Re: [DotGNU]DotGNU vs .NET (was Re: flexible for users, or...) User Interfaces)

[Barry Fitzgerald]

On Tue, 9 Jul 2002, Norbert Bollow wrote:

> You don't get something better by deciding to ignore
> technological advances made by proprietary software
> companies.  The way to go is to do two things:
>
> (a) Make something similar to the good parts of what they
>     create, while improving on it as we go along.
>

Like Portable.Net, GNU-RPC, and in some ways, the auth projects FrePort
and MACS (alphabetical order).


> (b) Make good things that don't exist in proprietary
>     software yet.
>

Like PHP-GW (sort of), in many ways the auth projects (because their
design is superior and not a replication of proprietary software per-se),
Forum, VRS, etc...

It really is a two fold approach -- not only do we need to design new and
great things, we also need to maintain compatibility (where feasible) with
the rest of the world and incorporate those technologies when necessary.

A good example of this is our dual embrace of the HTTP and Jabber
protocols.  Both HTTP and the Jabber protocols are open standards, but
ultimately Jabber is still largely a product of the Free Software
community and ideology.  HTTP has been embraced and extended by literally
everyone.  It's a common protocol -- and we have to support it for
compatibility purposes, regardless of people's feelings on it's technical
merit.  We support Jabber because some people find advantages in using the
Jabber protocols for the reason of transport.

In the end, the technology that wins the day will be the one that offers
the most choice.  The one that offers the most options is the winner in
the long run.  But it does not do to offer only options that nobody knows
and is familiar with.  That is a very strong argument for continuing to
support older industry "standard" (for lack of a better word)
technologies, not to mention also-existing proprietary software.

Hope that that made sense from a strategic point of view.

        -Barry