The whole business of software patents is very likely to explode at any time. There are several software patents on cds and dvds which are so vague that it is impossible to tell exactly what is going on. The licenses to use these patents allow the company issuing the patent to control cd content and to examine it. The first time one of these companies tries to use the license to access DOD classified information there will be a resolution. The easiest way to resolve this business with Borland is to contact Borland. They may also have an interest in Wine. If they have no objection to what Wine wants to do then there is no one else to complain. They might even help Wine. Borland probably doesn't even realize that Wine exists although some of the people in Borland are probably very familiar with it. I think you will find this to be true of all big software companies such as IBM and Microsoft.
gslink wrote:
Borland probably doesn't even realize that Wine exists although some of the people in Borland are probably very familiar with it. I think you will find this to be true of all big software companies such as IBM and Microsoft.
IRC borland had a delphi for linux named kylix which if I remember correctly used wine. They had 2 sets of widgets, native and qt classes to ease development and porting. Here is the article. It also mentions codeweavers... Really old news.
http://open.itworld.com/4907/IW010305tckylix/page_1.html
So it's not possible that they have never heard about wine before...
.Bill
gslink wrote:
The easiest way to resolve this business with Borland is to contact Borland. They may also have an interest in Wine. If they have no objection to what Wine wants to do then there is no one else to complain. They might even help Wine. Borland probably doesn't even realize that Wine exists although some of the people in Borland are probably very familiar with it.
Havn't they even used Wine to port Kylix? ...let's hope they don't blame Wine for Kylix' failure. ;)
-flx
On Thu, May 12, 2005 at 08:04:57AM -0400, gslink wrote:
The easiest way to resolve this business with Borland is to contact Borland. They may also have an interest in Wine. If they have no objection to what Wine wants to do then there is no one else to complain. They might even help Wine. Borland probably doesn't even
Borland may have some intrest in Wine (it can be used to run code produced by their compilers, which helps their potential market-share a bit). However, it could also view Wine as a competitor (especially to their Kylix product). I doubt they'd sue Wine developers over this, but it's possible that they have a reciprocal license that would require them to defend patent #5,628,016.
Would Wine be able to proceed if we got the following statement from them?
"Borland hereby grants a worldwide, royalty-free, non-exclusive license to use the methods described in U.S. patent 5,628,016 to the current Wine developers when they use the current LGPL source-code for wine, and to anyone who properly recieves a copy of that source-code or properly creates a derivative work based on it when they use said copy or derivative work."
I think this would satisfy the conditions in section 11 of the LGPL. In absence of communication from Borland, we might also infer this based on their future public statements.
I just don't want to see this turn into another situation like GIF files or PGP 2.x, where one company's patent put all "compatible" implementations on shaky legal ground.
David Lee Lambert wrote:
Would Wine be able to proceed if we got the following statement from them?
"Borland hereby grants a worldwide, royalty-free, non-exclusive license to use the methods described in U.S. patent 5,628,016 to the current Wine developers when they use the current LGPL source-code for wine, and to anyone who properly recieves a copy of that source-code or properly creates a derivative work based on it when they use said copy or derivative work."
I think this would satisfy the conditions in section 11 of the LGPL. In absence of communication from Borland, we might also infer this based on their future public statements.
I'm not a laywer, and I don't really want to start a debate on this, but my opinion is:
It would need to be granted to gcc, not Wine, since gcc would be generating the exception handling code. To be properly compatible with the GPL, I'd guess it would have to be granted to any project that wished to use it with GPL licensed software, not just gcc.
Mike
Its highly likely that GCC and WINE are already infringing on some software patent somewhere (since its well nigh impossible not to in the current "patent everything you can" climate inside a number of big companies)
What makes this particular borland patent any different?
Jonathan Wilson wrote:
Its highly likely that GCC and WINE are already infringing on some software patent somewhere (since its well nigh impossible not to in the current "patent everything you can" climate inside a number of big companies)
What makes this particular borland patent any different?
My guess: many projects have a "don't tell us about patents, we don't wanna know"-policy to plead ignorance in case they get sued (would this work anyway?) - this one is known (now).
Did people actually try to get the code into gcc (without saying "unfortunatly this patch infrings on patent #XXX") before?
-flexo
Jonathan Wilson wrote:
Its highly likely that GCC and WINE are already infringing on some software patent somewhere (since its well nigh impossible not to in the current "patent everything you can" climate inside a number of big companies)
What makes this particular borland patent any different?
Borland has more money than Wine. There are several groups besides Microsoft and IBM that would benefit from not having to write and maintain an extra codebase to run their product on Linux. Wine needs to partner with some of these people both for money and also for development help. Wine has gotten to the point where it is becoming a very excellent piece of code that looks rough. From past experience this means it is about ready for big time use with all the problems that entails.
On Thu, 12 May 2005, gslink wrote:
The whole business of software patents is very likely to explode at any time.
I assume this is somewhat related to the Winelib article on Slashdot:
Winelib Hobbled by Exception-Handling Patent http://yro.slashdot.org/yro/05/05/12/1947213.shtml
If not I'll make just a few comments about that article anyway.
* The Slashdot summary makes it sound like Wine is infringing on the patent and that we will have to remove some code. This is absolutely not the case. The infringing piece of code does not belong in Wine so noone ever even tried to put it there. Exception handling belongs in gcc and some patches exist to add this feature to gcc. But obviously the gcc developers (who are pretty strict about this stuff) never allowed this code to get into the gcc codebase. This means Wine cannot assume it's available or use it.
* Some posters on Slashdot seem to think it's a recent discovery. It's not. This issue was discussed more than a year ago. I'll let interested parties dig the wine-devel archives for the relevant threads. This also means there's no need to worry about 'plausible ignorance of th issue', it's been discussed already so the cat's out of the bag anyway.
* Some posters on Slashdot seem to think this 'discovery' is related to the recent news about Eben Moglen helping Wine. It's obviously not the case since this issue was known long before the agreement with Eben was even in the making.
On Friday 13 May 2005 15:09, Francois Gouget wrote:
- The Slashdot summary makes it sound like Wine is infringing on the
patent and that we will have to remove some code.
Wine duplicates many of the data structures in Windows which are described in the patent. Code using the C syntax described in the patent doesn't compile in winelib yet, but this has nothing to do with the patent. Give me a week or two...
I don't think you can implement a Windows compatible environment without the SEH datastructures described in the patent. IANAL, but saying "windows compatibility is patented by Borland" is just silly. I, for one, am not worried about it.
On Friday 13 May 2005 19:07, Gregory M. Turner wrote:
Code using the C syntax described in the patent doesn't compile in winelib yet [snip]. Give me a week or two...
Imagine that, it seems I have once again imagined myself to have more spare time than I really do. Honestly, folks, I can't wait to lay into __TRY & family -- especially since that work brings me closer to __try & family for winelib, which, as far as I'm concerned, is unfinished business -- but I simply am not in a position to do anything about it until I finish my dissertation and take care of some other pressing matters.
Regardless, I am still alive and still plan to do this, sooner than later. In the meanwhile, I'm glad to see that the rest of you are being better wine hackers than I!
"I'll be back,"
-
David Lee Lambert -
Felix Nawothnig -
Francois Gouget -
Gregory M. Turner -
gslink -
Jonathan Wilson -
Mike McCormack -
Vassilis Virvilis