search

webrtcftw / goals

Let's use issues to map out goals for this group
22 stars 0 forks source link

documentation: getting started #6

Open fippo opened 9 years ago

fippo commented 9 years ago

There is quite a lack of documentation about even the basics. The codelab @feross and I used during LXJS is pretty good, but we noticed some problems with it.

getUserMedia is pretty easy to understand but signaling (being BYO) seems to be the issue here. Much as I like trickle-ice, it is harder to explain than a simple "offer, answer, you're done". It is a great optimization, but without any examples that show what it optimizes (call setup time) it is very hard to understand why it is done.

samdutton commented 9 years ago

I think I mentioned to Philipp that at Google I/O this year (with help from a couple of Google tech writers) I built a codelab derived from the one at bitbucket.org/webrtc/codelab. It's a definite improvement – I'll port it back again early next year some time.

On Fri, Dec 19, 2014 at 5:20 PM, Philipp Hancke notifications@github.com wrote:

There is quite a lack of documentation about even the basics. The codelab @feross https://github.com/feross and I used during LXJS is pretty good, but we noticed some problems with it.

getUserMedia is pretty easy to understand but signaling (being BYO) seems to be the issue here. Much as I like trickle-ice, it is harder to explain than a simple "offer, answer, you're done". It is a great optimization, but without any examples that show what it optimizes (call setup time) it is very hard to understand why it is done.

— Reply to this email directly or view it on GitHub https://github.com/webrtcftw/goals/issues/6.

lisamarienyc commented 9 years ago

I'd be happy to help out here with high-level intro documentation as needed. My latest talk is published here: https://vimeo.com/113475929. While mostly evangelism for diving into WebRTC, my aim was to give web devs some context and explaining the elements of a WebRTC app. Hopefully helpful.

samdutton commented 9 years ago

FWIW the Google I/O 2014 WebRTC codelab is here http://io2014codelabs.appspot.com/static/codelabs/webrtc-file-sharing/#1, the repo for it is here https://github.com/googlesamples/io2014-codelabs/tree/master/webrtc-file-sharing, and the instructions are here https://drive.google.com/folderview?id=0B9xlQg9XpugsNzBQU2NTTVFyb3c&usp=sharing. It's oriented to file sharing.

This codelab could, of course, be improved, but I think it's better than the 2013 I/O codelab on Bitbucket https://bitbucket.org/webrtc/codelab/.

One big piece still missing: there's very little about WebRTC infrastructure and nothing about STUN and TURN.

Sam

On Tue, Dec 23, 2014 at 4:14 PM, Sam Dutton samdutton@gmail.com wrote:

I think I mentioned to Philipp that at Google I/O this year (with help from a couple of Google tech writers) I built a codelab derived from the one at bitbucket.org/webrtc/codelab. It's a definite improvement – I'll port it back again early next year some time.

On Fri, Dec 19, 2014 at 5:20 PM, Philipp Hancke notifications@github.com wrote:

There is quite a lack of documentation about even the basics. The codelab @feross https://github.com/feross and I used during LXJS is pretty good, but we noticed some problems with it.

getUserMedia is pretty easy to understand but signaling (being BYO) seems to be the issue here. Much as I like trickle-ice, it is harder to explain than a simple "offer, answer, you're done". It is a great optimization, but without any examples that show what it optimizes (call setup time) it is very hard to understand why it is done.

— Reply to this email directly or view it on GitHub https://github.com/webrtcftw/goals/issues/6.

feross commented 9 years ago

The codelab that @fippo and I ran at LXJS 2014 is here: https://github.com/LXJS/training-webrtc

(It's also derived from the https://bitbucket.org/webrtc/codelab codelab.)