[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: [Debconf-team] [Debconf-announce] First videos are now online

On Wed, Sep 3, 2014 at 9:14 PM, Steve Langasek <vorlon@debian.org> wrote:
> Hi Richard,
> On Mon, Aug 25, 2014 at 10:10:29PM +0200, Richard Hartmann wrote:
> > the first videos are now online and ready for your consumption:
> > http://meetings-archive.debian.net/pub/debian-meetings/2014/debconf14/
> One of the things I wanted to address post-DebConf14 is to make sure the
> event pages in summit.debconf.org include direct links to the video
> recordings.  Can you tell me if there's a reliable naming scheme for these
> files?  E.g., it appears spaces are replaced by underscores, and several
> punctuation characters are removed.  Is the filename schema available in
> code form?


but, that is run on the first version of the title.  If the title is updated (I am sure many have been), the code is not re-run, so really you can't use that code with the current titles in summit.

the good news is the string and summit db key is here:

"slug": "HPs_OpenStack_Helion_Debian_all_the_way_down",
"conf_url": "https://summit.debconf.org/debconf14/meeting/134/hps-openstack-helion-debian-all-the-way-down/",
"conf_key": "134",

"state": 3 or above are the videos that have been encoded and I expect to find them in

> The videos for 2014 are also in webm format, where
> http://meetings-archive.debian.net/pub/debian-meetings/2013/debconf13/ shows
> a series of different formats.  Should we expect other file formats to be
> coming later?  Is there a reason to link to one file format vs. another as a
> default?

More formats are pretty easy to create, the encoding parameters are defined here:


We get webm because of this .cfg line that lives on the box doing the encoding:

set it to
upload_formats=webm mp4 ogv
use the django admin to change all state 3 back to 2, run enc.py, and a day or two later we should have dirs full of mp4 and ogv  and re-encoded webm.  no we can't check to see if the file exists cuz then we could never fix anything, and yes, I suppose someone could add a "last modified" time stamp that could be checked against the file time stamp.

Just because we can does not mean we should.  If someone wants to argue in favor of it, now would be a good time.


Reply to: