Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Social dynamics of open source were already discussed, maybe few more words on technical side.

Most (all?) examples on that article are in MIDI format, which is not a trivial sound clip. The MIDI file doesn't contain any actual sound, just 'meta' information on which insturment is used, which notes are being played, dynamics of each note and so on. The rendering of this data into actual sound is left to MIDI player.

Usually the player will have support for 'soundfont', which is the other ingredient needed. It is a bundle of few hundered of instrument samples. To produce music, those samples need to be pitch-shifted, amplified and reverbed using the information from MIDI, and mixed together. The collection of instrument samples requires much work to make it sound well in various contexts. There are precious few open ones and others are burdened with copyright.

The action of 'seeking' through MIDI file is non-trivial because music state/context is collected from events (just as OpenGL is sometimes called a state machine). So to play the last bar of music you always have to fast-forward from start.

MIDI as a standard is considered a bit obsolete, but doesn't yet have the replacement that covers all its different uses (synchronizing and controlling musical equipment, reprograming equipment's presets, capturing musical notation, reproducing music file or from input stream). In 90s the MIDI on web reached its peak because of small file sizes and hardware-accelerated support on sound cards. Since then it is in decline and its future is uncertain.

Being able to just click and hear these music samples on wikipedia, without installing browser plug-ins or hunting for players/soundfonts? Don't take it for granted.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: