Re: GNOME Office list, and Sun/AbiWord

On 30 Aug 2000, Havoc Pennington wrote:

> Martin Sevior <> writes: 
> > From what I know of staroffice it
> > does not have Abi's GUI independence which makes this simultaneous
> > development possible.
> >
> I have heard that Sun intends to continue to make StarOffice cross
> platform. Remember that it is XP now. I'm not sure how they will
> implement XP, but you could talk about it with them. (Right now I
> think they have an XP toolkit in the Java/Swing tradition, maybe they
> are planning to move to GTK as XP toolkit. I don't know.)

Who do we talk to? As far as I know their previous code was based on
3-rd party cross platform toolkit.

> > We can share code and ideas but I don't see a merger happening unless Sun
> > adopts the Abiword code base and helps us develop it.
> > 
> Don't decide until you talk to Sun. Maybe they are willing to adopt
> some of the AbiWord goals.

I agree. However this was my impression only.

> > It would have been nice to have been in the loop about this earlier.
> > Abiword has contributed over 300,000 lines of code to Gnome and provides
> > the 4th most useful program to normal users (after filemanager, email and
> > web browser).
> >
> > Maybe it was our fault for not joining these mailing lists earlier.
> > 
> I don't think you've missed anything; nothing is decided yet. It's not
> "too late" or anything like that.
> I don't know what the end goal for the AbiWord hackers is. The GNOME
> goal, as I see it, is to get a complete office suite as soon as we
> can. I'm not interested in an office suite that's "pretty good for a
> free one", or a "small, fast but feature-incomplete suite." I think we
> want a suite that can compete with proprietary offerings, especially
> Microsoft Office. It has to have all the features end users need, it
> has to be very stable, and it has to have a nice GUI.

The goal of Abiword is to be "World's Word Processor". We want abi to
available to everyone with a feature set that is full but not bloated.
We want it to be available to everyone no matter what sort of computer
they use.

We don't want to be "not bad for free". Every feature we implement we want
to be the best out there. We re-worked our dialog structure to allow
modeless dialogs so "Insert Symbol" did not not block typing. (Who want
that!) This also allowed us to have an auto-updating Word Count dialog.
(Much more useful than the MS Word modal dialog). We could have put out a
Lists feature that was not too bad, instead we are re-working it to be the
better than MS Word. These are just a few of the examples of attention to
detail we're doing. We want every feature to work great.

That said it is obvious to everyone that we cannot get all of Word's
features all in one go. So we aim to beautifully implement a useful subset
of features in a rock solid 1.0 release. This will fill useful niches 
for people who don't have MS Word and provide a colonization base for 
continued growth. I hope we can get Abiword included on every CD
given anyway with PC magazines. After that we will keep adding
features till we overtake all other Word Processors and World Domination
will be ours :-)

If we are successful this will certainly help Gnome. A question from a
nervous employee/mum/dad might well be: But what happens to
all my abiword files when I switch to Gnome? Not a problem Abiword is
there. When someone sends you a MS Word document, read it with abiword and
send them back a *.abw with a link to download abiword to read it. Since
abi is cross platform you know this will work.

This is our vision.

> I think maybe at least some of the AbiWord hackers share this goal,
> and Sun shares this goal. It makes sense to work together, because we
> can achieve a lot more that way. I hope everyone will try to put aside
> the ego issues, and take a hard look at what their goals are and how
> they can work together to achieve them. 

Well I think our vision is pretty powerful and I hope we can persuade them
to help us. I can't see how they can unless they add features to the
Abiword codebase though. My own ego is talking here a bit but unless GUI
independence is designed into their framework I don't see that they can
achieve our vision. I agree that ego will need to be overcome. Sun's ego

> There's huge value in my opinion to having the current GNOME Office
> developers work with Sun; it will jump-start a vibrant community for
> the combined office suite, and keep the project from being 100% Sun
> engineers. 

I agree.

> StarOffice is something like 6 million lines of code; this is not to
> be sneezed at. It brings us a couple years closer to a complete suite,
> without StarOffice and the Sun engineers we were probably looking at 4
> years or so, now we can hope for 2. But only if we take advantage of
> the opportunity, and build a strong free software community around the
> combined GNOME/Star office suite. So I'm hoping the AbiWord hackers
> will be interested in helping out with that.

I don't know about 4 years. I think we can have Word Processor/Spreadsheet
combination with all the useful MS Word features in less than 2 years with
just the current progress of Abiword and gnumeric but I'm not a
Professional Software engineer. This is an extrapolation on on how fast
just a few core people I trust could implement the missing the missing Abi
features and assuming that bonobo holds up to its promise.

Anyway how much is their wordprocessor? It's probabally not much different
in size to our code base. (within a factor 3).

> > Well topics like this would be a natural thing to discuss. As Bonobo
> > integration rolls forward we would like to hear from the gnumeric team,
> > gnome-print, guppi and anyone else who wants their app to be embedded in
> > Abiword documents.
> > 

Whoops I forgot to say any application that wants to embed Abiword too.
Evolution? Nautilus? Pan? Anyone want a WP in their app?

> OK I have created Subscribe in the usual
> way. Let's follow up to that list.
> Martin, can you ask the people working on UNIX AbiWord to join
> gnome-office-list?
> is the URL.
> Also, can you guys maybe think about the goals that you have, why you
> want a word processor, what features are "must have" in the end
> product? Then you can reasonably talk to Sun about what combination of
> code will work best to achieve that.

See above.

> It's possible most discussion will need to wait until you can look
> over the Sun code Oct. 13, though.

I agree.

Martin Sevior

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]