Aryeh Gregor writes: > > To clarify, the subject needs to 1) be reasonably doable in a short > timeframe, 2) not build on top of something that's already too > optimized....
Integrating a subset of RTMP (e.g. the http://code.google.com/p/rtmplite subset) into the chunk-based file upload API -- http://www.mediawiki.org/wiki/API:Upload#Chunked_upload -- would be an example of parallel I/O that we really need if we ever hope to have reasonable microphone uploads for Wiktionary pronunciation collection. I know Flash sucks, but it sucks way less for microphone upload than currently nonexistent HTML5 audio upload support, client side Java, or any other alternative, and probably will suck way less than any of those alternatives for years. Soon GNU Gnash should have microphone Speex upload on all three major platforms, assuming the Gnash programming team doesn't starve to death first. _______________________________________________ Wikitech-l mailing list Wikitech-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikitech-l
