From: Jim Freeze Date: 2002-09-05T19:52:49+09:00 Subject: Re: suggestions to the Ruby community On Thu, Sep 05, 2002 at 02:30:15PM +0900, Hal E. Fulton wrote: > ----- Original Message ----- > From: "Jim Freeze" > To: "ruby-talk ML" > Sent: Wednesday, September 04, 2002 11:02 PM > Subject: Re: suggestions to the Ruby community > > > > 1. Form a ruby-doc mailing list to begin discussions > > related to documentation. > > Fine idea. I support it, but I'm too overcommitted > to actually join it. > > I wonder if it should be kept with the main list > for a while? Maybe split it off only when traffic > gets too great. That is fine, but I don't want to overburdern this list. BTW, who could setup a ruby-doc@ruby-lang mailing list for us? > > > 2. Form a Ruby Document Organizing Committee (RDOC) > > for purpose of officially reviewing the document > > standardization process and voting on > > decisions (of course Matz has ultimate veto power). > > Calling it RDOC is cool and clever... but I'd suggest > using some other name. We don't want people saying "the > committee, not the tool" or vice versa. And we don't > want searches for one to turn up the other. How about RDSC (Ruby Documentation Standardization Committee) Not particularly clever, just an acronym. > > 3. Start a web page to publicize the activities > > of the documentation project. > > Great. > There are several that already exist. Should this effort be on the main ruby-lang site (ruby-lang.org/docs) or on its own? > > 4. Start the paper work for the standardization process. > > a Gather Requirements and write Requirements specification. > > i. get feedback from community > > ii. have committee review and approve > > b. Write functional specification > > i. get feedback from community > > ii. have committee review and approve > > c. Create design document > > i. get feedback from community > > ii. have committee review and approve > > 4. Implement RDOC process and develop tools > > 5. Integrate 4 into releases of Ruby > > This is good, but it sounds a little monolithic > to me. I think an XP-like approach might be > better. > Well, if its make you feel better we can reword it: a. Gather stories b. Programmers prioritize stories c. Programmers choose stories and outline their effort :) I agree that a-c sound monolithic. My intent was to get the scope defined so more people could help. My particular style is that the requirements document in (a) is just a page or two. The functional document in (b) defines the user interface (for now probably mostly command-line). And the design document is (dare I say it) optional. > That's up to the people working on it, of > course. But beware of two things: 1) The > process-for-the-sake-of-process syndrome, > where more paperwork and red tape are > generated than actual progress made; and > 2) the "scope creep" syndrome, where nothing > is ever releasable because a cool new feature > is currently being added. > > > I welcome your comments along the lines: > > > > a) this is a good idea, let's get going > > and I want to be on the committee > > b) what, are you crazy?? > > Are a and b mutually exclusive? :) > > > c) good idea, but I can do a lot better > > job than you... > > d) it's all hopeless, I am going back to > > programming in perl. > > I think it's great that this is turning into > a project of its own. > > And just think, it started from someone on the > list saying, "Why can't you guys get your act > together?" :) No flames! It's a paraphrase! > -- Jim Freeze ---------- Programming Ruby def initialize; fun; end A language with class