ruby-core

On Monday 01 December 2003 06:44 pm, Yukihiro Matsumoto wrote:
> I'm sick of proposals that sounds like "hey, matz, I'm a Ruby newbie;
> thank you for the language; by the way, Ruby's design here is not in
> Ruby-way; it should be fixed somehow; I don't know how, because I'm a
> newbie; but *you* should fix".  Sigh, don't preach me Ruby way.
>
> Sorry if I offend you, George.  I don't mean yours was one of them.
> But I want concrete proposal, its rationale, and effect estimate of
> the change.  They are what I want to see in the Ruby2 RCR.

I know I'm probably the last person on earth you, or anyone, wants to hear 
talk on this matter, but I've concluded, that this is exactly why I am going 
to, rather then hold my tongue as I had originally thought to do.

I have three, IMHO important, points to make:

1. Lets not mistake *suggestive ideas* for *proposals*. They are not 
proposals. People make suggestions as a way to LEARN. Which is why many of 
them are from newbies. This is often a misunderstood form of self-education, 
very much like a scientific process (ref. http://jan.ucc.nau.edu/~gaud/
bio372/class/behavior/sciproc.htm) People observe their own Ruby-related 
problems and make suggestive hypothesis toward solutions, often times 
directed right at Ruby itself. They present these in the mailing lists to 
"put them to experiment". In doing so they learn. They learn how thier idea 
may be is misconcieved, or how there are good alternatives to it, or how the 
idea might be improved upon, and so on. And we all learn from this, not just 
the submitter of the idea.. Then sometimes, a good idea is actually hit upon. 
Thank goodnes they posted that suggestion! And the community response gives 
the suggester sound footing to actually make a real proposal.   All of this 
is a way to learn and explore Ruby. To feel apart of a vibrant language that 
actually encouges its users to ask the questions, to make the sugestions, to 
feel just a little bit "like matz". So please, do not mistake these things 
for sins against the Ruby Way.  For they are not impinging demands to change 
Ruby. They are but the "protoplast" for what may become so when they are made 
"real" as an RCR on the Ruby Garden. Then and only then.

2. And so I am led to make a suggestion now: That we, as a community, might 
think about how to better this whole process. I for instance, was thinking of 
a wiki-ish web knowledgebase of well catagorized suggestions, good search 
functionality, perhaps automatic poles for "late stage" ideas (those being 
prepared for RCR), etc. Something that really moves ideas through a process 
from early suggestion and exploration through a perculation (mod Perc ;-), 
upward into final RCR. And with this, likewise, a much more exacting RCR 
process itself. (I believe David is working on the latter presently). The 
rationale behind this is of course to organize the seemingly chaotic way in 
which it is approached now, thus lowering repetition of same-old-same-old as 
all we have now is a poor mail-archive search; and at the same time improve 
the quality of the really good ideas that make it to the top. This would help 
matz, I think, quite a lot.  (But mind you, even if we gain #2, please do not 
forget #1!)

3. My last point is a bit harder for me to come out and say, because I do not 
want to come across sounding "sacreligious", for a lack of a better term. I 
am not presenting this to point fingers, or any such address. I am only 
asking for an honest answer to "why?", and to point out that it furthers what 
I have said above.  -- I recently submitted an RCR on the Garden, and you, 
matz, rejected it within moments of my posting. In fact it was rejected so 
quickly that your reply wasn't even complete, but was chomped off half-way 
down, so that I had to come back to the mailing list and say, "what, did you 
say?" In the course of that thread, and my constant proding for a good reason 
for rejection, I was finally told that my idea was not good because the whole 
notion was simply bad programming. Well, the "funny" thing about this, is 
that "my idea" wasn't my idea. Rather it was your idea (or at best, one of 
your -dev compatriots), because the idea is listed in Ruby's latest source as 
TODO. Why would you reject an idea that you had already planned to do? It 
dosen't make sense. There may be good reasons and I'm not going to speculate 
on the matter, beyond the fact that it is just more evidence that we need to 
address the ideas above, becuase obviously the current route is becoming a 
burden for our beloved matz.

I hope I have presented my ideas clearly, and further, I hope have not come 
across too brazen --that's furthest from my intentions. I simply want, if I 
can, to help.

T.

Thread

Prev Next

In This Thread

Prev Next