Hi people,
I just came back from holidays and red all this thread at once.
The main joke is in the initial posting where Jon goes from telling Costin
that "It really scares me that you are the only person..." in its beginning
to "We just don't have enough overall developer resources to support two
different forks".
I red it and I thought that Jon was trying to recruit Costin (the "only"
TC3.3 man) to TC4.x because Jakarta had not enough developer resources to
afford "wasting" Costin on 3.3 (only Costin since Jon says he is alone).
Man... this got me confused: I thought Sun had a bunch of people working
full time on Tomcat and then I learn that Jon is concerned about wasting
ONE part-timer's work with 3.3!!! (Remember again that Jon says that
Costin is the ONLY person interested in 3.3.)
=:o)
I am glad that further postings confirmed that there is a bunch of full
time developers working on 4.x and that there is a lot of other people
interested and working on 3.3. I am glad because:
- Tomcat 4 is a very interesting evolution;
- I can not afford to hold my breath until 4.x gets stable and I hope
that 3.3 "gets there" soon(er) enough to provide me with a good
production Servlet engine that supports JSPs.
I would like to see a finished production Tomcat server soon. However,
there is NO Tomcat version yet that provides all the functionality and
robustness that most serious developers expect from a Servlet engine.
The fact that production experience is being injected in Tomcat 3.3 is
a reassurance to me that such a version will happen soon enough. Forcing
the people that are doing that work in 3.3 to quit from 3.3 and learn
4.x is not the shortest path (in time) to use their knowledge in having
_some_ Tomcat version REALLY production ready.
I will believe that 4.x is production ready when it happens. The
time-to-become-stable estimation is one of those that fails more often
and by larger amounts.
In the meantime, 3.3 looks closer. Refactoring is different from
such a large redesign/rewrite as Tomcat 4.x is - and this holds true
even if you call it Catalina or JServ 2.
Lets see:
- There are to many new ingredients, too many new components and too
many new developers in Tomcat 4. Too many things can fail;
- OTOH, in 3.3 there is Costin and 3.x-experienced guys cleaning up,
refactoring and fixing what they know well, only adding new stuff
where essential.
Agin:
- On one hand I see a lot of new pieces on 4.x that can provide a
lot of unforeseen issues.
- And OTOH I see work over a better known base with much less new
pieces on 3.3.
So, I have to believe that 3.3 is the shortest path to have some
production ready Tomcat.
I respect Craig a lot. A LOT really. But his judgment might be a bit
biased since 4.x is his child. It is so easy to be (over)optimistic
about the time-to-production of one's own child!
I just don't know what to think about Jon. Maybe he got overexcited
about 4.x new features and he wants to have it ready ASAP. Maybe that
is why he makes this kind of political move with such enthusiasm that
he even becomes incoherent.
My political move is to defend 3.3, since I only get overexcited by
having a Production Quality Tomcat AsSoonAsPossible!
Yes, I like the future in Tomcat 4.x, but I need a PRESENT really
soon too and 3.3 would be good enough for that.
I will be clear - if Costin is forced to move away I will try to use
anything that him and the other 3.3 guys post on SourceForge, on any
other site or mail me privately.
BUT if that happens I will be much less motivated to come back to
Tomcat since Tomcat would have become a feud where there is no longer
room for the competition and complementarities of good ideas.
Remember that Tomcat 4.x wouldn't exist so soon if the argument
against dividing resources would be applied with the focus on getting
a production quality 3.x server out. And one could argue that such
focus would better fulfill the responsibilities of the project towards
its users.
In the end, I think we are better with this division of resources and
interests since it ensures that products get fully matured (as 3.3 for
the 3.x line) while future more sophisticated designs evolve (4.x).
Have fun,
Paulo Gaspar
> -----Original Message-----
> From: Jon Stevens [mailto:[EMAIL PROTECTED]]
>
> Costin,
>
> It really scares me that you are the only person (as far as I can
> tell) that
> is seriously interested in maintaining and developing Tomcat 3.x into the
> future. It is not good to have the entire rest of the core developers work
> on Tomcat 4.x and having you sit here and say that you are going to work
> towards back porting everything that the Tomcat 4.x people come up with on
> your own. Talk about a complete duplication of effort by only a single
> individual.
........
> We just don't have enough overall developer resources to support two
> different forks of the same project going on at the same time! This isn't
> good! :-(
>
> thanks,
>
> -jon
>
