Re: [aspell-devel] VC6 and BCB5.5 patches

[Jose Da Silva]

On Monday 21 February 2005 06:28 am, Gary Setter wrote:
> I can search for issue surrounding it, such as patents. For me the
> culture wars is not so interesting.

Searching for patents now isn't important since you may have no patents yet, 
the problem is that it is possible to later add such features, then, all of 
a sudden you have a patent problem. .NET is still a case of the tail wagging 
the dog from an Open Source perspective since microsoft could add whatever 
features it wishes and open source can do little but follow or exit.

> Fact is there are lots of ordinary people who 
> have to do projects based on decisions others make for them, and
> if it comes down that .NET support is required, there is nothing
> they can do.

Yes, there is little point in working on a project that is moving in the 
wrong direction. If everyone is using .NET, and you can't make it work, then 
you are moving away from the majority and nobody is going to be able to use 
what you have. However, if you choose your terms/commands carefully, you 
could get things to work without having to resort to specific special 
commands. For example, gcc has moved up to 3.x, yet Kevin is trying to 
maintain 2.95 compatibility for as long as possible to keep it compatible 
across the biggest range possible.

> So the only question is, do those guys deserve our 
> support. And the most important question, will they give back?

This is only a question only the maintainer can answer, we could give an 
opinion, but that is as far as it goes. The GNU license allows you to fork, 
but there is no real point in doing that, since the best place to fix or 
improve something is at the root, and this is the root Aspell tree.

> If .NET is not compatible with the LGNU, then I can understand
> that providing .NET support would be like aiding and abetting a
> violation of the LGNU license. Bummer.

If you've followed the idiotic acrobats happening between SCO and IBM (and 
indirectly affecting the open source community) you will probably have noted 
that some open source projets that used to have SCO support have stopped 
supporting their OS. So even though you may not have much interest in 
politics, some people do.
You will also have people demanding that their portions of programs get 
pulled out. I once watched how some written documentation grew into 
something fairly decent, but due to disputes and other issues, people 
started demanding their contributions be pulled out, by the time things 
cleared, the documentation was a skeleton of it's best form since some 
fairly good additions had to get yanked out.

.NET wasn't created in a vacuum, it should have some way of you to be able to 
pull older code into it and get it permanently sucked-in, so looking at it 
that way, what you could probably do is see what commands will work across a 
fairly large range (including on .NET), for example, instead of following 
very specific .NET commands, you choose what commands have the best effect 
across the largest range possible so it will work equally well on Borland, 
Visual C, .NET, Visual Age, gcc, etc... sort of like Kevin is doing by 
trying to maintain 2.95 compatibility.
However, like I pointed out earlier, we can render an opinion, but final say 
is with the maintainer.

> > Today, many computers are running windows, but just reading
> > www.slashdot.org today, there is a posting mentioning how cisco
> > is moving a fair bit of stuff to linux as well.
>
> You can thank me for that ;-) Every tech. that I get involved
> with becomes obs. Rember Pascal and VAX/VMS? Of course every
> tech. becomse obsolute eventually.

...and some of them seem to remain timeless, for example, you probably know 
what I'm talking about if I mentioned trying to follow K&R C standards.  ;-)

Desktop OSes may be "pushed" in the directions of C++, NET, or whatever 
technological new wanna-be, but if you look at smaller technologies such as 
palm OS, embedded, smallish stuff, K&R C is still the way to go, and with 
plain C leading the pack in terms of portable across a huge arena.

Would you believe it if I were to compile some parts of Aspell to run on a 
cell phone? Well?, what's the point in that? Well, 10 years ago, you may 
remember the Game Doom required high tech 486 running at 66Mhz, while 
nowadays you've got phones that outdo those specs. looking at a bigger 
picture, perhaps you'll be carrying your personal computer with you and 
plugging it into the home TV, who knows? The point I'm hoping to say here 
is, that if you try to keep to a known root language, you are then trying 
your best to keep portability across as large a range as possible and you'll 
be surprised to see where you'll find things in a few years time.