[Mulgara-dev] Priorities

[Paul Gearon]

Hi all,

We obviously have a lot of things to do in Mulgara, both bug fixes and
new features.  I've just been giving some thought to what the project
needs more than anything else.  I think just 2 points rise above all
the others:

1.  Fast loading.
This was Kowari's killer feature.  Anyone can query data quickly, but
LOADING it fast is tough.  In the long term we still want to move to
XA2 (I have some preliminary code for this), but for the moment we
need to make the XA code as fast as possible. It appears that we may
have had a slowdown in our loading speed at some point in the last 18
months.  We need to track this down and fix it ASAP.

2. SPARQL support.
I think we're dead in the water without this.  iTQL is pretty close to
SPARQL, but that's not enough.  We need a whole new parser.  I know
someone is working on this, but it's been a low priority task until
now.
Parsing is easy enough, but I think that the only people who can do
the AST properly are Andrae, Brian, or myself.  Andrew or Simon could
do it, but I don't think they have much time for Mulgara at the
moment.  Those of us who are capable will have to make this happen
soon.  Anyone else who can help will be appreciated.

I know that it's hard to get up to speed with the codebase.  To help
with this I'm making a public commitment to individually help anyone
who wants to learn about it.*  Just get in touch with me.

I'd encourage anyone with code that is near completion to complete it
and check it in soon (2 people have told me that they have something
here), but other works in progress should probably be given reduced
priority until we've addressed this issue.  I'm in that category
myself, as I'm currently working on a logic interpreter.  (All the
better to do OWL with, my dear).

Regards,
Paul Gearon

* Contrary to popular belief, infinitives can be usually split, though
it should be avoided if it sounds awkward.