its present state unbootable. I don't believe that 'cuz it'll work on Windows and of course, this here Red Hat clone. Clear this up, anyone?
I'm trying the driver now (okay, still have to patch the kernel and rebuild).
I have an ABIT BE6, not BP6. They both use HPT366. Any differences in the implementation that BE6 users should be aware of besides using the latest BIOS?
The above comment brought to you by the Schemeing mind of Kenneth.
I have seen a few programs (ssh comes to mind) that have a 'make uninstall' option if you download the tarball. Some uninstallers are, of course, better than others. Perhaps a general program that would analyze a Makefile with an 'install' "phony target" and either (a) add a 'uninstall' target to the Makefile, or (b) allow the user to uninstall the program through a separate utility. This progam would have to allow a user to uninstall personal configurations (rc files, directories, etc.) for a specific user or all users, or even all users except those who have specially requested to keep their configuration. It could even be implemented as a wrapper around make and monitor the installation (or at least make backups of any files that make would replace), sort of like CleanSweep SmartSweep for Windows. I am thinking kernel module to monitor a specific process. This is starting to sound a little overkill, but how else do you think it was done in Windows?
Make me feel stupid and say that a program like this already exists. After feeling stupid for a second I will be happy.
Slashdot Mirror
They've had that for a while now. If you look in the Makefile, you'll see that `uname -r`=2.3 gets mapped to the 2.4 defines, so for a while there I was compiling a module for a 2.4 kernel! I went back to the stable because I had some very serious filesystem corruption (like stuff in /lib ... ick!), giving me a good excuse to change distros to Debian.
On the subject of sound, I noticed that Debian does not have a save/load of sound settings (volume in particular) like Red Hat and company do. So I wrote my own little script to do that. If you want it, e-mail me, but it's really very simple so you could probably write it yourself. Another Debian quirk -- ahem, feature -- is that the sound module is not loaded automatically like in RH, but if you put it in /etc/modules with all other modules you want loaded at boot time, there it will be. If you have to give the module any parameters, you can do it there or put them in /etc/modutils/$modulename (I have in ./ne, 'options ne io=0x200', to match my isapnp setting it to io 0x200, for example).
There's a UDMA patch available for the 2.2 kernels. It's linked to in the UDMA mini-howto (see ldp.org). It's basically the same as what's going into 2.4 (aka, in 2.3 right now). However, I haven't had any luck getting it to work on my ABIT BE6.
Last time I checked, netfilter was optional. You don't have to compile it in. This makes transition easier.
HEY! Why does that say Anonymous Coward when I'm logged in?!?!
Stupid Mozilla M14.
Good point... MS hardware does tend to be good. I have a MS Sidewinder joystick and two MS Intellimice. I like that wheel. And of course, check out 'imwheel' for X (Window, not Box... big difference).
btw, it's a "modified Windows". Most likely Win98 w/ some junk removed and a few little things changed.
> I have this sneaky feeling that as part of the settlement deal on the US v. Microsoft case, Microsoft will provide the specifications necessary to run gaming applications written completely in Linux on XBox.
Like we (?) need specs... How different can it be from a standard (whatever that means) x86 machine? Windows, especially Win98 (which IMHO is what the "modified" Windows will be based on), depends on all sorts of subtleties about the x86 platform -- PnP, for example (I know that isn't the best example, but I'm not a architecture lawyer). The only thing that I could possibly imagine being different is the BIOS. (Which one are they going to use anyway?) In that case, modify LILO as necessary, possibly modify isapnp to handle different defaults, and recompile kernel. The only possible problem might be getting it to boot the kernel to start with... but then there's always LOADLIN (they will keep MS-DOS in Windows-XBox). Tada, awesome Linux box. If it wasn't Microsoft, I would actually consider buying one.
If you think that I am wrong, I probably am. Okay, Kenneth, stop being so modest...
Kenneth
It also has a well-defined notion of what is a configuration file and how to deal with these when (un)installing or upgrading a package. There is never any question about where a file goes. It hides lots of messiness from the user, and an inexperienced user can easily set up the configuration program to only ask high-priority question. This allows the inexperienced user to be hidden from much of the complexity that is going on, but if needed it is trivial to access this complexity. Now that is what I call a good user interface.
As you can probably tell, I like Debian. Even upgrading from RH6 to Debian 2.2 frozen.
...
So it can't even run DOS.
Though when it does, dosemu is going to have competition.
From screenshots on the site, the interface looks a lot like Bochs's VGA-in-X interface. Though in the Release Notes it said something about removing a UI.
Right now it doesn't do much except run kernels that say "Hello World."
I seem to remember it being a LinkSys. But maybe you're right.
Oh man, that ruined a few years of my early life... I remember one of my first non-pointless programming projects was an extensive modification of Nibbles. I figured out how its speed-checking routine worked (it used For...Next loops for delay in the main subroutines to avoid the floating point ops and to get better precision, but timed a similar loop to determine a multiplier for the iterations of the For (excuse me, FOR) loops later on. This was before I took any algebra or even pre-algebra ... a good guess is probably age 7 or 8... and I'm sure that understanding the little formula that it used and some other things was the root of my interest in mathematics from that point on). I discovered its key-getting-and-parsing section, and once I had read over it at least 50 times, I understood it and added my own keys. I tweaked some other routines too so that you could turn off walls (i.e., go right through them), start at any level (that was easy), return the snake to the middle of the screen, change how much it increased after hitting each number, and lots of other stuff that I dreaded leaving.
/usr/games in the default path!!!) (the Debian package is gnome-gnibbles, fyi). It's even cooler! Pass-through walls (I never thought of that!), bonuses that make you shorter, totally new levels, up to four snakes (once me and one of my friends played four-player GNibbles, each one doing two snakes... that was fun!), and unmatched speed!!! And more overjoyed was I to find a Gorillas clone for the TI-89 over at ticalc.org! Unfourtunately, it doesn't do the little sun-smiles-when-you-throw-a-banana-at-it thing (screen's too small) that I just loved (hacked it once to do that at random...). Anyone know of a Linux port?
Of course, now that I use Linux, I very quickly found gnibbles (Debian maintainers please put
You missed money.bas. I didn't hack that one very much (too boring), but I did see that it had a small machine code section for something like fast screen scrolling. It even had menus. High-tech BASIC coding!
Before leaving BASIC and moving on to more advanced languages, I undertook a very large project, specifically to fix up and enhance my friend's program called "Vocab Quizzer" (for English). I cleaned it up a lot, sped it up a lot, made the code a lot neater, fixed bugs, and (this took up most of the time) added lots of awesome graphics. I found probably the fastest and least memory-intensive way (in BASIC, that is) to draw and maintain a starfield bitmap in the background of everything in the program. I added in a grade-keeping section that was password-protected, and even devised a checksum algorithm to allow grades of home or lab-based quizzes to be verified by the teacher. Way-cool for a 6th grader like me. Okay, so it was pretty pathetic. But that wasn't important. It's what it led to that was important. Scheme, C. And of course I always liked CLIs, so Linux came naturally. Did I mention I discovered Slackware at about that time? Now I'm using Debian, and very very happy with it. Look how far we've come!
And to bring it all full-circle, I can run DOS from Linux. Vocab Quizzer -> QBASIC -> DOSEMU -> Linux -> P3-450.
I was looking around in CompUSA a few weeks ago and saw a LinkSys NIC box that mentioned that TurboLinux was bundled... I didn't get it so I can't say whether it was correct or not, so either they messed up and put the sticker on boxes too early or the announcement is a little late.
Just a little side note, but Linux itself can sleep to disk. Check out one of the patches linked to on Kernel Notes.
Finally, go grab some filesystem drivers that can read Ext2 ...
Explore2Fs
I'm posting this from Debian 2.2 frozen, Mozilla M13. I WANT CRYPTO! Okay, okay, not that badly. But this shows that Debian quite surely includes Mozilla. Actually, I had not installed it in my initial install (which, btw, was downloaded over a 38.4 modem), so:
apt-get mozilla
Then wait... and wait... and start X, bring up a terminal, and type 'mozilla'. Simple. Then it crashes. Not that often, fourtunately.
HDTV (err, DTV) is not at all dead. Why? Corporations are spending big bucks to upgrade their transmitters, facilities, etc. for high-bandwidth difital broadcast. They are not going to simply throuw away all of this. The problem was probably that the FCC was too aggressive with its timetable and thought that we could transition a lot quicker than we have been. But it'll come, I guarantee that. There will be some rocky spots with the transmission format (we are going to have to switch over to COFDM), but the US will have a digital broadcast system operating in the near future. Now whether or not the broadcasters choose to carry digital video over this network is a different story, but TV is something that can't be replaced anytime soon. There is the prospect of using only a few megabits of the 19.2 MBit/sec data stream as a standard-definition TV signal (which will still be better than today's signals) and the reast for push services and possibly even net downloading (of course it couldn't be 2-way).
Good. Other people have said that [insert your registar here] has tons of problems if you want to do your own e-mail but they also provide e-mail services.
What if you want to do your own email?
What's that in USD?
The list of accredited ICAAN registars is at InterNic.net. This is just to let you know the alternatives, and if some registration company is really allowed to register domains.
So in their present state, you can't really read anything outside the largest circular region on the disk. Darn.
Okay, so I don't really know that much about CD technology (what a shame), but I do know that you can record data in a number of sessions, and that newer drives (MultiRead drives, including most all CD-R[W]s) can read more than 1 session. Anybody know if you could somehow use that unused space by creating some more sessions to fill it in? You would need a lot of sessions to fill in all of it (how many sessions can you have on a CD anyway?) if this were indeed possible. I am assuming that the only reason that you couldn't just structure your iso9660 fs just right to avoid read/write on the clipped-off parts of the disk is because the laser is just going to wander over there anyway (read, Bad Thing).
Of course, a DVD in this size...
This is a big problem with lots of binary-only Linux software: Linux runs on many different kinds of processors. Wine constrains you to x86 (at least for emulation; I have never used Wine as a porting library.). On the other hand, that might not be that bad for MS -- "We released a port! (it only runs under Intel processors, so tough luck)"
That means Unix, and once that happens, how hard is it to make a Linux/BSD port?
Really hard. Why? 'cuz Gates has to swallow it.
And remember that OS X has Aqua. I'm willing to bet that MS is going to take very full advantage of Aqua's prettiness to hide some mess. So what if it doesn't work? It looks nice!
And remember that MS likes to take advantage of lots and lots of undocumented features released to them under NDAs (okay, that's just what I observe). Linux probably doesn't have 95% of those "features".
Nope. Didn't work. I booted the kernel off of a disk; it detected the controller fine and even got most of the way through reading my partition table and then suddenly stopped, then after a short time said "Timeout waiting for DMA". I sort of have an idea what that is, but on the contrary I have no idea how to fix it. This is starting to get really annoying. I bought a board for its UDMA, and its UDMA isn't working!!! Is there anyone with Gentus who has gotten it to work on the BE6? If so, can you please send me the diffs between your kernel (source, of course) and the standard 2.2.14 (or is it .14? whatever it is.)!?!? I tried all the different UDMA modes on the HPT366 controller, and some of them work even worse than UDMA Mode 4 (which was what I tried first), but none any better. (btw, the HPT version for the BE6 latest BIOS is 1.21 (or some mix of decimal points among that)). This says nothing about the patch linked to in the parent comment working on the BP6.
Anyway, you probably deduced from the above that I am not happy.
Oh well; that's just what I get.
Don't like Red Hat? I'm pretty sure the significant modification to the kernel can be found here.
I hope it works with the HPT366 okay.
An HPT366 driver is built into the 2.3.x kernels (at least 2.3.45, which I have tried and then dumped because of nasty fs corruption). The docs for that say that the HPT366 chipset is in its present state unbootable. I don't believe that 'cuz it'll work on Windows and of course, this here Red Hat clone. Clear this up, anyone?
I'm trying the driver now (okay, still have to patch the kernel and rebuild).
I have an ABIT BE6, not BP6. They both use HPT366. Any differences in the implementation that BE6 users should be aware of besides using the latest BIOS?
The above comment brought to you by the Schemeing mind of Kenneth.
I have seen a few programs (ssh comes to mind) that have a 'make uninstall' option if you download the tarball. Some uninstallers are, of course, better than others. Perhaps a general program that would analyze a Makefile with an 'install' "phony target" and either (a) add a 'uninstall' target to the Makefile, or (b) allow the user to uninstall the program through a separate utility. This progam would have to allow a user to uninstall personal configurations (rc files, directories, etc.) for a specific user or all users, or even all users except those who have specially requested to keep their configuration. It could even be implemented as a wrapper around make and monitor the installation (or at least make backups of any files that make would replace), sort of like CleanSweep SmartSweep for Windows. I am thinking kernel module to monitor a specific process. This is starting to sound a little overkill, but how else do you think it was done in Windows?
Make me feel stupid and say that a program like this already exists. After feeling stupid for a second I will be happy.