RAND !== F/OSS compatible [LWN.net]

RAND !== F/OSS compatible

Posted Feb 21, 2008 17:23 UTC (Thu) by leoc (subscriber, #39773)
In reply to: RAND !== F/OSS compatible by kirkengaard
Parent article: Microsoft announces changes to promote interoperability

It also says:

Microsoft is providing a covenant not to sue open source developers for development or non-commercial distribution of implementations of these protocols. These developers will be able to use the documentation for free to develop products. Companies that engage in commercial distribution of these protocol implementations will be able to obtain a patent license from Microsoft, as will enterprises that obtain these implementations from a distributor that does not have such a patent license.

I hate to sound optimistic, but this appears to be good news all around.

(Log in to post comments)

RAND !== F/OSS compatible

Posted Feb 21, 2008 17:41 UTC (Thu) by proski (subscriber, #104) [Link]

You must be sarcastic.

RAND !== F/OSS compatible

Posted Feb 21, 2008 17:43 UTC (Thu) by JoeBuck (subscriber, #2330) [Link]

No, it isn't good news all around.

The terms you quote are incompatible with the GPL, both versions 2 and 3.

From GPL v2 (where I flagged the relevant clause with italics):

7. If, as a consequence of a court judgment or allegation of patent infringement or for any other reason (not limited to patent issues), conditions are imposed on you (whether by court order, agreement or otherwise) that contradict the conditions of this License, they do not excuse you from the conditions of this License. If you cannot distribute so as to satisfy simultaneously your obligations under this License and any other pertinent obligations, then as a consequence you may not distribute the Program at all. For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.

And it's not just the GPL: the idea that any commercial distributor would need a patent license contradicts the DFSG (and therefore the OSI definition of open source).

RAND !== F/OSS compatible

Posted Feb 21, 2008 18:05 UTC (Thu) by drag (subscriber, #31333) [Link]

Most of Microsoft's patents should be trivial to code around. Like any other program there is
a ten thousand and one ways to skin the cat.

For stuff that can't be coded around all you need to have is for a program to have a exception
in it's license allowing you to use those specific plug-ins that would be made compatible with
Microsoft's patented API requirements thus working around the issue in a legal fashion.

This is morally acceptable because.. one; the person using the plugin stuff would already be
running a lot of proprietary code in the form of Microsoft software with it's own costs and
issues so it won't be adding any additional restrictions on those people that didn't already
exist.. and two; because that sort of thing will be mostly used by people trying to gradually
migrate away from proprietary software or introduce more free software into their businesses.

RAND !== F/OSS compatible

Posted Feb 21, 2008 21:07 UTC (Thu) by roc (guest, #30627) [Link]

It may not be easy to code around Microsoft's patents when they are determining the protocols,
and hence the detailed functionality, that your software must implement.

RAND !== F/OSS compatible

Posted Feb 21, 2008 23:54 UTC (Thu) by drag (subscriber, #31333) [Link]

Samba has done that. So you have precedence. At least thats what they are saying they are
doing and that most Microsoft patents are not that good.

Of course for patents for protocols that you can't code around you'll have to setup some sort
of proxy, translator, emulation, or plugin system that is under a compatible license that can
be a buffer between OSS and Microsoft's patents.

RAND !== F/OSS compatible

Posted Feb 22, 2008 6:21 UTC (Fri) by ikm (subscriber, #493) [Link]

It feels that 'patents that cover protocols' are supposed to cover the protocols themselves,
and as such, it should be impossible to work around them without deviating from the protocols
by definition. An example of such a patent would probably show the real deal here.

RAND !== F/OSS compatible

Posted Feb 22, 2008 15:01 UTC (Fri) by smitty_one_each (subscriber, #28989) [Link]

>This is morally acceptable 
What's badly needed is an overarching theory of 'moral'.
I track the GPL well in a common-sense mode, and for that reason wouldn't touch MS's license
cocaine with a 10-foot wookie.
Arguments based upon 'moral' get squishy very quickly, and lead to much strife, ironically.

RAND !== F/OSS compatible

Posted Feb 22, 2008 15:38 UTC (Fri) by drag (subscriber, #31333) [Link]

It's because it's human concepts and ideals that dictate morals and ethics and those change
based on who you are talking to. My moral code is obviously going to be different then your's
and most everybody else's. In life there are very very few things that do not quickly become
'squishy'.

It's probably why so many people here like programming.. It's something that has absolution.
There is a 'right' way and a 'wrong' way. When it is 'right' it works, and when it does not
then it's wrong. It's very simple, it's testable, and generally there is a hard line on
correctness. 

There are very very few other areas in society that can be that.. obvious between right and
wrong. When ever software touches those other parts of human life (business, law, human
interfaces) then it can't help but become 'squishy'.


In my eyes extracting the maximum amount of freedom while working around the limitations of
living in a modern government-controlled society is a very good thing that a software project
can do. Except when it becomes appropriate to break those laws in acts of civil disobedience
which then is 'more ethical'.

So providing MS-patent-compatible open software (as open as is allowed by MS) that is designed
with the expressed goal of allowing people to use more truly free software in a practical
real-world environment is a good thing as long as that sort of software is kept to a minimum.
Of course this is only good approach when those patents can't be otherwise worked around.

RAND !== F/OSS compatible

Posted Feb 21, 2008 20:15 UTC (Thu) by leoc (subscriber, #39773) [Link]

You are totally correct, and I hereby rescind my expression of all around goodness, and would
like to apologize to everyone for being way too optimistic.

RAND !== F/OSS compatible

Posted Feb 21, 2008 17:47 UTC (Thu) by thoffman (subscriber, #3063) [Link]

I would like to be optimistic, but decades of history of Microsoft's actions make me cynical.
The problem is, what does "non-commercial distribution" mean?

I suspect in practice it will mean that Microsoft will want payment for every installation of
(just for example) Red Hat Enterprise, as that is a "commercial distribution".  

Microsoft might even define "commercial distribution" to cover Fedora and Ubuntu, since Red
Hat and Canonical both are for-profit companies.

In short, the "covenant not to sue" seems very suspicious unless it covers all open source
code regardless of (a) how it is distributed, and (b) who is "behind" the distribution.

RAND !== F/OSS compatible appears to be good news ...

Posted Feb 21, 2008 18:03 UTC (Thu) by TxtEdMacs (subscriber, #5983) [Link]

"... engage in commercial distribution ..."

That would include Red Hat, Ubuntu, [whatever Mandrake calls itself now], etc. Essentially, all those that refused to buy into the previous promise not to be sued. Sorry, same drivel in new look packaging.

What's the cliché on slashdot? "Nothing new here, move on!"

RAND !== F/OSS compatible

Posted Feb 21, 2008 19:11 UTC (Thu) by einstein (subscriber, #2052) [Link]

Right - microsoft will agree not to sue the hobbyist working in isolation, but they will
expect and insist on royalty payments for any attempt to use this for anything meaningful.

Hard to see how this could be interpreted as anything but a cynical and self serving gesture
from microsoft.

Well, it's significantly better than nothing

Posted Feb 21, 2008 19:33 UTC (Thu) by sepreece (subscriber, #19270) [Link]

While the terms won't satisfy everybody's requirements, they're clearly ad advance over what
Microsoft permitted/provided before.

And the part about declaring what patents bear on what APIs/protocols seems like a very useful
bit of information to give the community, since it tells you just what you need to work
around.

scott