MS' hidden API

[Francois Gouget]

On Wed, 3 Jul 2002, P. Christeas wrote:

>
> > Yes, the goal of wine is to eventually run all windows applications
> > (except for device drivers). However, for Wine 1.0, the focus will be on
> > running a set of widely used applications, such as Microsoft Office,
> > Lotus Notes, etc. Currently, many applications will work with a little
> > tweaking.
>
> I do have an objection to that. Linux has many native applications, including
> an office suite, a mailer, a perfect browser etc.

Ahem, I would disagree with the perfect browser part<g>. But that's
besides the point.

Apparently, what you want is to see Linux succeed on the desktop.  Then
Office, Outlook, Money & co are a requirement. The issue is not whether
they have perfectly good Linux replacements, but whether people are
going to switch if that means giving up the applications they use now.

Some people have good reasons not to switch to the Linux apps: for
instance some user use Word add-ons for reference management,
bibliographies, etc. Until similar specialized add-ons (this specific
one or others) are available these people just won't switch. Other
people have not so good reasons (that's not what I use), but that's
irrelevant. It means they won't switch to Linux just the same (or
obstruct any switch in they company).

And until people switch to Linux these specialized add-ons, and the tens
of thousands of specialty/professional applications you mentioned will
not be ported to Linux, thus preventing large numbers of people from
switching too.

So the best strategy is to add support for the applications that will
allow the greater number of people to switch to Linux. And it turns out
that if you order applications on this criteria, they will turn out to
be pretty much in the order of their popularity in the Windows world.


But all the above is irrelevant since the premise "for Wine 1.0, the
focus will be on running a set of widely used applications" is
incorrect. This is not the focus of the 1.0 release. The focus of the
1.0 release (as far as the Wine project is concerned) is to get Wine to
a stable state where we can go a fix bugs without having to reorganize
three quarters of the code and to break half the applications and
existing Wine/Winelib applications out there.

This is why dll separation, window management and stabilization of the
server protocol is on the 0.9 list.

http://bugs.winehq.com/showdependencytree.cgi?id=35

We will have reached 0.9 when the the above tasks are completed. Then
during the 0.9.0 to 1.0 period we will work out any left-over issue in
the above items and gradually freeze the code for a 1.0 release. After
that we should be able to improve Wine without breaking compatibility so
much, and without having too much regression. But there is not official
list of applications that 'must work in 1.0'.


--
Francois Gouget         fgouget at free.fr        http://fgouget.free.fr/
               RFC 2549: ftp://ftp.isi.edu/in-notes/rfc2549.txt
                IP over Avian Carriers with Quality of Service