On Sun, Feb 02, 2003 at 08:47:18PM +0000, John Chambers wrote: > > > The main problem with making such a suggestion is that most > of the suggestions I've gotten for enhancements have been > based on the assumption that I have all the abc sitting on > the disk. This isn't true. My Tune Finder doesn't save any > tunes from the (275 now) other sites where it finds abc > files. It saves the URL and a few header lines. When you > make a request for a PS or GIF or MIDI version of a tune, > it connects to the remote site, downloads the abc file, > extracts the tune, and converts it on the fly. Anything > that requires searching through the tune collection would > take several days. Even a limited search of another site > could take many, many minutes. I have avoided implementing > anything that requires more than just downloading a remote > file. And my database is now up to around 60 MB, on a free > "guest" account. The folks at the MIT EE department are > supportive so far (and some of them are folk musicians ;-), > but I do need to be somewhat careful of how much of a disk > hog I become.
Just out of interest, how long do you guess it would take to actually grab all the ABC you're indexing (and how big would it be) ... do it once and it would speed searching through them up. The problem of which, apart from resources, is cacheing obsolete versions ... maybe a protocol, or convention, or something, for grab-an-ABC-*if*-it's-changed, would be useful ? -- Richard Robinson "The whole plan hinged upon the natural curiosity of potatoes" - S. Lem To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
