> Sometimes I feel the need to find a tune by composer, but it > doesn't seems to be possible from JC tune finder. For example, > all the "tunes" by Marin Marais (there is some on the web).
Wouldn't anybody transcribing stuff that complicated put it into a separate file or identifiably distinct site and advertise the fact? In which case the Tune Finder isn't the way people would get to the music. And I doubt if anyone has ever typed a title of one of the alchemical pieces on my site into the Tune Finder - people search for that sort of thing by genre or composer, the titles are too long and error-prone. > It could be nice to be able to search your tune finder by any abc > fields, like by D: for example if you like Malicorne and want to > find some traditional tunes they've made a cover of. "D:" has a logistical problem - the transcriber may not know who's recorded the tune. I expect all of the G.S. MacLennan tunes on my site have been recorded, some of them hundreds of times, but I've no idea when or who by. (And as I can't validate the information I'm not greatly interested in being told). I can't imagine any mechanism that could propagate discography data (e.g. from cddb) to the maintainer of an ABC site, can anybody? Remember, the titles of the recordings can be ambiguous or not match those of the scores. "D:" is fine for a band, fan club or record label putting up trans- criptions of specific artists' stuff but isn't easy for anyone else to use. > Could we directly insert a little code at the beginning of an abc > file to exclude it from your tune finder, like for the robot > exclusion header in html files ? I hope people won't use it to > fool your tune finder, but to prevent it from indexing useless > tunes (for example I've made a 'favorite' for my own purposes and > it's a replication of other abc files, In which case it's a partial mirror of some other site, and that other site might go down. Better to have it available. A keyword in the header that said the tune was a mirror would be more use than hiding it. The user might want to eliminate mirrors from retrieval requests, only asking for them if the first request failed, but could only do that if the tune were indexed. Mirroring is a major headache when using the Tune Finder - the way the ABC corpus is now, you can get up to twenty copies of the same thing. Something needs to be done to stop this getting any worse. Google's "do you want to see similar results?" is sensible. > and I've also some tunes which are still works in progress or > temporary files (I sent it to my website for backup only) Simply putting them physically on your site but left un-linked to any public page would do it. About half the content of my site (not the music; medical-related material) is like that. No need to notify an indexer of their existence in the first place. =================== <http://www.purr.demon.co.uk/jack/> =================== To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
