| 00:04 | <Dashiva> | These web standards are open, reliable, highly secure, and only supported by Safari. |
| 00:06 | <Dashiva> | I accidentally opened the chatlogs from 2008 instead of 2010, and I didn't notice for a while |
| 01:16 | <TabAtkins> | Man, wth. 4 freaking "google promo" emails have gotten through the spamblocker today. |
| 01:36 | <jwm> | google knows your spamblocker |
| 01:36 | <jwm> | be afraid |
| 04:21 | <MikeSmith> | I'm trying to understand what the validation errors at http://validator.nu/?doc=https%3A%2F%2Feyeasme.com%2FJoe%2FMathML%2FHTML5%2Fextras.html |
| 04:22 | <MikeSmith> | the initial error is because you can't have an HTML img element in SVG content served as text/html |
| 04:22 | <MikeSmith> | right? |
| 04:24 | <MikeSmith> | er, make that MathML content |
| 04:27 | <MikeSmith> | I thought for this case the parser is supposed to just ignore the img, then pop the open MathML annotation-xml element from the stack and proceed from there |
| 04:48 | Hixie | tries to decode http://lists.w3.org/Archives/Public/public-html/2010Jun/0001.html to work out what the reason is for removing the text |
| 04:48 | <Hixie> | does anybody know what the actual reason behind the "widespread belief" that the paragraph is harmful might be? |
| 04:48 | <Hixie> | nothing seems to actually explain what the reason is... |
| 04:49 | <aho> | huh? :) |
| 04:55 | <jwm> | Hixie: ? |
| 04:55 | <jwm> | you there? |
| 04:56 | <Hixie> | here |
| 04:56 | <jwm> | cool |
| 04:56 | <jwm> | is there anyone wanting to do websocket serving in browsers? |
| 04:56 | <jwm> | I know the peer to peer provisions got removed |
| 04:56 | <jwm> | but I have some big plans for doing web p2p |
| 04:56 | <Hixie> | the peer to peer stuff got moved to a differnet part of the spec |
| 04:56 | <Hixie> | but it's still there |
| 04:56 | <jwm> | ohh it is? |
| 04:56 | <Hixie> | search for PeerToPeer |
| 04:56 | <jwm> | I didn't see it anymore |
| 04:56 | <Hixie> | it's very immature so far |
| 04:57 | <jwm> | yeah anyway I could help? |
| 04:57 | <Hixie> | sorry, "peer-to-peer", not "PeerToPeer" |
| 04:57 | <jwm> | I'm really serious about p2p on web as a platform |
| 04:57 | <Hixie> | interface name is ConnectionPeer |
| 04:57 | <jwm> | I think it's the only way to attempt to avoid firewalls/packet inspection properly heh |
| 04:58 | <Hixie> | i envisage some sort of mechanism that would let peers connect to each other |
| 04:58 | <jwm> | what about doing that coding in javascript |
| 04:59 | <Hixie> | but with the protocol defined in a spearate working group |
| 04:59 | <Hixie> | coding? |
| 04:59 | <jwm> | I mean relying on client coding to handle peer connections |
| 04:59 | <jwm> | just keep the spec real simple |
| 04:59 | <jwm> | maybe even utilize websockets? |
| 04:59 | <Hixie> | websockets isn't really applicable here |
| 04:59 | <Hixie> | there's nothing about web sockets that makes sense for peer to peer |
| 05:00 | <jwm> | well just being about to serve websockets from a browser should be real simple to do |
| 05:00 | <jwm> | about/able |
| 05:00 | <Hixie> | it would also not solve like 90% of the use cases :-) |
| 05:00 | <jwm> | it's like 5 lines in nodejs |
| 05:00 | <Hixie> | like, you wouldn't be able to connect to it |
| 05:00 | <Hixie> | i think we need a much better solution than that |
| 05:01 | <jwm> | can't we build on websockets? |
| 05:01 | <jwm> | I know we need a firewall breaker utilizing upnp or some other methods |
| 05:01 | <jwm> | but that can be subject to browser implementation |
| 05:02 | <Hixie> | i see nothing in websockets that would be even remotely useful in doing this |
| 05:03 | <jwm> | sad to hear that |
| 05:03 | <jwm> | I think websockets is a nice simple protocol |
| 05:03 | <jwm> | it'd be nice to have a simple low overhead protocol even one already implemented in some form to use and get p2p jump started |
| 05:04 | <jwm> | I guess it can be coded as a plugin for now |
| 05:04 | <jwm> | see if anyone finds it useful |
| 05:04 | <Hixie> | websockets is a very simple protocol, yes -- it has like three features: a handshake with origin protection, a client/server architecture, and framing for text |
| 05:05 | <Hixie> | for peer-to-peer we don't have origins, the architecture is client/client, and the framing needs to support video. |
| 05:05 | <Hixie> | it's like completely different. |
| 05:05 | <jwm> | websockets 2.0 :) |
| 05:05 | <Hixie> | i don't think trying to use websockets is a helpful way forward here |
| 05:06 | <jwm> | it's just a simple way forward |
| 05:06 | <jwm> | I don't think it's bad to be able to share text/base64 coded images |
| 05:07 | <Hixie> | if people start sending base64 encoded data over 8-bit-safe network infrastructure, we've failed. |
| 05:08 | <jwm> | well could websockets be expanded upon and be made binary safe |
| 05:09 | <Hixie> | man, you're like someone with a hammer who sees everything as a nail :-) |
| 05:10 | <Hixie> | we should design solutions to fit the problems we have |
| 05:10 | <Hixie> | not try to take our existing solutions and bend them to all similar problems |
| 05:10 | <jwm> | I like extending solutions we have already |
| 05:10 | <jwm> | websockets is pretty new and it'd be nice to boost it up :) |
| 05:10 | <Hixie> | no, it really wouldn't :-) |
| 05:10 | <jwm> | I'm actually conforming to the web since I see it as the main future of transporting most content |
| 05:10 | <Hixie> | it would just make it be more complex |
| 05:11 | <Hixie> | why can't we just make a more appropriate protocol for this use case? |
| 05:11 | <jwm> | really I say screw the web even and let's make more and more efficient protocols tasked for the problem at hand |
| 05:11 | <Hixie> | (or use one of the many existing ones?) |
| 05:11 | <Hixie> | anyway |
| 05:11 | <jwm> | well I'd like the traffic to go across regular protocols/ports |
| 05:15 | <jwm> | ok then I see you don't believe in the same direction |
| 05:15 | <jwm> | could you point me to the new section with the peer to peer area? |
| 05:17 | <jwm> | http://www.mail-archive.com/whatwg⊙lwo/msg19815.html |
| 05:17 | <jwm> | I guess there have already been discussions on this |
| 05:17 | <Hixie> | http://www.whatwg.org/specs/web-apps/current-work/#peer-to-peer-connections |
| 05:20 | <jwm> | really big on video |
| 05:21 | <Hixie> | yeah it was originally intended for video conferencing |
| 05:21 | <jwm> | ahh ok |
| 05:21 | <jwm> | I don't like the third party server concept |
| 05:22 | <Hixie> | we need it, no other way to do double NAT traversal |
| 05:24 | <jwm> | hybrid p2p |
| 05:25 | <jwm> | nat traversal can be done with upnp right? |
| 05:28 | <Hixie> | jwm: not on my network |
| 05:28 | <Hixie> | upnp is disabled on my router (because it's a HUGE security hole) |
| 05:31 | <jwm> | secure when your computers can be trusted :) |
| 05:31 | <jwm> | (which is never) |
| 05:31 | <jwm> | heh |
| 08:16 | <scotfl> | raham: aah! I forgot about sectioning root. doh |
| 12:51 | <boblet> | scotfl: what was your comment on sectioning root in reference to? |
| 16:26 | <henrikl> | When dealing with legacy systems that depend on RTMP for live streaming of events, is there any support for showing these streams in HTML5? |
| 18:35 | <Lachy> | Hixie, Requiem 1.9.4 is out. This should work with iTunes 9.1.1 |