Re: TM/JSON first draft
Robert Cerny <[email protected]>
| Newsgroups | gmane.text.xml.xtm.general |
|---|---|
| Message-ID | <[email protected]> |
Miles, Am 24.08.2010 um 06:33 schrieb Thompson, Miles: > Thanks for your proposal which seems a pretty straightforward/sensible > way to achieve this aim. I think it may well have a lot of appeal to a > certain crowd. Thank you. That is encouraging to put more energy into it. I will post some more concrete question regarding TM/JSON shortly. > ... > Apologies if this is already answered... but all this talk of JSON > makes > me wonder - has anyone implemented TopicMaps on top of CouchDB or > something like it. Any thoughts on whether such data stores are a good > fit for TM (Seems to me they would be). I recently came to a similar idea when reading up on the NoSQL movement: to use such a document database as a cache for a topic map engine (or any other statement oriented storage technology). > If so, would it be possible to base this implementation almost > entirely > on serializing back and forth this new TM/JSON stuff or is there a lot > of 'housework' that would also need to be done at a layer about > CouchDB > for a real 'topic map engine' to work? TM/JSON documents and CouchDB seem to be a good fit. It would for sure need some 'housework. Yet alone to try it out and see where the difficulties are. I would start off using a Topic Map engine below. Although attaching to an engine later is also an option. I personally have never used CouchDB, but i can imagine it gets messy at a certain project and model size. Robert