Re: 2nd try at date/format.rb & pty.c
Mat Schaffer <[email protected]>
| Newsgroups | gmane.comp.lang.ruby.documentation |
|---|---|
| Message-ID | <[email protected]> |
On Aug 22, 2006, at 5:56 AM, Hugh Sasse wrote: > On Sun, 20 Aug 2006, James Britt wrote: > >> I've started a document on Writely. >> >> People can post comments and suggestions to this list; I'll update >> the Writely >> doc. Anyone interested in editing that doc, please send me your e- >> mail >> address and I'll add you as a collaborator. > > I've just done my first set of edits. I've fixed a few typos, added > a little info to clarify things, but the most drastic change is that > I've added a paragraph at the start about the goal of the > guidelines. This is hopefully not controversial, but will help us > shape the document as it grows. > > Now, I was wondering if it is worth putting links in to information > for beginners about CVS. Obviously turning this into a CVS tutorial > is inappropriate, but people who want to get involved will need to > know where to look for information about how to use CVS update, etc. > I'd be inclined to mention the "Pragmatic Version Control with CVS" > book as well: I think with the large role the Pragmatic Progammers > have had in our Ruby community this need not be confused with > commercial endorsement. I only raise that in case anyone thinks > that might be a problem. > > One more point that I'd like to raise somewhere: at present there is > no way to keep *internal* documentation with the code. Rdoc doesn't > provide a means to support it apart from turning off Rdoc processing > (which is fair enough, it wasn't a goal of the project), but if > there were a means to markup documentation as being internal would > this be a Bad Thing? The present guidelines are almost perjorative > about internal documentation. Is this simply because it will be > exposed to the general user, thus leading to breaking of > encapsulation, or does it have detrimental effects on code > maintenance? I'm wondering if Rdoc should support internal > documentation in the future? I'd have thought that details of > choice of algorithms, tradeoffs in design, etc would be useful for > future maintainers. For one thing, constraints change. Available > memory and processor speed changes since the 1980's are extreme > examples. I'm all for walkthrough-style stuff in this document. I'd like to see this turn into a "Getting Started" type document as time goes on with relevant links to things like learning CVS. But maybe we should be more equal opportunity and link to a page that contains a number of CVS resources (if there such a page) rather than just go straight to the Pragmatic Book. Either way, I'm with you on that sentiment. Thought I disagree with the placement of the paragraph. I'm thinking the final document should be structured something like: - Getting set up (tutorial style) -- getting CVS copy -- setting up stdlib doc and compiling docs - Style guidelines -- Where and how to write, indenting (bulleted lists) - Submission guidelines -- Where and how to post your changes (bulleted lists) Lots of example terminal sessions, links to extra info, etc. And probably a link near the top that skips to the guidelines sections for people who just need a quick reference, maybe a table of contents like thing. I didn't quite get what you were saying about internal docs. But I think I see now. I think it could be nice for rdoc to support some sort of special marking as internal documentation that would be hidden by default (similar to the show/hide source) thing. A place for notes on tricky implementation. If that's what you're getting at, I like it. But I'm not sure it's something we need to spend energy on right away. For the time being, the show/hide source function is probably good enough for most people. And we don't want to implement it in a way that would encourage internal documentation getting out of hand (usually a sign that refactoring is needed). -Mat