endless automata

…and now on with part 2.  I’ve put together a basic web interface to use the JSON service.  You can check it out here:

http://artlogic.nfshost.com/ojw.html http://artlogic.nfshost.com/ojw/

The scanner is asynchronous, and very fast, but also not the most reliable thing - there’s no retry built in and it’s sort of difficult to stop it once it’s sent out all the requests.  If you end up going over the maximums (either set by you or built in) you will begin queuing links instead of sending new scan requests.

It’s possible to stop and restart a scan but you will have to uncheck the “Clear existing scan data?” box.  When you stop a scan results will continue to flow in - even if you start another completely different scan.  I will eventually fix this.

The second part you need to keep track of are the Jabberwocky Settings - the only active settings are the average paragraph length (which is actually the exact paragraph length now) and the randomize checkbox.  After you collect sentences by scanning the generate button will continue to make up random paragraphs.

In the next couple of days I’m going to try and get the statistical sentence choosing working and clean up the other problems.  Please let me know if you have any feedback or notice any problems.