Ideally, admin users should have control over what gets distributed:
recordings should not be trimmed (e.g. to the shorter track's length) without
their permission and if they want/need to distribute a recording that contains
tracks that have content of different lengths they should be able to do so.
One solution:
1) If it's detected that tracks are of different lengths (beyond some
configurable and/or user-specified tolerance), the admin user should be warned
when they get to the Trim/Review Hold state, assuming they've set the hold. If
they haven't set the Trim/Review hold, I think it makes sense for workflow to
be put into the hold state automatically ("conditional hold state"), with
Status column informing them why it's in the hold state ("Track length
difference detected").
2) In the hold state UI (regardless of how they got there), the admin user sees
warning about tracks being of different lengths (beyond the defined tolerance).
The warning would go away if they "fix" the problem, i.e. by setting a trim
point within the shorter track's length. However if the admin user doesn't fix
it, the system should respect the out-point in place (making sure they are
aware that their outpoint is beyond the end of the shorter track) , and let
them continue, having been duly warned (and likely having edited metadata in
some way to explain the track discrepancy to viewers).
(Alternatively, it might even be okay to preset an outpoint in the hold state
that's identical to the shorter track's length, so that the user has to go to
the effort to set it beyond the shorter track's length. In this case, user
needs to be alerted to the pre-setting of outpoint.)
4) If other services are expecting identical lengths, then the next step in the
workflow could be to pad the shorter track, so it *does* have an identical
length.
5) If we believe there are institutions that are okay with mediapackages
*always* being trimmed to shorter track's length (i.e. taking it out of the
hands of the admin user), there should be a configuration setting to this
effect. I just don't think it should be a given that Matterhorn will act this
way.
Judy
On May 9, 2012, at 6:29 AM, Schulte Olaf A. wrote:
> This "ideal solution", is that a notification on different track length (plus
> conditional hold state) only? Or does it include options for further
> processing? As far as I remember, Tobias mentioned other services expecting
> "identical" length, so you might have to work around this as well.
>
> O
>
>> -----Ursprüngliche Nachricht-----
>> Von: [email protected] [mailto:matterhorn-users-
>> [email protected]] Im Auftrag von Judy Stern
>> Gesendet: Dienstag, 8. Mai 2012 22:30
>> An: Matterhorn Users
>> Betreff: Re: [Matterhorn-users] Fails in trimming due to duration differences
>>
>> On May 8, 2012, at 3:49 AM, Rubén Pérez wrote:
>>> I totally agree with your last paragraph, except for the "we have no
>>> conditional
>> hold" part. I do think we could include a conditional hold if the inspection
>> operation
>> detects an difference in track durations too large, and perhaps allow the
>> admins to
>> edit metadata (e.g. to change the description) or take adequate measures.
>> This sounds like an ideal solution (I had assumed it would be too difficult
>> to
>> accomplish; glad my assumption was wrong.)
>>
>> Judy
>>
>> _______________________________________________
>> Matterhorn-users mailing list
>> [email protected]
>> http://lists.opencastproject.org/mailman/listinfo/matterhorn-users
> _______________________________________________
> Matterhorn-users mailing list
> [email protected]
> http://lists.opencastproject.org/mailman/listinfo/matterhorn-users
_______________________________________________
Matterhorn-users mailing list
[email protected]
http://lists.opencastproject.org/mailman/listinfo/matterhorn-users
