Bug#381263: gyachi 1.1-61-1

2009-04-04 Thread Greg Hosler


--- On Thu, 4/2/09, Eddy Petrișor  wrote:

> From: Eddy Petrișor 
> Subject: Re: Bug#381263: gyachi 1.1-61-1
> To: "Greg Hosler" 
> Date: Thursday, April 2, 2009, 5:58 PM
> 2009/4/2 Greg Hosler :

> > The md5.* files DO have copyright, and I don't see
> these files being in conflict w/ gpl.
> 
> Indeed, it is some sort of BSD-like lincese.

I have re-looked at the md5 files (md5.[ch])

I do not see the license of the md5 files conflicting w/ the GPL license, and 
I'm not inclined to replace these files unless someone can point out me the 
conflict that needs resolving.

:)

The license follows.:

  This software is provided 'as-is', without any express or implied
  warranty.  In no event will the authors be held liable for any damages
  arising from the use of this software.

  Permission is granted to anyone to use this software for any purpose,
  including commercial applications, and to alter it and redistribute it
  freely, subject to the following restrictions:

  1. The origin of this software must not be misrepresented; you must not
 claim that you wrote the original software. If you use this software
 in a product, an acknowledgment in the product documentation would be
 appreciated but is not required.
  2. Altered source versions must be plainly marked as such, and must not be
 misrepresented as being the original software.
  3. This notice may not be removed or altered from any source distribution.

--

I'm open to replacing the files, but not for the sake of replacing, and not 
"just because". Show me the conflict, and I'll replace them.

All the best,

-Greg

 


  

Bug#381263: gyachi 1.1-61-1

2009-04-03 Thread Greg Hosler

Sorry about the poor formatting. ymail just plain sucks in the reply dept.

I'll tag my replys with "**"

--- On Thu, 4/2/09, Eddy Petrișor  wrote:

From: Eddy Petrișor 
Subject: Re: Bug#381263: gyachi 1.1-61-1
To: "Greg Hosler" 
Date: Thursday, April 2, 2009, 5:58 PM

2009/4/2 Greg Hosler :
>
> Got your file list. Thanks. It's really not as bad as it looks, and I can
> fix most fairly easily and fairly quickly.

Great. You'd probably want to foward this info t the Debian BTS (just
CC the mail to 381...@bugs.debian.org).
I am not cutting anything from this reply to make this easy for you.

> All the gyvoice files are from an earlier version of wine. not
> sure why the copyright was yanked. actually I have some guesses.
> the code necessary for gyachi was a small subset of wine, and the
> files were just hacked, removing anything not explicitly necessary.
> In the process of the hacking, that's almost certainly where the
> copyright (and a lot of other extraneous stuff) got dropped. Adding
> back in the respective copyrights is probably doable.

Or is possible that wine didn't had those at that time. Can't you use
some library instead of that embedded copy of the loader?

** Wine almost certainly had copyrights at the time. :)

BTW, is gyachi usable without any windows dll files? I see that there are no 
dll files in the source, but a wine interface, why is it needed?

** GyachI needed a way to load and callback a dll. Wine
** does a WHOLE LOT more than that. In fact, last year
** i looked at replacing the gyvoice directory with winelib
** calls. The problem is that wine *is* an emulator, DESIGNED
** for running window exe's (as well as for loading dlls).
** GyachE only needed the DLL loading capability, and so the
** necessary wine suport was *GREATLY* hacked to eliminate as much
** of wine as possible. What's left is stubs of the original
** (except the dll loading portions).

** This, by the way, is what other codec players have ALL done (as
** pointed out to me by people on the wine-devel list. Projects
** like mplayer, avifile, etc, all hacked wine to get the dlls
** loaded. One would think that someone would have made this a
** separate library, with use that it has seen...
**
** :)

** Re: the dll files.

** the DLL are only for voice chat, and in fact, there is a
** configure option for disabling voice chat altogether (which
** in fact eliminates the gyvoice PE code from ever being compiled.
** I added that configure option when GyachI was added to Fedora.
** it became necessary in order to compile on X64 (and PPC). Anything
** where wine won't run.

** there are 2 codec files. They are available if you hunt around on the
** internet. The true speech codec dll files. They are a side project
** of the gyachi project, and you can find them at sf. They USED to
** be part of the code base. Obviously that was a problem. I received
** a LOT of vocal flak from one particular individual (but a lot of
** support from others) when I conditionalized, and removed the codes
** from the codebase.

** if gyvoice happens to be enabled, and the codecs aren't found, then
** you will see a popup stating which of the 2 coded files (or both)
** cannot be found. gyvoice is then disabled. So no harm is done
** by leaving it enabled

> the client/gyachi_icon_defs.h file i actually added (with other files
> along the way. I just forgot to update it with proper copyright before
> commiting it. That can and will be done.

Cool.

> The md5.* files DO have copyright, and I don't see these files being in
> conflict w/ gpl.

Indeed, it is some sort of BSD-like lincese.

> The crc32 - worse case I have a public domain version that I can replace
> this with if I can't locate a gpl'd version (or I'll borrow the gpl
> version from coreutils).

Either one of those two (public domain or GPL will do).

> the sha.* files. I'm tempted to just make those 2 files a library, which
> gyachi links against. Seems like that would solve that particular issue

Unfortunately linking GPL code against a MPL-ed code is not a compatible
mix of licenses. Simply put, you can't mix MPLed and GPLed code together,
neither by static nor dynamic linking. You must replace this sha
implemenation with another one that is GPL-ed, LGPLed, BSD or public
domain licensed

I am sure there are many sha implementaions with have such a license.

** I actually spent a bit of time looking last night. It wasn't as
** straight forward to find as one would think. The actual reference
** work is out there, and this actual code. some perl code, and a lot
** of wiki stuff, and papers.

** I'm also aware that coreutils has implementations of crc32 and sha160
** so worse case, I'll borrow that code...

> The autogen.sh script. I'll add the necessary copyright / attributions.

-- 
Reg