"{{abc.def},rdf:you.klf,{{!named:kkk}.{fgh.lmk}},Named:Named:jkl}"
I call that lazy Json. I am testing my parser against that.
The result goes into the console table as Sqlson, the nested store format. More important, I can leaave the graph structures open after laoding that, and I have access to all sub graphs, start and end point rowids. The machine can do this with any nested store. Having the graph structures in place, the machine ahs all the potential names indexed! Just a nLogn walk gets me to any subgraph, at Sqlson level, in sqlite3. On opening the machine will index the names. This is what gives the machine such powerful search capability, the ability to take sqlite3 on an nLogn walk through the table store, the table, the urls, the web. All it will need it the convolution syntax from the Json crew, think regex on two Bson streams.
1 comment:
Its like you read my mind! You appear to know so much about this, like you wrote the book in it or something. I think that you could do with some pics to drive the message home a little bit, but other than that, this is great blog. A great read. I will certainly be back.
Post a Comment