I can't say much about this project yet, but its essential goal is to track who owns things. It will ask the community-at-large to update it with news of acquisitions, mergers, deals, etc.
I have posted preliminary source code at http://sunir.org/src/crystal. I will update that as I progress. Once I complete this, I should be able to complete TheProgrammersBooklist quickly.
You can now play with the script at http://sunir.org/apps/crystal/crystal.pl. Please don't break it. It's fragile.
-- SunirShah
See also
January 2002...
Are you still interested in this, Sunir? --StephenGilbert
Yes. Should I be taking it more seriously? (i.e. are you going to help me? ;) -- SunirShah
I can help with content and user-level testing of the software. --StephenGilbert
Same for me. This sounds like a project that could turn into a very useful resource. --SebPaquet
Project diary
Task list
Midway
Done
Bugs
By the way, if you want to see basically what the file format of the object database is, here is a (not-so-good) example:
Crystal_20Palace.tpb
{Entity} <name>Crystal%20Palace</name> <type></type> [relationships] [/relationships] #editstamps# <comments>7</comments> #/editstamps# <comments>This%20is%20a%20useless%20comment.</comments> {/Entity}
This is for an object of class Entity which has the attributes name, type, and comments which aggregate data structures relationships (an array) and editstamps (a hash).
Bonus question: Can anyone guess why the extension is .tpb?
Cookies for you!
Still interested in my having a crack at this, Sunir? -- ChrisPurcell
I was envisaging setting it up on a PeriPeri-deviant, using the existing MetadataSyntax to add information. That means making an ontology of sorts, and since I'm not sure what you were aiming for, I can't do that by myself. So, starting simple, here's something that should fit in:
To add a company-job-position then, one makes a new page describing (at a minimum) company, person, job title, and dc:coverage.
A merger/split/spin-off needs original and resulting companies, and a dc:date.
Putting all these on their own pages allows one to gossip about them (cite references, et cetera) in the right place. The RDF engine will handle the implications, and preserve Wiki:OnceAndOnlyOnce. What do you think? -- ChrisPurcell
I'm not convinced an objective ontology makes a lot of sense, but more like an OpenCoding system. -- SunirShah