Re: More Political Stuff

[Matthew Berg <galt gothpoodle com>]

On 25 Aug 2000, Sean Middleditch wrote:

> So what exactly will make GNOME 2.0 so different that it can't just be
> 1.6?  If it's going to use the same libraries, etc. what's so important
> about it?

I would assume its thinks like:

  imlib depreciated and replaced with gdk-pixbuf

  goad depreciated and replaced with oaf

  bonobo as a central feature of the desktop

  replacement of gmc with nautilus

There are a lot of major changes to the base framework.  I won't list them
all, as they can be gound on the gnome developer site.
 
> GNOME DOES need some streamlining.  I can tell its built upon something
> built upon something built upon something... at least integrate it for
> the end-user, so all they need to know is "there's a new GNOME-libs, new
> GTK, etc." and not have to keep track of 10 million packages.  Even the
> Helix-Updater requires people to select the packages, and it can be
> daunting to the newbie.  Yes, OSes like Windows require 10 million
> different "components" to be upgraded, but almost no-one on Windows
> does.  Linux or UNIX, you have to.

Having a layered API != bloat. Though I'd agree that some steamlining
would be beneficial. :)

As for package tracking, what is necessary beyond what Helix does?  Click
helix-update, click gnome-libs and it merrily downloads iand installs all
the necessary RPMs for you. 

> > AbiWord is barely useable, Gnumeric is not yet complete. StarOffice,
> > as I can see it, is a way for us to be able to say "hey people,
> > GNU/Linux is now even useable for end-users; you can now drop you
> > proprietary programs and switch to our free system." without them to
> > have an excuse.
> 
> So in short, yes, we're stuck with StarOffice?  Not to start pointing
> fingers are starting a war or anything, but, for example, the KDE team
> put together an excellent office suite all on their own.  Seriously,
> many people who say Linux is ready for the desktop get criticized not
> over the lack of office suites, but the lack of quality.  Bringing in an
> inferior system wont make GNOME any better, its just going to show that
> we cant develop our own software that works better.

Neither AbiWord nor Gnumeric have stopped development people. Look at the
Changelogs.  And I personally find both applications very usable.

As for KOffice, I can't really comment as I've never tried it.  But I have
heard that it is far from stable at this point in time.

And how can you say that StarOffice being developed as a Gnome application
is a bad thing? It's just another choice. 

> Seriously, it is going to be a TREMENDOUS amount of work to port
> StarOffice to GTK/GNOME/Bonobo.  In that time, perhaps developers could
> make AbiWord usable, complete Gnumeric.  And they both have a lot more
> potential than StarOffice.  I'm certainly not pleased over the prospect
> that GNOME is simply going to make use of inferior technology just to
> get something done quickly.

Again, look at the Changelogs.  Abiword continues to have hundreds of CVS
commits a week.  And they're ramping up for a new release RSN.

Though I personally prefer Abi myself, there are still a lot of missing
features.  For me, it doesn't hurt. But for people who need what's
missing, StarOffice is a reasonable choice.

Gnumeric is also still under active development. And Havoc has even stated
that it may replace StarCalc in the suite.

Even if it doesn't, both Gnumeric and the upcmong Gnome version of
StarCalc will be using open XML file formats.  Either converting between
them or, even better, making both use the same format, would not be THAT
difficult a task.

As for porting difficulties, I won't speculate, as I have not seen the
current code base, nor do I know what will be done for the 6.0 release in
October.  

> Also, GNU/Linux isn't even close to ready for the desktop.  Even if
> GNOME was complete, there are still way too many problems.  The whole
> issue of "root" for example is confusing to newbies.  You'd be better
> off developing a stable gnome-su and whatnot so that there is never a
> need to open a terminal and su as root.  If someone has to do that even
> once, then the desktop isn't ready.

I tend to agree. Add gnome-su support to the admin tools. Between that and
the capability existing in Helix Update, and you shouldn't need to touch a
command line.

Actually, something like sudo would also be helpful for business
environments in particular.
 
> GNU/Linux is also extremely unstable, in comparison to other
> UNIXes.  While Linux itself doesn't crash that often (in 1.5 years of
> use, I've it crash once, and that was because of a faulty patch I put
> in), almost all of the software on it does.  XFree86 locks up, a
> lot.  Even the old 3.3.x series.  Most of the 'stable' GNOME crashes a
> lot.  You can throw in all the nifty programs you want, all the office
> software you'd ever need.  But other OSes already HAVE all that.  So
> what would be the point?  Really, the two reasons I stopped using
> Windows was the crashing and lack of control.  I have lots of control in
> Linux, but not by using a GUI: GUI's dumb the system down, no matter
> WHAT you try to say, it's true.  The stability is only apparant if you
> run in a console, too, or run the venerable x-lib software most distros
> still come with.

Ummm, I haven't had an X or Gnome crash in over 6 months.  Not to say that
you might not have issues, but I would be prone to suspect hardware and/or
configuration issues.

> And finally, GNOME is about a lot more than Linux... if adopting
> StarOffice is to make just to make Linux SEEM ready for the desktop, why
> are we screwing over the OSes that use GNOME?

I'm not entirely sure what you mean here.  How is adding an office suite
(that's already available in one form) "screwing over the OSes that use
GNOME"?

> > If people still wish to develop other software, then I think there
> > will be place for those. It just takes developers willing to spend time
> > on those programs coding and *maintaining* them.
> 
> And are these developers that don't maintain their own software going to
> want to maintain the StarOffice ports?

Because Sun is paying them?  
 
> 
> OK, that was a large rant about GNOME/Linux/etc. all in one.  Really, I
> tend to bitch a lot, but aren't these valid points?  Are these kinds of
> things taken into consideration by most developers?  What reasons do we
> truly give people to use GNOME?  Yes, I use it.  But I'm not
> dumb.  Windows is still easier, KDE is still faster, more stable, AND
> easier...  GNOME is trying to put together a million features, a million
> pieces of software, and we're rushing it.  No usable system was ever
> made in 3 years.  KDE wasn't.  Windows wasn't.  MacOS wasn't.  We can
> try to get everything done NOW, and screw it all up in the process.  I
> love GNOME.  I love all of it.  But it feels like in the last couple of
> months everything's been sort-of heading downhill... 

I'm not going to get into a war about what is the easier environment. I
personally find GNOME to be far simpler for what I use it for. Especially
now that Helix has provided a service that makes it MUCH easier to get
updates and new packages.

I'm curious to hear what has been rushed, in your opinion.  

As for it going downhill, I can't imagine where that impression comes
from.  The last few months have ushered in Helix as a reliable simple way
to do upgrades, the release of Gnome 1.2, which is a major improvement
over previous versions, the depreciation of older libraries with new
cleaner ones, etc.

Not to mention the developments of a lot of exciting new components, such
as Evolution, Nautalis and yes, even StarOffice.
  
> Sorry for being critical, but I want a clear understanding of where
> GNOME is heading, and I want to make sure all of you have a clear
> understanding of where you're going.

This is the point of the Gnome Foundation. :)

Matt