I totally agreed Peter. But for the developers, if they are simply too busy, there is 
nothing we can do about. Maybe we can take a different approach here. Fix up things 
that are *critical* and then release (I'm interested in a velocity tools project beta 
release and it is sooo... close except a couple of issues). For things that can be 
done but not already a function of velocity, leave it for the next release (ie 
numerics, etc...).
 
I really think that Velocity is super nice and elegance as is right now. If we need to 
substantially improve it, there might be lots of re-desiging work not to mention 
keeping backward compatibility.

        -----Original Message----- 
        From: Peter Romianowski [mailto:[EMAIL PROTECTED]] 
        Sent: Sun 12/08/2002 22:15 
        To: 'Velocity Developers List' 
        Cc: 
        Subject: The most friendly PITA (long)
        
        

          Another week is gone and I start to feel like the PITA of
        some of the readers. :|
        
          I also start to consider the possibility of "fork your own
        codebase and shut up" (as basicly provided to me on the list
        some time ago) as a bad but maybe the only choice left :(. I
        would still like to work on velocity (improving the number
        support (has anybody tested it already?) or propose a #filter
        meachanism or help coordinating thins - see later) but I do not
        see any sence in doing so!
        
          I could live with a paradigm that says: "Well, please
        propose your patch 3 times - we're too busy" as it is on tomcat-dev
        for instance. But *there's* a living discussion and development
        going on! And that is *totally* missing here for way too long
        now! From time to time there are people discussing things -
        most times without any result. Why isn't there anything like
        "Vote for #local" or "Vote for number support proposal"? With
        rules like "X +1 means - go for it" and "any -1 from the core
        team means: no go - more discussion needed" (As I think that
        was the main reason Geir dislikes votes). That way we (you)
        could step out of this "we're talking about things for a while
        but if the thread's over..."-thing. This list is starting to
        become some kind of "academical discussion list" (more details
        later on).
        
        A proposal for further development
        ----------------------------------
        
          So maybe someone could work out something like a "Road Map"
        with things like:
        
        1. What to do with 1.3.1?
        2. What to implement for 1.4?
        3. What would come in 2.0?
        
          I mean: It's enough there! 1. and 2. can be nearly extracted
        from the list archives as there were so many discussions and
        even proposals (Map-Support, Number-Support, #local, Whitespace-
        stuff (#filter)...) And enough people seem to be interrested
        and come up with good ideas and even proposals / patches!
        
          That could be discussed (every piece of it separately, since
        the first two things are really "only" feature lists, whereas the
        third thing could contain something like "rewrite this or that" or
        "add a total backwards compatibility killer"). The best thing would
        be to assign a single person to the task of updating the document
        and bringing it to the discussion again and again until it's time
        to vote on something. And then we could go into details for every
        point on each list and discuss implementation details. I would
        really like to see more "solution driven development" instead of
        "philisophical discussions". Some things do not need that much
        discussion (like a #local directive or improving the #macro
        capabilities (using macros from #parsed files)) IMHO. Or the
        discussion is over and now it's time for actually doing it (number
        support?) Don't get me wrong: Discussion is good! But it could be a
        bit more pragmatical (since Velocity is still a *tool* and not a
        whole new technology or science - IMO :). Don't get me wrong
        again: I don't want to leave the "Simplicity-Paradigm" and implement
        every "please come into my mind"-feature in the core!
        
          The above proposal of working on a "Road Map" (really not the
        right word - found no other) is basicly a thing I wanted to throw
        in for a long period of time since I think it is really time to
        move things forward (and thus not only maintaining what's there).
        
          Note to JR (who's still reading the list I guess): I *know*
        what you're thinking! :)
        
          And as always: No personal assult.
        
        Peter
        
        [One sentence: I like Velocity and I *really* want to help to improve
        it further, but I will *leave* disappointed if we cannot manage to
        bring back life to it.]
        
        
        --
        To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
        For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
        
        

<<winmail.dat>>

--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>


Reply via email to