Jon and I have begun to think about the future of LORLS. At the moment we seem to keep coming up against issues with the data design whenever we try to implement a new feature. So we’re thinking it might be time to (warning scary thought coming) throw anyway what we’ve got and start again from stratch.

Having spent a couple of late Monday evenings discussing this we’ve come up with a new data model. This is still very much at the play stage – so things are likely to change.

One thing we’d be keen to do this time round is to seperate the backend that provides access to the databse from the frontend, or rather frontends –  assuming that with a XML+HTTP based API at the backend there could be multiple frontends for different tasks, institutions, mash ups, etc.

Of course a well defined XML+HTTP API could even mean that there could be multiple backends.  For example traditional Perl based CGI scripts for use with Apache for those of us with a modicum of common sense, or Java+Tomcat for people who like smacking their foreheads against walls (you can tell we might have a small bias, can’t you? 🙂 ).

Interestingly we’ve just had an email from one of the other LORLS sites asked about forking the existing code. So maybe this is an opporuntity for some cross institutional development of LORLS.