On 03/03/2008 20:00, Chris Morley wrote:
> The gpx file is:
> http://www.openstreetmap.org/user/Chris%20Morley/traces/80947
> 
> I extracted the fist 5 minutes of the wav file:
> http://www.gaseq.co.uk/OSM/20080227afivemins.wav
> (not a great example of an audio track)
> 
> This plays in JOSM, but without an orange play head in the latest 
> version. It was present in the previous version of JOSM.

Thanks!

I was right on both counts. The explicit waypoints don't have time 
stamps at all, so that's why it never shows you the playhead - it's 
treating the time as 0, which is some time in 1970.

And there is indeed an empty track at the beginning of the file. So the 
fix I released a few minutes ago for tomorrow's build fixes the 
underlying problem with trying to use the markers in the first place (I 
just confirmed it works with your file).


> I'm on Windows XP, Java version 1.6.0_03, JOSM 574
> 
> I know its irritating to have ignorant people pontificating and not 
> doing the programming (I've never programmed in Java), but...
> 
> Is the program structure trying to go against the grain here? Would it 
> not be better to have the audio player on the gpx track? The audio 
> waymarks would then just move the play head. They would look the same to 
> the user as the original ones but would have a different (quite 
> lightweight) implementation. I guess you implied this in one of your 
> previous posts.

That's almost exactly what's happening now. There's nothing particularly 
"heavyweight" going on, there's just a single marker in the non-waypoint 
case to hold the start, and another to sync. I'd need a structure of 
some kind like that to hold this information, whether you see it or not, 
so the marker implementation is as good as any in that respect I think. 
To allow the waypoint case to work properly I still need the markers, 
and I don't want to change how markers work overall, because it's not 
just about audio. I also think it is useful to see the sync point 
specially, though there's no particular reason why audio markers need to 
be as chunky visually as they are now.

> I think the way I would really like to work is to start the voice 
> recorder and simultaneously make a waymark. Thereafter the audio would 
> be hands-free and and subsequent waymarks would be used (or not) as they 
> would be without audio. Then if the default synchronization was at the 
> first waymark and the start of the audio, there would be nothing to do. 
> It would be easy to explain, and easy for the user to understand. The 
> existing synchronization methods would remain as an option. The user 
> would turn on audio markers only if he had made waymarks with 
> audio-significance subsequently.

That ought to be fine except that you don't have timestamps on your 
waypoints. It would work nicely with my GPS, but yours doesn't seem to 
put timestamps on the waypoints, so it wouldn't work for you - there's 
no time on the waypo. Maybe there's an option on your GPS for this, or 
maybe there's an option to name individual track points.

I think I could solve this though - if I set unset waypoint times to the 
first trackpoint time, all would be well. It would also solve the 
problem of not playing from the start that you had. I can't do that 
tonight, but I'll let you know when.

David


_______________________________________________
josm-dev mailing list
josm-dev@openstreetmap.org
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev

Reply via email to