Re: Focusing on innovation re: mono, python et al

[David Nielsen <david lovesunix net>]

ons, 19 07 2006 kl. 12:48 +0200, skrev Claus Schwarm:
> Hi, 
> 
> On Sun, 16 Jul 2006 18:57:04 +0200
> Lluis Sanchez <lluis ximian com> wrote:
> 
> > [...]
> > If there are memory and performance problems with Mono or Python,
> > excluding them from GNOME is not a solution, because like it or not
> > users will still use them to run applications. 
> > 
> > GNOME should adapt to this reality if it wants to survive. [...]
> 
> 
> I don't understand the argument: You're right that some users use Mono
> apps but others don't. So, why should GNOME adapt to one part of the
> user community, and ignore the other part?

Users want good applications, Mono is a really good foundation to build
good applications rapidly on. Tomboy, Banshee, Beagle - all applications
that have good testing, good stability and a rapid development cycle
meaning we can deploy functionality with the users with every cycle.
For developers using Mono means we get a free development environment
that integrates with GNOME, it's basically unlike python, java, etc. a
platform that cares about GNOME and provides GNOME style tools for us to
work with - if we bless it today we get the tools today as well, no
waiting required. One thing I miss from my college days is visual studio
for development, it was a great tool and to date only MonoDevelop comes
close within GNOME - you'd be surprised how much this means for some
developers.

> Users can already use Mono applications if they like to; it's just an
> 'apt-get install * ' away. No problem. Why should they care about Mono
> apps being included (in the desktop release)? Also, developers can
> already use Mono without GNOME depending Mono so why should the policy
> be changed?

Users don't care if an application is based on Mono, they care if it's
functional, stable and snappy. Blessing Mono is largely a decision for
future developers - e.g. I was trained in C++ and Intel ASM when I
attended college but I absolutely refuse to write another line of C/C++
in my life after discovering C# - it makes programming fun and it does
most of the boring work for me so that I can spend my time making
applications rather than chasing common idiotic problems.

> On the other hand, if GNOME depends on Mono, it will be hard to
> deinstall it without breaking GNOME. Just wait a few releases.

So.. why should users care, if Mono provides the user with good
applications that does what he/she wants why would there be any reason
to remove it - if you don't run the applications that require Mono all
you lose is the bit of space it takes up (much like many GNOME distros
ship QT as well) - unless you are on the OLPC or Maemo platform this
shouldn't be a big issue.

> This will fragment the user community, and we don't need any more
> fragmentation in the desktop: It's already bloody complicated to write
> an article for a journal when considering the differences between KDE
> and GNOME. It's frustrating to explain every time: "Under KDE, you
> do this to get X, and under GNOME you do that to get X." If you don't
> write it, you just frustrate new users. Reading all this stuff is even
> more frustrating!

As if the question of Mono's inclusion doesn't already fragment GNOME,
those who are opposed to it because it's MS technology or similar, IMHO,
silly reasons (read: not based on technological merit) won't let it in.
Then there are people like me who have been waiting for the right moment
for Mono to get included so that one day I might rely on Mono for any
development I do - I'm frankly getting to the point where if Mono
doesn't get included, I'll simply stop using GNOME. I'm getting tired of
this debate, I think Mono is great and I think continuing to push a C
based platform is a mistake.. We are handed great technology to move
GNOME forward, do we want that or are we willing to bet that C is a
viable option for Topaz. If you think that not including Mono is a way
to keep users you are mistaken, there are plenty out there who will not
switch over or leave GNOME if we don't take this step.

> Including Mono will just lead to another desktop being used widely,
> namely XFce.

Why would this be? remember so long as applications rock users will want
to use them - be that in GNOME, XFce, KDE, etc. Mono apps across the
board pretty much rock already, f-spot is the best photo management tool
out there, banshee for my money is a much more snappy application than
rhythmbox (not to mention more stable), Blam! can't be beaten for RSS
feed reading. 
Yes, we need to optimize those applications, just like every other
application we ship - Evolution being the prime example that even a C
application written by good programmers can be a horror to use - it's
slow and eats your ram for breakfast.
Untill we have good data we can't really say anything about Mono'
ressource consumption longterm, I believe that Mono based applications
can be made to run at least as well as 90% of what we have currently in
C - some claim that we could actually optimize better in a high level
language than in C but let's not make any assumptions on performance or
ressource use till we've had time to gather data.

> Splitting our user community will also lead to less influence.
> Third-party projects already ignore very basic HIG recommendations. And
> in fact: Why should they bother? It's not that GNOME appears to be the
> leading Linux desktop, isn't it? If we split due to the desktop release
> depending on Mono, it will be even harder to convince third-party
> projects to follow our example.

We offer ISVs and core developers yet another piece of technology to use
when they want to integrate into GNOME - how is this bad - C# is after
all one of the languages that's being pushed hard in college classes
currently and looking at job ads the demand for .NET development is
there. Providing this for GNOME is good, it will bring more applications
and more programmers to our platform.

> Mono seems like a good platform so it should be able to sell itself. On
> the other hand, the risks of forcing users to opt-out instead of
> letting them opt-in are immense.

What risks? we add more technology, get better apps - the only users we
risk loosing are those do listen to the anti-MS crowd and frankly if you
use GNOME because you hate Windows and not because you love GNOME maybe
you should reconsider your reasoning. GNOME is a fantastic desktop, I
love it dearly, I wouldn't feel at home in another desktop but I want it
to move forward and I want it to live after the current generation of
developers move on. We simply won't do that on C.

- David Nielsen