[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

The LCC is a bad idea, but that doesn't mean the LSB doesn't have any issues



I think we're looking at this from the wrong end.

Using Free Software, it's easy to produce more Free Software in such a
way that it will run on all Free platforms. This is normal; most, if not
all, Free Software is built by people who mainly (or only) use Free
Software, so they do not usually look at the specific needs that
developers of non-free software have.

At some point, developers of non-free software came along, and tried to
produce non-free software for the Free platforms. These non-free
developers had a background on non-free platforms, where "a platform" is
a bunch of binaries that were compiled at a single "vendor". Since
everyone running one of those platforms has the exact same compiled
form, it is easy to produce a standard so that one can compile software
that will run on /all/ versions of the same platform; after all, there
is only "one" version (not considering later or earlier iterations of
the same platform).

This is not true if you're running a free software platform, where
technically everyone can create his 'own' version. As a result, people
who write non-free software are having issues.

To address these issues, the Free Software people created the LSB: a
standard that defines what a binary form of the source "out there"
should look like. This way, non-free developers can theoretically write
against that standard, and distribute a compiled binary that will run
"everywhere".

Obviously, that hasn't worked. The non-free developers are having issues
with their software if they write against the standard, so that it does
not, in fact, run everywhere. They look at the non-free platforms and
say 'it works over there, so it should work here'. They look at the
differences between the non-free platforms and the Free platforms, and
see that the non-free platforms are bit-by-bit the same thing
everywhere whereas the Free platforms are not. Thus, they conclude that
to solve the issue, the Free platforms should be bit-by-bit the same
thing everywhere, too.  

That is, indeed, one way to solve it: throwing out the Freedom we, Free
Software developers, have, and slowly starting to move towards a
non-free platform. But is that really what we want? I would think it is
not.

Instead of pointing us towards the non-free platforms with their
binary-identicalness, I think the non-free software developers should
tell us why the LSB has failed. They should tell us where the standard
is not clear enough, or where the issues with the LSB are, so that they
can be solved. This will require them to exhibit a level of cooperation
that we are not used to see from non-free developers, but does that
matter? After all, they want to get Free Software modified for their
purposes. That is possible -- it is the very heart of Free Software that
it can be modified for your own purposes -- but they should do it by
using the rules of Free Software, not by trying to mold the Free
Software in something which approximates the (to them more familiar)
non-free software.

Thus, the answer to the failure of the LSB is not "the Free Software
people should be more helpful to the non-free people"; the correct
answer is "the non-free people should be more helpful to the Free
Software people".

-- 
         EARTH
     smog  |   bricks
 AIR  --  mud  -- FIRE
soda water |   tequila
         WATER
 -- with thanks to fortune



Reply to: