Slashdot Mirror
USB Forum Becomes Too Greedy?
I just saw this on the Linux USB Mailing List: Up until a few days ago all the class specification documents were available for anyone for free. Those classes are important for the Linux USB development. It appears now, after the USB group got the Slashdot beany award, that the access for those documents is now restricted to members only, which are required to pay $2,500 per year. Is this just a coincidence? You can test it here.
http://www.usb.org/app/db/search/contacts/ Look at the source to get a bunch of Vendor IDs. Then append "-admin" and use that as a username and password.
um... I think you're the one getting confused. See http://www.usb.org/developers/docs.html where it says:
USB Device Class Specifications
(USB-IF member username and password required)
If you click on the link you are indeed prompted for a member username and password.
-Andy
Just 7 months after the whole /. community was screaming over "Apple licensing" of Firewire, I find this most ironic.
:-)
USB 2.0, be it an eventual product or vaporware announcement, was designed to steal mindshare from Firewire.
USB was designed for cheap low-bandwidth plug in parts. That's what it is good at. Trying to scale up from that detracts from the original design. It makes absolutely no sense to have a mouse on the same bus as an external hard drive or video capture.
Now that Firewire has been FUDDED into a mostly-Apple expansion port, Intel moves in for the harvest. Nice. You didn't see it coming?
FOr those of you that don't know, "Firewire" is the same as "Sony i.Link" and 1394 ports. Sony just calls it that to prevent mindshare from Firewire.
Bah. No more coffee this morning.
> It should be noted that many other operating systems also currently support USB as well. BSD's not with standing.
... but FreeBSD has had USB support for some time now.
I'll suppress the flames, it didn't look like a troll
I've finally had it: until slashdot gets article moderation, I am not coming back.
REDMOND, Wash. -- March 1,2000 -- Microsoft (Nasdaq "MSFT") announces the release of DirectExperience(tm), a revolutionary new system designed to provide an unparalleled audiual, visual and for the first time, tactile experience while providing optimal protection for the audio-visual-tactile content.
By inserting a series of IntelliPlugs(tm) into the flesh of the arms, legs and spine along with a DirectExperience(tm) cranial socket, content providers can download audio, visual, and tactile sensory experiences into the cortex of the brain, providing life-like experiences unrivalled by any other streaming medium released to date.
In addition to providing an encrypted stream of information right up to the integrated dermal and cranial sockets, reducing the chances of a successful side-channel attack, Microsoft has worked closely and secured an unparalleled deal in which all infants born after March 31, 2000 will have the plugs introduced in-vitro and connected to a DirectExperience terminal shortly after birth. Microsoft plans to use the profits from the licensing of its technology to content providers to allow for intravenous nourishment for these infants through the course of their natural lives. It is estimated that within 50 to 70 years, every person in the world will be upgraded to handle DirectExperience technology; after that, production of new consumers will be off-loaded to a separate farming location.
Early product testers, when asked about the DirectExperience(tm) system, offered the following comments:
"*belch*"
"Glurgh..."
"There is no spoon..."
Microsoft President and CEO Steve Ballmer, when asked about the DirectExperience(tm) system, said "It's the smell, you see..."
Jay (=
Since it's pretty unreasonable to expect individual OSS developers to ante up $2500 for the privilege of writing software, what about some organization such as the FSF paying the fee for the specs and signing up people to develop drivers, etc.
Is there some sort of nondisclosure that restricts anyone from doing this? Or are only commercial software vendors invited to the USB party?
This smells a little like the fracus that resulted when Intel closed some specs a while back. Didn't they eventually relent?
Finally, the wiseguy who thought the editor should research before posting should have checked the links on the URL he posted. The link to go to the class specs was indeed password protected.
--
CUR ALLOC 20195.....5804M
btw if you want to make a mirror of the members area off usb.org w/wget do something like:
wget -m --http-user=1211 --http-passwd=cypherpunk -np http://www.USB.org/members/devclass.html
the -np will make it so you do not get parent directories.
- ixx -
didn't mean to post anonymous on that last mirror post.
Just my conspiracy theorist notion.
"Cause there's 40 different shades of black, so many fortresses and ways to attack, so why you complainin'?"
I suspect you might've gone and confused 'open' and 'free' and 'free' yourself, too?
;) The fact that "there are a lot of people who'll never need to know the contents" doesn't permit a financial elitism. That is a business/commercial mentality which has no place in the Open(-Source) arena.
Let's use "Free" for "idealogically Open", and "free" for money for the time being.
Then I'd say that a specification should be Open - must be! - or Free, and it should be cheaply available.
Now, if it's Open then there are no restrictions on redistribution: therefore you can sell it to my friend for $2500 and he'll give me a copy because he's nice like that. All without license breach. And more to the point, I'll take a look at it, change it in a few places, and send it back both to my friend and the original authors as diffs.
(Incidentally, I think your use of "GPL" there is unfortunate. LGPL, possibly, or something else. GPL is a bit inapplicable when it comes to specs, because if it's going to be a Spec, then it needs to be relatively fixed.)
Now, this $2500 thing: is the figure reasonable? Is the documentation for USB still Open? Is it restricting distribution or biassing itself against a particular set of users? (E.g. the GPL states there should be no division of users on such things as country/ethnic/religious grounds - can I extend that to "financially able"??)
*I*'m not "a professional developer", though. I don't see why I should have to work for A Big Linux Named-Company in order for them to afford a copy of the Spec for me. In fact, that scenario is no better than the current commercial scene. Bad.
And while I'm here
~Tim
--
Rushing on down to the circle of the turn
FOr those of you that don't know, "Firewire" is the same as "Sony i.Link" and 1394 ports. Sony just calls it that to prevent mindshare from Firewire.
Firewire is not "Sony i.Link." Firewire also provides a line of power to the devices connected, whereas products plugged into Sony's i.Link do not. A subtle difference, but one that could make a difference when buying a firewire drive WITHOUT an AC Adapter.
-Nick
Perhaps the Slashdot article doesn't make it clear - it's not the Linux-USB people who are restricting access, it's USB-IF who are the culprits in the matter.
My question is, why would someone want to restrict the development of drivers for their products? Surely this is simple maths - no support, less people buy. Take the SBLive for example - this has excellent support under linux, and many Linux weenies went out and bought it [GRiN] ...
.my 2p
While source code is GPL'd, documentation isn't. Anyone could look at the source and document the class specs and charge for it. But in the spirit of open development and abstraction(I wouldn't want to spend time interpreting what a function does if someone else could tell me), perhaps future docs need their own public license. The pace of development would quicken and the barriers to entry (spending the time to figure out wtf something does so you can write code based on it) would lower.
But isn't the purpose of the Doomsday machine lost if you keep it a secret!
Linux can't just drop USB. If I go to the PC World next to my office I can buy USB keyboards, mices, scanners, printers, mp3 players, digital cameras etc. Most (if not all) new PC's and motherboards come with built-in USB. Linux needs to support USB. Therefore the developers need the specs. Whilst the specs should be freely available, $2,500 is peanuts to companies like IBM, SGI, Corel, VA, Red Hat etc.
HH
Yellow tigers crouched in jungles in her dark eyes.
Yellow tigers crouched in jungles in her dark eyes.
She's just dressing, goodbye windows, tired starlings.
Makes me wonder if the Beanie can be revoked.
As far as I understand, it was the Linux developers, not the Forum, that got the beanie. They'll need (and deserve) it more than ever.
-
__
Comment submitted. There will be a delay before you understand what you posted.
Start recruiting hardware manufacturers, take the last open version of the spec, Fork, and start working from there. Come up with a sticker for device and motherboard manufacturers (OpenUSB?) and refuse to buy any hardware that doesn't have the sticker.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
Linux companies could indeed join the members list (but probably not redistribute the obtained material, shutting out everyone else) - but (aside from the possibility of looking at Linux code and guessing why it's doing that) how would this help other OSes, like the various BSDs?
For something like USB to become a <i>real</i> standard, it has to be truly cross-platform, usable on every OS. I don't intend to switch hardware every time I boot into a different OS.
This message is provided under the terms outlined at http://www.bero.org/terms.html
"Methinks the editor should have done some research himself and put some of these in the story."
If you had read all the messages you would have seen that HeUnique was not only reading those links, he posted to the thread asking for more information and their (usb developers) opinion on posting this story to slashdot.
Interestingly, Alan Cox suggested waiting for a week before posting to Slashdot to give them a chance to explain and/or change back graciously before being "hit by a nuclear warhead"
Maybe Slashdot could make some sort of community watchdog section where things that seem suspicious could be posted, but in sort of a beta state before going to the main page.
HeUnique:
What's your response to this message? http://electr icrain.com/lists/archive/linux-usb/2000/02/msg011
You know...I really should start making posts of actual substance, instead of continually tearing down Slashdot, but I think Slashdot could be an excellent medium if some issues with it were addressed - such as journalistic integrity (as if there were such a thing).
FireWire supports both synchronous stream data, like video, and block data, like disk drives. Most of the action is in synchronous stream data, primarily video. Block devices for FireWire are rare, but they do exist. I just saw a Zip drive, of all things, for FireWire. The big problem isn't Apple, it's Intel and Microsoft. Intel has backed away from block-device FireWire support and is now pushing USB 2 for that purpose. Microsoft was late with FireWire support in the OS.
There are a huge number of FireWire devices.
Here's a short list (not for 28K modems).
Some related articles/sites:
The GNU/Linux IEEE 1394 Development Project
FireWire: The Fast, Easy to Use, Multimedia Interconnect Standard
IEEE 1394, the A/V Digital Interface of Choice
FireWire -- For Footage That Flies
Papers on IEEE 1394 Technology
Texas Instruments Interface Products
--
He lives in a world where those who do not run the client software of the omnipresent meme are unacceptable.
In the links thread where people broke in, a "coward" broke in, got the specs, and noticed exactly why they want this rev hidden. It has nothing to do with the money that the USB folks just got, and everything to do with trying to do security the stupid way.
Cheers,
Ben
My usual seat in the cluetrain is at A HREF="http://pub4.ezboard.com/biwethey.ht
FOr those of you that don't know, "Firewire" is the same as "Sony i.Link" and 1394 ports. Sony just calls it that
to prevent mindshare from Firewire.
No, sony calls it that because APPLE won't let people use the word "firewire" without paying a fee. That's why it goes by so many names.
Apple killed firewire -- nobody needed to help.
Recursive: Adj. See Recursive.
One thing should be noted also, since the posting is somewhat unclear on this topic.
The suggestion is not that the group who received the award has now closed off some specs. See #15 for example.
The suggestion is now that this USB group got $10k to use, the specs suddenly cosy money. (See post from Alan Cox).
--
It should be noted that many other operating systems also currently support USB as well. BSD's not with standing. I think it should be opened for EVERYONE not just Linux.
What makes Linux more special than the rest? And why should Microsoft have to pay instead of Linux? Oh, being not-for-profit is entirely false. Ask Redhat how much their stock is worth.
Not to mention once Linux, or a BSD used the standards in opensource, other places would just take the API calls from there.
Whats the object of this message? Stop fighting a Linux vs. the world battle, and fight an opensource vs. the closed source world battle. You might find you have better luck as the number of opensource people is greater than linux people. Especially as far as large corporations are concerned.
Rod Taylor
Before they fix their security leaks, I've mirrored the members only pages at
http://www.geocities.com/usb_dev/.
Feel free to mirror this...
Additionally, another AC has posted a mirror where the docs can be viewed (scroll down if you're interested). [I have personally been fucking with mixmaster trying to send off some mails, damnit] Of particular interest are two particular documents: "Device Class Definition for Content Security Devices 0.9a" and "Content Security Method 4 - Elliptic Curve Content".
The first of these docs outlines roughly why: "Protected Content typically refers to copyrighted content. Content Protection Methods (CPM) have been developed for the controlled distribution of protect [sic] content." and also uses an example of a pair of USB speakers which allow protection of content up until the hardware induces analog playback.
The second document discusses the technical details of Elliptic Curve cryptographic methods to provide authentication, and authentication + encryption. This is no CSS. This is good old fashioned high grade badass crypto. Some of the inline commentary regarding illegal intent there is somewhat chilling.
Anyone still have doubts as to why this dinner is being held behind closed doors and costs $2500 a plate?
I am deisgning a USB device in VHDL and I downloaded some of the Device Class specifications previously - it sucks that the otherwise very open (there are no licensing fees to create USB devices, and you can download the protocol specification from the website for free; they also run a very active and useful Developers Webboard for free) USB site should clam up on the critically important Device Class specs like this.
The Device Class specifications outline standard ways to use the USB protocol for, say, Modem devices, or Mass Storage Devices. If you make your hardware compliant with these Device Class specification ''APIs'' then you don't actually need to write a device driver! The major OSs write inbuilt support for the defined Device Classes, and your new Modem or whatever is recognized as being compliant with the ''APIs'' and ''just works''.
It is a bit numb to publish the physical and transport layer specifications for free but not the higher level ones! I wonder what made them decide to change?
Anyone concerned should email admin@usb.org (this is the correct address from the website) and request that the Linux implementation efforts are allowed a complimentary membership as they are not-for-profit.
-Andy
Frankly a year charge of $2500 can only affect individuals or volunteer groups. Companies and major software development groups may feel this as a bless as it shrinks USB potential market to their products.
Sincerly I am not absolutely against charging such things. Anyway things must be supported, one should pay for common expenses and many people would like to have conferences or meetings to settle up questions. and this demands money.
However USB group makes out of this an elite club. There are no chances. Either pay $2500 or hit the streets. And this does not just end here. If anyone has taken a look at their application form, then they consider ONLY companies as possible members. Besides:
"Renewal and Size of Forum. The Forum Sponsors reserve the right to limit the number of participants or to discontinue the program upon written notice. Upon any such, Company shall receive a refund of a portion of its subscription fee proportional to the amount of time otherwise remaining in its enrollment period."
The Forum sponsors are: Compaq, Digital Equipment Corporation, IBM PC Co., Intel, Microsoft, NEC, and Northern Telecom. So one may well think were this beautiful family of sponsors are pushing us to. Specially if we consider that this membership allows "early access to specifications".
In any case I would highly recomend anyone to read this membership form. It presents some more interesting points that may develope discussion:
http://www.usb.org/developers/data/usbifapp.pdf
It seems to me that the rampant mirroring some people are suggesting isn't going to be particularly effective, in the long term at least.
These specifications are 'in various stages of development' and the folks working on this stuff would probably like to keep up with these developments, not be stuck with what was freely available yesterday.
This needs to be sorted out sensibly and diplomatically, hijacking accounts and mirroring (potentially superceded) documentation is going to win us no friends nor be of much long term use. What it will do is make us look like a bunch of petulant children.
Boffoonery - downloadable Comedy Benefit for Bletchley Park
I've read all the comments on this article, and visited a few links. There's quite a few rumors and half-truths going around, so I won't try to address most of them. I will, however, address the simple act of trying to close off the device classes.
If we look back on the path computing has taken over the past two decades, we see it littered with dead technology. What does most of this dead technology have in common? At one point or another, it was closed off. Ridiculous fees were charged for redistribution of specifications. Those who paid were prohibited from sharing their information with anyone but the others who paid, and sometimes not even them.
By contrast, if the steward of a particular technology practices open computing, whether the technology lives or dies depends on technological rather than political (and no, I don't mean governmental) merits. If these stewards need money; it would stand to reason that whomever stood to benefit from the technology's wide adoption should contribute more to it in the hopes that their investments would be returned.
USB's move here could undermine its previously open stance, and that would be a bad thing for USB. I hope they reevaluate it. USB is a very useful technology in its own right and need not be squelched in this manner.
Oh, and I can't pass up contributing to a USB thread without telling you that salespeople in a local electronics store recently told one of my friends that the USB port on some home entertainment equipment was for the brand-spankin'-new ``Universal Stereo Bus''. :-)
- http://www.usb.org/developers/docs.html
- http://electr icrain.com/lists/archive/linux-usb/2000/02/msg010
3 5.html - http://electr icrain.com/lists/archive/linux-usb/2000/02/msg010
5 5.html
Methinks the editor should have done some research himself and put some of these in the story. A slashdot story with no links? Bah!--
Although it appears people are confusing the group that got the award (developers), and the people who are charging for access to docs (USB-IF), perhaps it's time to split the Beanie Awards concept in twain.
- Slashdot Beanie Awards
- Slashdot Weenie Awards
(Slashdot Weenie Awards =anagram> So, what ideals answered?)For laudable service to fellow geeks.
For all the FUD-spewing, patent abusing sorts.
[
I've contacted the individual authors of the documents being restricted, and most of them appear to agree with my points. I think we can expect this situation to get fixed with either usb.org putting it back into a publically accessible spot or the same documents being published elsewhere.
This message is provided under the terms outlined at http://www.bero.org/terms.html
That was the net/net of this situation so if you are concerned with getting at the information there is nothing further to worry about. If you are interested in the background, I will try to explain.
The USB-IF charter has always been to support the development of USB products. It has also always included the FOCUS on members and anything that was made available to the general public was fine as long as it didn't become a support or financial burden. This is based on the belief that USB technology is close enough to "rocket science" that most all individual companies will have trouble if they don't have support from the spec architects in the form of education, information centralization, market message commonality, a compliance program.... These things just ain't free. Ergo the membership organization does most of these things for its members and charges an annual membership fee to cover the cost.
USB-IF is a non profit corporation that spends over 20% of its budget on web information distribution, over %30 percent of its budget on a compliance program, slightly under %20 of its budget on administration. Whats left is now destined for on of the most often requested services from it's members, an improved compliance and logo program. The members feedback has been for a long time (this is an expensive change and it took a lot of effort to make it) that we should have a better way to stop the slipshod manufacturing and design of things that call themselves USB products. I don't know how this will work out, but we are doing our best to accomplish what the members are asking within our budget.
Now the place where this community can help!! As part of our compliance testing we verify product electrical characteristics and protocol capabilities. When it comes to the qualification of drivers and therefore the statement that a product works with a certain OS we only have one set of tests that we can use. They were developed and given to the IF for this use by the promotor companies (yes Microsoft is a promotor company). There were some big costs to develop these tests, by the way, that were covered by the promotor companies themselves. Currently for any other OS we have to simply take the peripheral vendors word for their drivers compliance to the OS. If the Linux community can find a way to judge driver quality for USB products the OS column for Linux on the USB-IF product list would mean much more. We are open to a proposal that would help seperate the functional products from the non functional.
PLEASE NOTE;Our compliance program is primarily a feedback tool to developers. We get more repeat attendees at workshops that come back even after their products have "passed" because the workshops allow them to learn so much and develop so much easier or quicker. As most of us heard in high school or college, "the score is not the goal of the test, it is the feedback on the learning that is of real value".
RSS