Apologies and requests (Re: (Meta) Guile and direction)

[Lalo Martins <lalo at hackandroll dot org>]

Read below for the sum-up of a private followup to this thread.
I want to apologize for bringing the subject of direction in
the context of a syntax flamewar started by myself; in
hindsight, my timing was really stupid.

I want to apologize also for saying there is no direction; what I
meant is that the goals/roadmap aren't formalized or published.
Which for me, as a software designer, ammounts to pretty much
the same, in a project of this size, but I should have realized
other people have other backgrounds and points of view.

I will not apologize for yelling. It's exasperating to have to
say something 5 times before someone really listens.


Just one small adendum: I think most people here seem to think
outside developers decide against Guile because it's slow or
it's poorly documented and this covers pretty much all the
cases. I have to tell you from experience this is not true.

To begin with a lot of people just don't even _consider_ Guile
because it's Scheme and well, who the hell wants to force their
users to know Scheme?

Then those who don't have this problem usually go away either
because it's poorly documented, or it's slow, or it's a moving
target (I found out it would be pretty much pointless to try,
since most real work will depend strongly on the module system
and the first thing I was told when I started looking was that
it was being changed).

But this is not my point here; if you want to discuss this
subject please start a new (sub-)thread.

What I'd like to do is make a formal request to the project as
a whole:

Please PLEASE write a TODO list, tasklist or something, so that
people can know what is the project working on, why, and where
it is supposed to get to. This tells new developers where their
contributions would be the most appreciated, attracts new
developers who agree with these goals, and wards off people who
don't. And then _publish_ the results in the webpage.

And PLEASE do something about the vapourware on the webpage.
Perhaps write that off as "long-term plans" or even remove it
until people at least agree that translators are a good idea at
all. Unfortunately, presenting a nice face (PR) is as important
for making people use Guile as documentation or startup speed.


On Fri, Apr 07, 2000 at 07:50:20PM -0400, you wrote:
> >>>>> "Lalo" == Lalo Martins <lalo@hackandroll.org> writes:
> 
> Lalo> And it took you and me quite an argument to find them all, and
> Lalo> even then it is possible that we are wrong. Begin to see my
> Lalo> point? Now assume you and me are newbie developers eager to
> Lalo> contribute. Assume we're having this argument because we want to
> Lalo> find out what would be the best area to contribute in. Uh.  See?
> Lalo> :-)
> 
> Well, *I* started off saying what they were.  *I* knew ;) Yes, I
> admitted they should be documented.  But saying there aren't any, and
> yelling, especially after starting a syntax flamewar, is hardly the
> way to convince anyone of anything.
> 
> >>  We are trying to make a general purpose extension/scripting
> >> language.  People are working on things we need.  We need faster
> >> startup times, we need a better module system, we want it to be
> >> fast, we want the C interface to be usable.  These are generally
> >> agreed on.
> 
> Lalo> And why do we need these things? Just because they are broken?
> Lalo> It looks a lot like maintenence work.
> 
> Faster startup because half the people who decide against using guile
> do so because it's "too slow" but when we benchmark it, it's not
> slower than anything else if you discount starup time :)
> 
> A module system because well, yes, it's broken.  Even hard-core
> guilers hate using it, and it's suckfullness is a major contributor to
> the (aforementioned) startup time.  And actually, we'll need a
> non-suckful module system if we ever ARE going to get a translator
> system up.
> 
> Fast is just good.  the C interface is what makes it an extension
> language :)
> 
> Lalo> Or perhaps the real goal is - "clean up the house first and
> Lalo> after it is working like a kittie we think what next". It's not
> Lalo> a bad goal either.
> 
> Well, always some of that :)
> 
> Lalo> So don't you think it _could_ be worthy to discuss the long-term
> Lalo> and mid-term goals and document them in a nice-looking roadmap
> Lalo> or anything so that we can attract more developers and ward off
> Lalo> people with different agendas?
> 
> Sure it could, but not by screaming that the problem is there *are* no
> goals (rather than that they aren't written down well), especially in
> the context of a discussion of an idea of yours people didn't like.

[]s,
                                               |alo
                                               +----
--
          Hack and Roll  ( http://www.hackandroll.org )
            News for, uh, whatever it is that we are.


http://www.webcom.com/lalo           mailto:lalo@hackandroll.org
                 pgp key in the personal page

Brazil of Darkness (RPG)    ---     http://zope.gf.com.br/BroDar