[Mono-list] Good New from Web-sphere

Daniel Stodden stodden@in.tum.de
06 Mar 2002 10:33:07 +0100

Content-Type: text/plain
Content-Transfer-Encoding: quoted-printable

On Tue, 2002-03-05 at 19:44, Miguel de Icaza wrote:
> >     This means only 20% of the job was done in about half a year (July-=
> > Though, not bad as long as you forget about other tools - ildasm, ilasm=
> > ASP.Net, ADO.Net, System.Windows.Forms (background to build it) etc.
> Well, the funny thing is that when we started Mono, I told my boss at
> Ximian `Do not expect any contributions to Mono to happen before the end
> of the year, because nobody contributes to a code base that does
> nothing;  we have to wait for the year's end and have something actually
> useful for people to actually contribute bug fixes'.
> I am glad I was so wrong about this ;-). =20

huh? you're surprised? but that's the most natural thing in the world.

first, it _does_ something, in that on can always fall back on the ms
framework. so you see what you're doing.

now, start browsing through the msdn class docs, which admittedly
represents decent documentation for an extremely decent class system.

everything you look is more or less screaming at you: you understand me.
it's like you've seen me before. this is your chance not only to use me
but to dig through the guts yourself. there's still tons of me so you
can still pick whatever you like with little chance of interfering with
other people. even more important, i'm mostly specified and proven so
absolutely no need for lengthy discussions and advocacy and ugly
flamewars throughout the mailing list on what the interface should look

this, forgive me, is not software engineering.
this is pacman.=20
most steps in designing a unix-style operating system kernel 10 years
ago must have been similar.

the ultimate pre-condition for open source development. tons of code,
few politics in between. the problem with open source is not lack of
capacity. the problem is negotiation and "corporate action". pick 10
people who know what they're doing and a _problem_ and you'll get at
least 5 different directions what things should look like. show them the
solution they can live with and things will take off really, really
quickly. implementing .net at the class level is such a solution. not a
problem. don't be surprised how fast the rest of this thing will grow as
soon as the runtime gets actually usuable for real-world applications by
an wider range of developers.

best regards,


Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part

Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org