Slashdot Mirror

2. Does not require PulseAudio, but can still output multiple channels from multiple apps at the same time. This was always a problem with ALSA.

dmix says hello, while flipping PA (and Pottering) the bird.

Would you mind telling me what brand/model your laptop is? (..) I've been worrying about having to use Pulseaudio one day/in a laptop.

It is a HP dv6535ep, the sound card is

$ lspci | grep Audio
00:07.0 Audio device: nVidia Corporation MCP65 High Definition Audio (rev a1)

a Intel HDA with a Conexant CX20549 (Venice) chipset. No hardware mixing.

From http://alsa.opensrc.org/index.php/Dmix

NOTE: For ALSA 1.0.9rc2 and higher you don't need to setup dmix for analogue output. Dmix is enabled by default for soundcards which don't support hardware mixing. You still need to set it up for digital outputs.

Pretty much everything uses dmix now a days, even gnome that was using the crappy esound interface moved to gstreamer long time ago, which supports pretty much everything (alsa, oss4, pulseaudio).

Unless you need some outdated piece of audio software, I don't think you have to worry about something hogging your sound device.

It's ridiculously complex, but I'm sure I can make it even worse. Audio production is JACK territory. JACK is an "audio connection kit" meant specifically for apps like Ardour (recording), Rosegarden (sequencer), Hydrogen (drummachine), JACK-Rack (virtual rack for effects plugins), ZynAddSubFX (software synthesizer), or Jamin (JAck Mastering INterface). When you run the JACK daemon, these and other apps can talk to JACK rather than fight over access to the sound hardware.

More interestingly, JACK lets you connect the MIDI and audio inputs and outputs of these applications and your sound hardware to each other in meaningful ways. For example, you could (1) let Rosegarden control Hydrogen and an external, physical synthesizer via MIDI, (2) let an external MIDI keyboard control ZynAddSubFX, (3) send Hydrogen's output to JACK-Rack, (4) send JACK-Rack's and ZynAddSubFX's outputs and the soundcard's line-in to Jamin, (5) send Jamin's output to Ardour or Audacity or whatever else you want to use to record the results, (6) and so on. (I don't know if such a setup ever happens in the real world. Perhaps you're now twitching in light of my audio naivety. But it was just an example.)

Check out http://jackaudio.org/intro and http://jackaudio.org/faq for a friendly introduction.

With QJackCtl, you get a graphical JACK front-end that'll let you set stuff up with relative ease. Install it.

What JACK then DOES all the audio buzzing about "within" it is, typically, this: It gives it to ALSA. Unlike Pulse, JACK, GStreamer, Xine, Phonon and many of the other funny names you'll come across, ALSA is part of the kernel (or rather, has parts in the kernel). So ALSA is not an optional convenience. The ALSA project provides soundcard drivers; through ALSA, applications talk to the sound hardware. (Unless you install OSS4, I suppose. Or do something else I don't know about.)

I usually attach JACK to the ALSA interface hw:0. The hw: interfaces provide direct access to the sound hardware, bypassing ALSA's own software mixing, rate conversion, and other niceties that may (?) slow things down or lower quality. hw:x,y means soundcard x, subdevice y. hw:x without a subdevice means use whatever subdevice is available. In my case, subdevice 0 has four hardware audio channels, and subdevice 1 has one. I suppose that's why I can let JACK talk to hw:0 without it blocking out any other, non-JACK applications like the Flash plugin, media players, or games.

There'll be other ALSA interfaces available in QJackCtl besides hw:, in my case: plughw:. And Audacity's preferences offer a different and larger ranger of ALSA devices besides hw:0,0 and hw:0,1, for example: spdif, iec958, dmix, and default. As far as I know, plughw:, dmix and default all provide software mixing, so they may have their uses if you find some application blocked by another (such as the JACK daemon) due to a lack of hardware mixing. And if ALSA's software mixing doesn't work for you either, PulseAudio or something like it comes in as another layer on top of ALSA providing software mixing.

ALSA woes can be mended and virtual ALSA devices with specific properties created in ~/.asoundrc, though if you're lucky you don't need that file at all. See http://alsa.opensrc.org/index.php/.asoundrc for the lowdown.

I said JACK attaches to ALSA, but you can also attach JACK to OSS, for example. OSS these days is something like a compatibility layer provided by ALSA, which has replaced OSS. OSS devices on my system are /dev/dsp, /dev/dsp1, and /dev/audio.

There's also OSS4, which isn't "legacy" and which some ALSA-frustrated users swear by. It doesn't come with any distributions I'm aware of, so you might have to install it yourself.

I suppose you could also attach JACK to (say) PulseAudio as an intermediary layer betwe

How do you handle per-application volume levels in alsa?

softvol, isn't perfect but surely its better than this and can be improved on, OSS4 also implements per-app volumes at API level.

How do you use another machine on the network as a sink in alsa?

google it. Outside of thinclients i don't see how that is particularly useful anyway, so giving everybody else glitchy audio, instead of just using PA on thin clients seams stupid.

How do you handle per-application volume levels in alsa?

softvol, isn't perfect but surely its better than this and can be improved on, OSS4 also implements per-app volumes at API level.

How do you use another machine on the network as a sink in alsa?

google it. Outside of thinclients i don't see how that is particularly useful anyway, so giving everybody else glitchy audio, instead of just using PA on thin clients seams stupid.

There is. ALSA's dmix has been enabled by default for a long time, years. Have you even tried Linux? I can't remember the last time I had to 'configure' sound on Linux. Insert sound card, mixer shows up, play sounds. From the ALSA wiki: "NOTE: For ALSA 1.0.9rc2 and higher you don't need to setup dmix. Dmix is enabled as default for soundcards which don't support hw mixing."

Yeah, it's supposed to work, but for some reason for me it doesn't.

And have you looked at that page? It's full of listings of arcane incantations. Really, I just want the darn audio to always get mixed, without having to get a degree in audio engineering to understand what's going on there.

If you bothered to try, you'd find that it does.

See the dmix page, which says "Normally (without hardware mixing) you cannot use /dev/dsp multiple times directly."

So it seems that if you have onboard audio, and want to have more than one app use /dev/dsp, you're out of luck.

ALSA supports LADSPA plugins. Example compressor/limiter setup here: http://alsa.opensrc.org/index.php/Ladspa_(plugin) I use this for watching movies at night without disturbing neighbors, works very well.

We are talking about a GUI subsystem that can't even fall back to a safe res/refresh when something goes wrong.

Except, apparently Ubuntu Hardy does exactly that.

Its config tools are terrible

You mean xf86config? Those are almost never done directly, and the vast majority of systems boot to a working GUI off the livecd. If a livecd can boot properly, there must be a tool somewhere that's generating a workable conf file, right?

And then there is the putrid audio architecture

I quite like it, actually, but then, I suffered through OSS.

I don't know if ALSA defaults to falling back to software mixing when it runs out of channels. I'm fairly sure it can. But I know I was grateful both to have a sound server (back when I switched from Windows 98, where I could play sound out of exactly one program at once), and then later, I was grateful to not need a sound server (back when doing the mixing in hardware was a significant speed boost).

Actually, a quick Google search confirms it: you can get software mixing in ALSA. I suspect that this will eventually become the default -- and according to this page, it already is, at least for some builds of ALSA. Do you have this problem on Hardy? (A fresh install, so it can properly autodetect?) The thing was less than two weeks from release when you posted, so it would have been fair to test...

Once you've gotten that bad at audio AND visual, then you're pretty much hostile to the end-user. Not focusing on these areas is sheer absurdity

Which is why they are focusing on them, I would think. Hardy apparently has some things to make X easier.

"Wikis are the lazy or uninterested programmer's way of doing documentation."

I think most packages have good installation instructions, OK usage instructions, and no troubleshooting information. The problem isn't that the programmers don't want to write docs, it's that they have no more idea what to write than you do. Take a look at Linux sound. The ALSA wiki is the only place to go to try to find what people have done to get specific cards working or problems solved.

Well, as other posters have said, Linux is just an OS kernel, not a distribution; having said that, there is a great deal of innovation going on in the open source world.

Linux:
1. User-space file systems. FUSE. This stuff is neat. Linux supports a panoply of filesystems that Windows users can only dream of, and a lot of these are worlds and worlds ahead of Windows stuff. Take a look at FunionFS, and Wayback FS.
2. Abstract, granular CPU and I/O prioritization and scheduling. Linux can be realtime in ways that NT can only dream of; which is impressive considering the scale of Linux.
3. LinuxBIOS. Anyone stuck an NT kernel into Motherboard firmware? No? Why not?
4. KVM. Linux kernel virtualization. Microsoft is talking about duplicating this for the NEXT version of NT.
5. A fully relocatable kernel. New in 2.6.20
6. How about a native IPv6 stack? Linux did it first.
7. How about boot time switching between 64-bit and 32-bit, or ACPI and noACPI? How about probing/autoloading of modules on boot? How about all possible drivers being installed, all the time, even ATI and NVIDIA's closed-source drivers now, using the Novell KMP system?
8. POSIX compliance (uncertified), AND Win32 compliance (uncertified). First OS to do this.
9. Support/scaling for an unlimited number of processors?
10. How about a flat memory model (4GB/4GB split), even on 32-bit?
11. Don't forget about ALSA. Wanna change how your sound is mixed, in userspace? No problem. Wanna reroute your mid-rear-left speaker to your record slot? No problem. Want 3D sound in older applications? OpenAL is there for you (unlike DirectSound in Vista). Here's a list of ALSA plugins, all of which are utilized in userspace: http://alsa.opensrc.org/ALSA_plugins .
12. Vast improvements in Kernel security all the time. Things like selinux, and AppArmor (AppArmor is really cool stuff) are worlds beyond UAC and group policy.

And that's just the OSS Linux kernel. Wanna talk about other subsystems?
CUPS versus Windows printing?
1. Autodiscovery of local subnet printers? Not possible in Windows, even Vista.
2. End to end Postscript printing, even on $15 crapprinters?
3. Out of box support for IPP, CUPS, LPR, SMB, and other kind of printing system you can dream of.
No matter how you slice, CUPS is worlds away from Windows printing. Never, ever have to deal with printer drivers as you move from network to network; this is a dream avaliable for years in the CUPS world.

X? Xorg is a thing of beauty.
1. Full network transparency (2D/3D). Not avaliable in Windows. Best of breed network performance using NX.
2. A fully modular windowing system. Remove or add components at will. No Internet Explorer required.
3. Extremely high performance, with decades of support for both 2D and 3D operations.
4. The sky's the limit in terms of scalability. 1 monitor? 4 monitors? 64 monitors spread across 12 systems? No problemo.
5. Xgl is the beginnings of a pure 3D windowing system with legacy support. Xegl is the future of this pure 3D windowing system, at performance levels that put Aero's hybrid 2D/3D setup to shame.
6. Yes, spinning cubes. And a whole lot more eye candy. On a whole lot less hardware than Aero requires. Geforce 5200 mobile with 32 MB of RAM? No problem.

GUIs?
I don't know much about Gnome, as I'm a KDE guy, but:
1. KIO-slaves. ftp:// ? of course. bzip2:// ? torrent:// ? fish:// (this one is amazing, directory browsing over plain SSH). beagled:// ? how about man:// or programs:// ? how about klik:// ? KIO-slaves are one of the coolest features in GUIs out there, hands down.
2. Kparts. Click on a PDF url, and you get KPDF in your Konqueror window. Click on a DOC url, and you get Kword in your Window. Click on an RPM, and you get either YaST2 (for SuSE), or KPackage. And all of these are user configurable, of course, on a user-by-user basis. This is something that neither OS X or Windows have worked out correctly.

Hmm. Your post convinced me to start this off even though I don't think I am ready to.

I have been thinking lately about "copyright pollution" where the copyrighted works of others gets in our heads and pollutes them to to point where what then comes out of us is "tainted" as it were.

Now this should not really be that much of an issue in a sensible legal environment, but I think we may not be in such an environment now and I also think that those forces causing that environment to deteriorate for a good while now are still at work trying to make it worse.

Are we getting to the point, or will we get there in our lifetimes, where an artist cannot afford to pay attention to, watch, look at, read, etc. any works that are not in the public domain or that carry a free license of some sort? Sort of like where people say we are today with respect to software patents.

all the best,

drew

http://musicians.opensrc.org/DrewRoberts

{'Well, perhaps it could. If trademark law was taken overboard,'

Well, there is a reason TradeMark and Copyright are different.}

You forget that this is all in the absence of any copyright law (from my point of thinking for this excercise) except where explicitly talking of copyright law.

{'This is just what people say about GPL code / Free Software and yet people do make money from it.'

Yes, but they have other streams in order to make money from [such as Red Hat having a support line and Asterik selling hardware to go with their PABX software]. People who are solely songwriters don't have other means to make money from their songwriting effort.}

I know, but the same hold for people who are exclusively code writers. I am about in that position of being only a songwriter myself (mostly lyrics actually as I can't really sing and don't play an instrument well enough either. I do work sometimes with a friend and sing to him and he tries to figure out from what I am singing what I am trying to sing!!!)

One problem is though is that the idea is out there, the licenses are out there in the wild and people are using them and experimenting with them.

http://musicians.opensrc.org/DrewRoberts
http://ccmixter.org/media/tags/attribution

Look around, there are more.

{So, unless there is some other means for them to make money, I don't have a problem with paying copyright royalties to perform their songs. [It's only one cent per performance, and most performers can afford that, even if they perform a concert with thirty to fifty songs, they should be making enough profit off each performance to afford 50 cents or so]. }

The trick is to find that some other means.

Is that how it is were you are? I need to look into it down here. Not that I need to know at this time personally. I do think that in some places it is a percentage of revenue though.

{Like I said, Copyright laws are consistent in what can and can't be copied.}

Perhaps by fiat, but not by some grand ste of consistent principles that anyone has ever pointed out to me or that I have found in all my searching.

Could you copyright a landscaped garden? Why is a cars coachwork not art? After all, at least in bygone days, sculptures of the designs were made before the actual cars were ever produced. This may all be done on computer now.

{The difference here is really in what is considered 'Performance art' and what is static art. It's not an inconsistancy as much as it is a differentiation as to the 'type of art'. }

But in the case of music or video which is recorded and not live, it is not in fact a performance. Not like a band playing a cover is a performance.

BTW - as an odd aside, I wrote a novel in november last year as performance art. I wrote in in an IRC channel on freenode. Some people even dropped in to watch the struggle at times.

The final unedited mess is here:

http://www.ourmedia.org/node/262954

{It's not legal to claim you wrote something in public domain, but that would result in a plagiarism case. The copying of the work wouldn't consist of a crime, only the claiming it as your own work. }

Gotcha, so you say there is overlap and cases are brought under copyright as it is simpler. In the absence of copyright law though (remember) cases of plagerism could still be brought under the plagersim laws. (I have no reason to doubt you, I have just never read any plagersim laws.)

{If my memory of the 'fake' group of books is correct, they dumbed down the songs to have easier chords/melodies than the original songs.}

So they would be claiming a copyright in the arrangement? What about the original words?

{I think they dumbed down the chords for easier playing. Often they transpose everything for guitar into the key of 'C' as well.}

Dumbing dow

"All the people complaining about DRM should actually DO something"

DONE.

Sayings - Deterred Bahamian Novel - http://www.ourmedia.org/node/262954

Tings - Anuddah Bahamian Novel - http://www.ourmedia.org/node/85937 &
http://www.ourmedia.org/node/111123

drew Roberts's Storefront - Lulu.com - http://www.lulu.com/zotz

Some tings for you from zotz : CafePress.com - http://www.cafepress.com/zotz

Now for some other stuff of mine:

http://www.youtube.com/results?search_query=zotzbr o&search=Search
http://www.archive.org/search.php?query=(creator%3 A%22drew%20Roberts%22)%20OR%20(collection%3A(ourme dia)%20AND%20%2Fmetadata%2Fauthor%3A(drew%20Robert s))
http://code.google.com/p/drsoundwall/
http://www.ourmedia.org/user/17145
http://musicians.opensrc.org/DrewRoberts
https://sourceforge.net/projects/zbcw

I am not the only one doing such things either. For instance:

http://ccmixter.org/media/tags/attribution

"so CREATE something yourself and see how it works voluntarily instead of forcing authors to agree with your politics."

Ah, I am not the one running to get copyright laws amended over and over. Retroactively. There was a legal (lopitical?) agreement made with the public, but it wasn't good enough for some. They wanted to change the agreement. Now it is wrong for others to change it back to something more like it was? Or even completely different?

Seems some people are trying to force us into new "agreements." Why should we not fight back?

all the best,

drew

"The entire value of the book is wrapped up in its IP, because copies have a trivial cost (compare to 200 years ago, when printing books had a significant cost)."

Well, I am putting my time where my mouth is and experimenting with other possibilities:

http://www.ourmedia.org/node/85937
http://www.ourmedia.org/node/262954
http://www.ourmedia.org/node/187924

"I'd be really ecstatic if there were stricter controls even than we have now--as long as the length of copyright was reduced drastically and keys were escrowed with the government and released at the end of the copyright term."

Well, I would say that you are not thinking straight or that the punishments in your country are no where near as bad as they are in mine. How does 4 or 5 years in a reputedly very nasty jail for each non-genuine CD or DVD in your posession sound? (By that I mean even ones that you may have purchesed in good faith and been ripped off in the purchase because you were sold a bogus disk.)

I am fine with the drastically reduced terms though.

Would you care to comment on the first comment at this link:

http://www.digitalproductions.co.uk/index.php?id=4 3#comment

I am also experimenting with music:

http://musicians.opensrc.org/DrewRoberts

and am thinking about messing somehow with a movie next year...

all the best,

drew

I am living where I come from, my people are from here from way back. I would rather not leave my home just because the guys in power pass some cockeyed laws.

Are all of your country's laws perfect? All the politicians honest over that way?

all the best,

drew
http://musicians.opensrc.org/DrewRoberts

"From this point on I aim to only listen to copylefted music. Movies and TV? I'd rather have fun making a copylefted movie than killing my brain cells and liberty with an MPAA offering."

I am trying on the production end of the copylefted music front:

http://musicians.opensrc.org/DrewRoberts
http://www.ourmedia.org/node/145261 (I really need to do this one.)

Would you care to brainstorm on the movie side of the equation?

all the best,

drew
http://www.ourmedia.org/node/262954
Sayings - Deterred Bahamian Novel
CC BY-SA (think copyleft)

"Hmm. Of course...who's working on that free music again?

www.anvilstudio.com"

I am, for one:

http://musicians.opensrc.org/DrewRoberts

May I suggest considering a copyleft type license for your Free music?

Oh and as to programs, people might check:

http://lmms.sourceforge.net/
http://ardour.org/
http://www.rosegardenmusic.com/
http://www.ferventsoftware.com/

"When it comes to music, listen to local bands playing at pubs or other venues."

Better yet, fund the creation of some decent copyleft type music. Raise some money and pay some of those local bands to write and record a song under a copyleft licnese. Then if it's any good, flog it.

all the best,

drew
http://musicians.opensrc.org/DrewRoberts
A small start - as yet unpaid.

"It's looking more and more like the only safe user content to host is plain old text, and I think that's a damn shame."

http://musicians.opensrc.org/DrewRoberts

Some people are making works that should be safe to host. Fund the creation of more stuff with licenses you like and host those works.

You will get a positive feedback loop going and things may have a chance of being different.

Talk to your local bands about funding some of their songs under Free licenses.

all the best,

drew
http://www.ourmedia.org/node/262954
Sayings - Deterred Bahamian Novel

Actually, ALSA supports software mixing (and hardware mixing on some hardware) of different sound applications; see information on its dmix plugin. When used in conjunction with ALSA's OSS emulation, this allows (most) sound-using applications to co-operate.

Fuck Creative Labs. Doesn't any one make cards that can do hardware mixing any more?

Dmix won't be good enough until it also works for applications using snd-pcm-oss's /dev/dsp emulation.

But ALSA does support mixing, even on cards that don't support it in hardware.

Dmix necessary for most sound cards? I don't think so, unless by "most sound cards" you mean "really freakin' old sound cards." Almost every modern card, including the on board sound on every single motherboard I've used in the past 2 or so years, my 5 year old Sound Blaster Live, etc, do not need dmix to mix multiple streams automatically, and hardware accelerated.

But, having said that, you're right, there are some cards that need software stream multiplexing. And in that case, check here. I've used this guide to set up dmix on a really old card I had laying around. The guide is very easy to follow, and took me maybe 5 to 10 minutes including the time it took me to google for the appropriate info, and play a couple songs (simultaneously) afterwards to test it out.

One thing I realize though... if a soundcard is so crappy as to not support hardware mixing, this is generally a good indication that its sound quality may be... less than desirable.

Are you referring to dmix? Just enable OSS emulation in ALSA and have OSS applications speak to that, works like a charm.

Windows XP installation requires no more effort than a modern Linux installation. In fact, it is pretty widely accepted that it is the Linux distribution developers who are playing catch-up to Microsoft in terms of the user friendliness and automation of the OS installation process. Thankfully, Linux installations are mostly very simple nowadays because of this. This might've been a valid argument when Windows 98 was around, but that was quite a while ago.

In response to your issues with patching, when was the last time you installed a Linux distribution that did not give you a friendly warning that you need to download tens of megabytes worth of patches and updates? I mean, sure, you can argue that Windows patches are more important because the standard installation of Windows XP is a lot less secure than the standard installation of any Linux distribution, but if you argue that, you'd be forgetting one key point: Your Linux distribution of choice will likely be several months old, at most. Windows XP, on the other hand, was released when? Around five years ago?

Also, while I'm sure that Microsoft's default hardware drivers would usually not give you full functionality of your desired device, Windows at least recognizes that the device exists and to give it some basic functionality. No matter how advanced, a soundcard with Microsoft's default driver will play sound. A videocard with Microsoft's default driver will display your desktop on the screen. The sound may not be as crisp, and the video may not go to desired resolutions, but at least these devices will have rudimentary functionality.

A visit to the manufacturer's website will usually let you download the official Windows drivers and install them fairly easily. Because they usually don't provide Linux drivers, you're stuck performing complicated system administration similar to what is described here. Leaving such things up to the user is simply unacceptable in a home desktop environment.

Unlike the 30 year old definition of the word "user" which most Linux devotees arrogantly cling to to this day, a user is a layman consumer, NOT the programmer.

When this stops being the reality of Linux, only then will it really be able to compete with Microsoft. All security issues aside, people just want their computer to WORK.

Take a look at this. Getting alsa's dmix plugin to play nice really isn't as difficult as it would seem from the documentation.

Even the latest Knoppix (2004-02-16) CD doesn't support my Aureal2 card. Yes, support sucks because Creative Labs bought Aureal and has refused to release support info for the cards, but there are drivers that work. The problem isn't the code, it's the documentation.

While looking for documentation on ALSA, I found this gem;

http://alsa.opensrc.org/index.php?page=AlsaComplai nt

To this day, I won't even look at ALSA. As long as this OSS driver is working, I have no motivation to change.

I don't like the idea of a GNU/Linux exclusive sound system anyway. I use BSD too.

This clever piece of code performs mixing using shared memory and IPC primitives - no kernel support, no sound servers. It's the Way Forward(tm)

You know, silly stuff like reliable, robust video and sound drivers.

It's funny, but Linux is in much better shape for video drivers than audio ones. Since the game-capable graphics market only includes two companies, Linux is already adequately usable.

But since soundcards are technically easier to make, there's many more brands still in active use. Many gamers who buy the latest NVidias to squeeze a few more FPS or pixels might still be satisfied using motherboard audio output, or a $2.50 PCI soundcard.

Linux audio support is close to adequate... but unfortunately, the Alsa Project's longstanding philosophical refusal to move software mixing into the central driver means you still can't expect Linux to run games on any random piece of desktop PC hardware.

The subject pretty much says it all ...

Read this or this for more info.

Death to ESD/ARTS today!! (and maybe even JACK, if we can low enough latency).

Sunny Dubey

AFAIK, from reading www.kt.opensrc.org, there is already a patch for this.

" Andrea Arcangeli posted a patch and announced that he and Gerhard Wichert had co-developed the patch to allow nearly 4 gigs of memory on 32-bit systems. A big discussion followed."

Located here

From what I've heard, correct me if I'm wrong, development of Gimp has really fallen off since the original developers graduated college.

OK, you're wrong :-) Take a peek at the Gimp-Devel summary page ( http://www.kt.opensrc.org/KC/gimp-devel / ). It summarizes a lot of the activity on the gimp-devel list.

In addition, the May issue of TheGimp.com has an article about some of the new features ( http://www.thegimp.com/articles/1999 05.html ).

If I understand the history correctly, frustration at the lack of 1.1 releases (since all work was being done in CVS) resulted in a semi-fork and the series of 1.1. A reasonably comprehensive guide to the changes can be found at http://sven.gimp.org/1.1/ . Pay particular attention to the feature list.

Development isn't stalled, and it is moving forward, but it's not being shouted from the treetops. Personally, I noted a marked difference when I tried one of the 1.1.x released, at least as compared with 1.0.x. There's a lot of exciting new stuff in there.

Maybe someone else with more information can comment, but I believe the problem with WinModems is not so much in writing the drivers, but in the licensing costs of the V.90 protocol.

Go see this page for a little more information.

Or, if you want to see why most of the real linux kernel guys hate Winmodems, follow this thread.

Alan Cox estimated 200k+ lines of code, running in kernel space.

So, forget about the specs. Even if you had them, they would be either too expensive to use or not worth the impact to the kernel to implement. Better off spending the few extra bucks, and taking a load off your CPU.

There seems to be discussions about capabilities though (check the last few Kernel Traffic issues). Which seem to do what ACLs do in other systems. Or am I mistaken?

"Bend Over, Boy, Because You Have It Coming To You."