Re: Setup of a new wiki - the beginning of a stronger userbase?

Andrew Suffield wrote:
> There is not a need for users to write low-grade documentation. There
> shouldn't *be* any users, and there's nothing worth documenting right
> now.

If there are any users, I'm very very surprised.  Before I can use Y as 
a desktop host to write more Y we need:

* to review and refine most of the existing code (this is why there is
   nothing which is currently worth documenting)
* a fully-working terminal (this may just be a keymap problem, but Ctrl
   doesn't work for me at the moment).
* gvim
* an mp3/ogg player
* a DVD player
* firefox
* irc/IM client
* thunderbird
* a Y video driver that renders to a Y widget (a la Xnest)
* accelerated drivers for my video card

So, only ten things.  Should be easy enough...

I wouldn't suggest that anyone else starts to use Y before at least one 
of the developers does -- until then there are going to be a _lot_ of bugs.

There are some things which I think could usefully be put on the web if 
someone feels like doing the work:

* A page with instructions on how to pull the latest version of Y from
   arch, including the locations of the main repositories.
* An archive of the tla logs and diffs from the main repositories.
   You'll need local mirrors of the archives to do this efficiently.
* It would be cool to have a tla archive explorer, to show the latest
   versions of the various Y files in the core repositories and the
   relationships/differences between them.  tla users can do all this
   from the command-line, but a good web interface might be easier to use
   for new users/developers.  Again, you'll need local mirrors of the
   archives to do this efficiently.

I agree with Andy that there is nothing much worth writing down at the 
moment -- every line of code is probably going to change before Y 
becomes usable.

That said, I would like to know of any bugs in either the trace engine 
code I added over the weekend or the new buffer-checking ymalloc/yfree. 
  If someone with time on their hands would like to write a more 
comprehensive automatic test suite for those functions I would 
appreciate it.

Cheers,

Phil