Slashdot Mirror

This is the problem, when a client, like XMMS, connects it negotiates the stream type. After this it just assumes all data after this point is of the same type. At no point can it switch content types. You can use something like a play list which lists multiple connections to simulate playing different formats in a row, the thing is this requires a reconnection to the server each time.

You can blame the two dominant protocols, SHOUT and ICY. One was created for the Shoutcast server and the other was created by the Icecast folks. Neither of them really considered the issue of carrying any other payload then MP3, or more to the point, changing content type in midstream.

At some point it would be great if Vorbis/Ogg became dominant because of the silly copyright restriction with MP3 that make the use of the lame encoder pretty questionable. It still has a way to go though since the code for bit peeling has yet to be finished and until that is completed, to down bitgrade an Ogg stream you have to decode it to some format like PCM and then reencode to Ogg (which is certainly not optimial for real time demands).

Everything that rises must come down under. spam-it-to-me-baby writes: "The Australian trial of a scramjet engine has fizzed. 'The experiment at the Department of Defence's Woomera Prohibited Area, 500 kilometres north of Adelaide, was not successful because the [United States-supplied] rocket experienced flight anomalies prior to the scramjet experiment,' an analysis of what went wrong says. Not to worry, another test is tentatively scheduled for next week, assuming researchers can work out what went wrong with this one on the way up."

Not to be confused with this previous scramjet test, also unsuccessful.

Ah, much better, I thought you were being unreasonable there for a minute. After Jamie drew attention to it in a Slashdot piece on Saturday, SafeSurf changed their legislative proposal. In Jamie's words, "Woo!"

That's not all he said, of course: "Please note that, now, they ONLY want to fine you thousands of dollars for failing to label anything you write that is harmful to an 8-year-old. What a relief! "The penalty for a first offense of failing to label or mislabeling material harmful to minors shall be limited to a fine of under five thousand dollars."

Bennett Haselton passed on this commentary as well:

"If you go to http://www.safesurf.com/online.htm in Netscape and "View Document Info", it shows it was last modified on October 29, 2001. (This function doesn't work in IE.)

The original OCPA is [at google]. SafeSurf apparently removed this paragraph from section 6:

Publishers may be sued in civil court by any parent who feels their children were harmed by the data negligently published. The parents shall be given presumption in all cases and do not have to prove that the content actually produced harm to their child, only that the material was severe enough to reasonably be considered to have needed a rating label to protect children.

and replaced it with:

Publishers may be sued in civil court by any parent who feels their children were harmed by the data negligently published. The parents/plantiffs shall be given presumption, if the case involves graphic images, and do not have to prove that the content actually produced harm to their child, only that the material was severe enough to reasonably be considered to have needed a rating label to protect children.

and then added three new paragraphs listing more exemptions from this rule."

Can you hear that pea through the mattresses? For the audio objectivists, a good update to CmdrTaco's recent MP3 v. Ogg Vorbis inquiry: E1ven writes: "Everyone is always arguing about whether Vorbis sounds better than MP3, or vice versa. Here is your chance to see who is right! ff123 is doing a set of Blind Listening tests and could use your help. The more ears the better!"