[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Star Office (was: opensource in Israel)
- To: Ilya Konstantinov <linux-il(at-nospam)future.galanet.net>
- Subject: Re: Star Office (was: opensource in Israel)
- From: Tzafrir Cohen <tzafrir(at-nospam)technion.ac.il>
- Date: Tue, 31 Jul 2001 17:33:34 +0300 (IDT)
- cc: Herouth Maoz <herouth(at-nospam)netvision.net.il>, linux-il(at-nospam)iglu.org.il
- Delivered-To: iglu.org.il-linux-il@iglu.org.il
- In-Reply-To: <20010731164716.A25131@goblin.galanet.net>
- Sender: linux-il-bounce(at-nospam)cs.huji.ac.il
On Tue, 31 Jul 2001, Ilya Konstantinov wrote:
> On Tue, Jul 31, 2001 at 02:07:47PM +0300, Herouth Maoz wrote:
> > To put it another way, are we evangelizing for code which is trustworthy and
> > which you can fix yourself etc., or are we evangelizing against code
> > associated with mega-companies?
>
> OpenOffice isn't "evil" from being associated with Sun. Yet, as a
> developer, I'd rather mess with KWord's or AbiWord's code since:
>
> 1. I'd like to avoid bureaucracy, obviously present when contributing
> to a project led by a huge corporation. You'll depend on few code
> reviews (not necessarily bad) and their execs for deciding what should
> go in.
>
> 2. I won't get the same community feeling I'd get with the KDE people.
> Sun's people are paid employees who probably forget all about the
> project as soon as they leave work -- not the community type of people;
> You not going to swap jokes with them on IRC :)
In other words: not enough hackers
>
> 3. They don't base themselves on existing toolkits and technologies but
> rewrite it all on the Linux platform. For getting a product release out
> ofthe doors on time - it might work, but it will feature minimal
> integration with the desktop and would be a bloat (Windows' OpenOffice
> build is 30MB, Linux OpenOffice is 60MB). They rewrite the XML parser
> (libxml?), the GUI toolkit (VCL, though there are talks about making
> the VCL widgets later wrap around GTK+ ones), the Object Embedding
> system (UNO instead of CORBA). Their own replacements might provide
> features superior to the existing solutions, but the "community" way
> would be to push improvements into existing projects.
>
> To improve KWord, I won't have to download the source to Qt or KDE, and
> I could use my existing experience with those toolkits.
>
There's another factor here: mozilla, OpenOffice and AbiWord are all
cross-platform applications. Koffice applications are KDE applications.
This means that they have many things given for free, but they depend on
the KDE libraries (can you run kword without all the nice kde daemons
running in the background?)
(BTW: all of those three projects were started by commercial companies)
That said, I don't know OpenOffice's code and whether or not some
unnecessery wheels have been invented there.
> You can see exactly the same issues with Mozilla, and I believe that's
> the reason it caught up non-Netscape developers so slowly.
Mozilla started getting attention from developers when it started getting
resonable performance (that is: not too far behind netscape 4).
It's probably not fun hacking a large boddy of non-useful code.
--
Tzafrir Cohen
mailto:tzafrir@technion.ac.il
http://www.technion.ac.il/~tzafrir
=================================================================
To unsubscribe, send mail to linux-il-request@linux.org.il with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail linux-il-request@linux.org.il