You're correct Bruce, I'm off base this time. I got contacted by the writer this morning who told me that the SFLC had told him that a fixed cap would work with GPLv2. So being in the middle of coding something (ie. not paying enough attention), and remembering the fixed price we paid to get access to the EU Workgroup Server docs, I just agreed that it sounded like this would be a work-around for v2, but not for v3 where section 11 is much stricter about patent licensing (explicitly the bits about extending the license downstream), and bingo - there goes the story with the quote. You know how these things go:-(. My fault, and I'll be more careful in future.
the devil is in the details. Someone just mailed me a comprehensive analysis and agreeing to this license, even with a royalty cap, would violate GPLv2 in several ways.
There is a field of use restriction : "Pricing for other device types can be negotiated with Microsoft."
Modification restrictions: "devices are fully compliant with certain required portions of the Microsoft FAT file system specification"
and a per-manufacturer limit: "a cap on total royalties of $250,000 per manufacturer".
So yes, I got it wrong and this license is in no way GPLv2 compatible.
Sorry for the mistake. Blame me, not the journalist who was just trying to get his story.
Remember I used to work for HP. The patent cross licensing deals done by companies like HP and IBM (don't know about Sun) explicitly exclude free software/open source use. They have very sophisticated legal departments that know the GPL *very* well. Their patent cross licensing doesn't include GPL code.
A bitter, abusive denunciation. [Latin diatriba, learned discourse, from Greek diatrib, pastime, lecture, from diatrbein, to consume, wear away : dia-, intensive pref.; see dia- + trbein, to rub; see ter-1 in Indo-European roots.] Word History: Listening to a lengthy diatribe may seem like a waste of time, an attitude for which there is some etymological justification. The Greek word diatrib, the ultimate source of our word, is derived from the verb diatrbein, made up of the prefix dia-, "completely," and trbein, "to rub," "to wear away, spend, or waste time," "to be busy." The verb diatrbein meant "to rub hard," "to spend or waste time," and the noun diatrib meant "wearing away of time, amusement, serious occupation, study," as well as "discourse, short ethical treatise or lecture, debate, argument." It is the serious occupation of time in discourse, lecture, and debate that gave us the first use of diatribe recorded in English (1581), in the now archaic sense "discourse, critical dissertation." The critical element of this kind of diatribe must often have been uppermost, explaining the origin of the current sense of diatribe, "a bitter criticism."
Again, your usage of the word is "now archaic".
Please learn the current use of the language before engaging in discourse with your betters, else learn to tug your forelock appropriately:-).
"The problem here is that the GPL sets up restrictions which no intermediate vendor can realistically comply with" should get you marked down immediately as -1 Troll".
Or at least as -1 clueless. Do you know how many intermediate vendors ship GPL code, both v2 and v3 ? It's a *lot*. You can even get patent cross licenses for all your other code so long as the patents you are licensing don't cover the GPLed code. Please post your ignorant long diatribes elsewhere.
Thanks dude. I guess I should take my complements where I can:-).
We are moving to a registry based config in later versions, but I'm not sure you would think that an improvement:-).
You have to remember Samba is 17 years old, and you can still parse original smb.conf files from the first version. These things do tend to acrete over time, and it's hard to break existing configs. Not an excuse, but......:-).
Sun CIFS isn't reimplemented from scratch by Sun, it was code they got from their Procom acquisition. It remains to be seen if putting a CIFS server into an otherwise stable kernel is a good idea or not:-).
> SHARE the SMB password system, make it available, so not every friggin windows machine has to do unencrypted passwords across the network to access SAMBA shares / > printers / whatever.
You're talking about a Samba PDC. That uses old NT4 technology, not AD. But as a member server we support AD completely. In fact the current Likewise code is based off winbindd (part of Samba). Jerry Carter, one of our release managers works for Likewise and supports it. It's open source too (at least the low end version is).
FYI: Channel #samba isn't visited by the Samba developers. #samba-technical is where we hang out. We don't use words like "wintendo" there. But then again we don't offer a lot of end-user help (mainly discussions on the code) so it might not be the right place for you.
I did point out that no other POSIX system behaved like that, but that didn't seem to make much difference:-). Eventually I just added a parameter that allowed our open directory cache to be turned off on *BSD. Once it got into the hands of Marc Balmer he took us seriously and fixed the bug.
They've already released the protocol info publicly. The catch isn't the protocol documentation, it's the patent license they claim is needed to implement the protocol It's moving the control point from trade secrets to patents. But I'm still really happy with the release of the info to the public. The docs aren't perfect yet, but they're going to be improved. It's a very positive step.
I loved that song. I keep trying to think of a way to make it the "musical" question in the Golden Penguin bowl. But it's too long and too obscene to work:-).
Cheers Bruce and FYI: I fully support your candidacy.
> Keep in mind that what we (and sun, and philips) needed to provide when bidding for contracts would have been compliance documentation, not down to the byte documentation of internal APIs, memory structures and file formats that were never > intended to be used by any third party.
If it's accessible from an on the wire protocol, then trust me it's available to be used by a third party. Not in ways you might like, but it's *definitely* available:-).
They way it will work is as follows. We'll read the docs and work on creating client-side test cases and embedding them into Samba4 smbtorture. Once that's in place, any competent engineer can create the server-side implementation without having to have access to the actual docs. We need the test cases anyway (remember, untested code is broken code), so this is the way we've been going about doing things anyway. This should just open up new protocols and new protocol areas to implementation by others.
Slashdot Mirror
You're correct Bruce, I'm off base this time. I got contacted by the writer this morning who told me that the SFLC had told him that a fixed cap would work with GPLv2. So being in the middle of coding something (ie. not paying enough attention), and remembering the fixed price we paid to get access to the EU Workgroup Server docs, I just agreed that it sounded like this would be a work-around for v2, but not for v3 where section 11 is much stricter about patent licensing (explicitly the bits about extending the license downstream), and bingo - there goes the story with the quote. You know how these things go :-(. My fault, and I'll be more careful in future.
Looking closely at the license here:
http://web.archive.org/web/20060207034921/http://www.microsoft.com/mscorp/ip/tech/fat.asp
the devil is in the details. Someone just mailed me a comprehensive analysis and agreeing to this license, even with a royalty cap, would violate GPLv2 in several ways.
There is a field of use restriction : "Pricing for other device types can be negotiated with Microsoft."
Modification restrictions: "devices are fully compliant with certain required portions of the Microsoft FAT file system specification"
and a per-manufacturer limit: "a cap on total royalties of $250,000 per manufacturer".
So yes, I got it wrong and this license is in no way GPLv2 compatible.
Sorry for the mistake. Blame me, not the journalist who was just trying to get his story.
Jeremy.
Remember I used to work for HP. The patent cross licensing deals done by companies like HP and IBM (don't know about Sun) explicitly exclude free software/open source use. They have very sophisticated legal departments that know the GPL *very* well. Their patent cross licensing doesn't include GPL code.
Jeremy.
I take great care with my language, which is why I'm bothering to respond, troll though you are (as I correctly categorized you :-).
From Mirriam-webster online:
http://www.merriam-webster.com/dictionary/diatribe
1 archaic : a prolonged discourse
2: a bitter and abusive speech or writing
3: ironic or satirical criticism
Note, your usage is flagged "archaic", much like your anti-GPL tirades :-)
More information (if such were needed).
http://www.thefreedictionary.com/diatribe
A bitter, abusive denunciation.
[Latin diatriba, learned discourse, from Greek diatrib, pastime, lecture, from diatrbein, to consume, wear away : dia-, intensive pref.; see dia- + trbein, to rub; see ter-1 in Indo-European roots.]
Word History: Listening to a lengthy diatribe may seem like a waste of time, an attitude for which there is some etymological justification. The Greek word diatrib, the ultimate source of our word, is derived from the verb diatrbein, made up of the prefix dia-, "completely," and trbein, "to rub," "to wear away, spend, or waste time," "to be busy." The verb diatrbein meant "to rub hard," "to spend or waste time," and the noun diatrib meant "wearing away of time, amusement, serious occupation, study," as well as "discourse, short ethical treatise or lecture, debate, argument." It is the serious occupation of time in discourse, lecture, and debate that gave us the first use of diatribe recorded in English (1581), in the now archaic sense "discourse, critical dissertation." The critical element of this kind of diatribe must often have been uppermost, explaining the origin of the current sense of diatribe, "a bitter criticism."
Again, your usage of the word is "now archaic".
Please learn the current use of the language before engaging in discourse with your betters, else learn to tug your forelock appropriately :-).
Jeremy.
Hmmm. That's not what the dictionary says. Let me check....
http://dictionary.reference.com/browse/diatribe
"a bitter, sharply abusive denunciation, attack, or criticism:"
Yep, that shoe fits. Wear it. Ignorant diatribe indeed :-).
Jeremy.
This :
"The problem here is that the GPL sets up restrictions which no intermediate vendor can realistically comply with" should get you marked down immediately as -1 Troll".
Or at least as -1 clueless. Do you know how many intermediate vendors ship GPL code, both v2 and v3 ? It's a *lot*. You can even get patent cross licenses for all your other code so long as the patents you are licensing don't cover the GPLed code. Please post your ignorant long diatribes elsewhere.
Jeremy.
Thanks dude. I guess I should take my complements where I can :-).
We are moving to a registry based config in later versions, but I'm not sure you would think that an improvement :-).
You have to remember Samba is 17 years old, and you can still parse original smb.conf files from the first version. These things do tend to acrete over time, and it's hard to break existing configs. Not an excuse, but...... :-).
Jeremy.
Sun CIFS isn't reimplemented from scratch by Sun, it was code they got from their Procom acquisition. It remains to be seen if putting a CIFS server into an otherwise stable kernel is a good idea or not :-).
Jeremy.
"and Andrew just had to leave Novel in a storm to push the idea"
No, that would be me, not Andrew Bartlett. Andrew has been happily working at Red Hat for many years now.
Jeremy.
> SHARE the SMB password system, make it available, so not every friggin windows machine has to do unencrypted passwords across the network to access SAMBA shares /
> printers / whatever.
Troll. This hasn't been the case since Samba 1.x.
Jeremy.
Sigh. The misinformation level in these threads is amazing :-(.
Yes, Win9x does encrypted passwords. They're not very good (lanman) but it is an encrypted password authentication.
Jeremy.
Read this :
http://www.samba.org/samba/PFIF/PFIF_agreement.html
for details on patent issues. It's not as black as you paint it.
Jeremy.
> As for contaminated IP, I would submit that Samba/CIFS and WINE have far more IP (patent) liability than mono does
Yes, but you would be wrong about that.
Jeremy.
You're talking about a Samba PDC. That uses old NT4 technology, not AD. But as a member server we support AD completely. In fact the current Likewise code is based off winbindd (part of Samba).
Jerry Carter, one of our release managers works for Likewise and supports it. It's open source too (at least the low end version is).
Jeremy.
No, Samba doesn't have a 2GB file limit. Samba is fully 64-bit clean.
Jeremy.
FYI: Channel #samba isn't visited by the Samba developers. #samba-technical is where we hang out. We don't use words like "wintendo" there. But then again we don't offer a lot of end-user help (mainly discussions on the code) so it might not be the right place for you.
Jeremy.
There was no exploit known in the wild before this was discovered and patched, so if you install the Debian patch asap you should be fine.
Jeremy.
No, we did report it. The answer at the time was "this is allowed by POSIX, deal with it", can be seen in the bug report here :
:-). Eventually I just added a parameter that allowed our open directory cache to be turned off on *BSD. Once it got into the hands of Marc Balmer he took us seriously and fixed the bug.
https://bugzilla.samba.org/show_bug.cgi?id=4715
I did point out that no other POSIX system behaved like that, but that didn't seem to make much difference
Jeremy.
Yeah, that's the point really. It's *fun* :-). And also everyone is being very nice :-).
Jeremy.
They've already released the protocol info publicly. The catch isn't the protocol documentation, it's the patent license they claim is needed to implement the protocol It's moving the control point from trade secrets to patents. But I'm still really happy with the release of the info to the public. The docs aren't perfect yet, but they're going to be improved. It's a very positive step.
Jeremy.
The slides will be posted at the conference website, but they're not up yet. Not sure yet about audio/video.
Jeremy.
I loved that song. I keep trying to think of a way to make it the "musical" question in the Golden Penguin bowl. :-).
But it's too long and too obscene to work
Cheers Bruce and FYI: I fully support your candidacy.
Jeremy.
> Keep in mind that what we (and sun, and philips) needed to provide when bidding for contracts would have been compliance documentation, not down to the byte documentation of internal APIs, memory structures and file formats that were never
:-).
> intended to be used by any third party.
If it's accessible from an on the wire protocol, then trust me it's available to be used by a third party. Not in ways you might like, but it's *definitely* available
Jeremy.
It's not true :
http://talkback.zdnet.com/5208-12558-0.html?forumID=1&threadID=43385&messageID=803908
Jeremy.
But I *loved* the goatse trolls :-). It used to be the only reason I /. :-). I miss sig11 and klerk and the rest of that crew.
/. fun, especially as they drove taco *nuts* :-).
:-).
came to read
They made
Anyone remember the Bruce Perens impersonators ?
Jeremy.
They way it will work is as follows. We'll read the docs and work on creating client-side test cases and embedding them into Samba4 smbtorture. Once that's in place, any competent engineer can create the server-side implementation without having to have access to the actual docs. We need the test cases anyway (remember, untested code is broken code), so this is the way we've been going about doing things anyway. This should just open up new protocols and new protocol areas to implementation by others.
Jeremy.