Call for Media Viewer Test: Windows Users Especially

Sivakatirswami katir at hindu.org
Thu Sep 7 00:53:46 EDT 2006


Disclaimer: I'm an novice at this SMIL stuff and don't know what i'm 
doing (smile)

doing this: (watch out for email wraps)
===========
on openstack
put 
"http://www.hinduismtoday.com/archives/2006/10-12/bd-pubdesk_2006-10-01_lessons-simple-sage-w-slides.smil" 
into tUrl
  set the filename of player "Player1" to tUrl
send startPlayer1 to me in 2 seconds
end openstack

on startPlayer1
  start player player1
end startplayer1
============
was literally causing revolution to crash on Windows machines...
I got one report from Australia and another from Texas.

You are welcome to check out that file... I've left if on the server, 
though it's no longer in use...the paths are all valid

Andre say it may possibly be the use of  durations in the slide seq...

<par>
                 <audio 
src="http://www.hinduismtoday.com/archives/2006/10-12/media/bd-pubdesk_2006-10-01_lessons-simple-sage.mp3"/>

                  <seq>
             <img 
src="http://www.himalayanacademy.com/taka/past/2006/June/June_17_2006/01-bd-6-17_DSCN2184.jpg" 
region="slides" dur="00:01:10.280"/>
              <img 
src="http://www.himalayanacademy.com/taka/past/2006/June/June_20_2006/01-bd-6-20_IMG_1644.jpg" 
region="slides" dur="00:01:10.280"/>

etc.

and he said he got better results but using "start" and "end" parameters.

it worked on "most" machines (my beta test group is only about 15 people)

anyway, I get better results with just doing this for now...
as i needed this thus up and functional by 9/1 and time for R  & D was over.

===========
global gConnectedToNet
local tNextSlide

on mouseUp
   testNetConnection
   if gConnectedToNet is not empty then
     stop player "player1"
     set the filename of player "player1" to 
"http://www.hinduismtoday.com/archives/2006/10-12/media/bd-pubdesk_2006-10-01_lessons-simple-sage.mov"
     statusMsg (" Publisher's Desk. Read it on page 10. Audio only, with 
slides of Satguru Bodhinatha Veylanswami")
     set the label of this stack to "Publisher's Desk, October 2006"
     set the top of player "player1" to 4
     set the height of player "player1" to 18
     set the top of image "audioPoster" to 24
     show image "audioPoster"
      start player "player1"
      put 1 into tNextSlide
     send "postSlide(1)" to me
   else
     connectionNotification
   end if

end mouseUp


on postSlide
   put the uSlides of me into tSlides # list list of absolute URL 
pointing to jpegs on the server
   put url (line tNextSlide of tSlides) into image "audioPoster"
   put (tNextSlide +1) into tNextSlide
   if tNextSlide = the number of lines of tSlides then
     put url (line 1 of tSlides) into image "audioPoster"
     exit Postslide
   else
     send "postSlide" to me in 40 seconds
   end if
end postSlide

====

OK, this works, but I'm not super happy with the architecture because 
the media package is now fragmented:
We've got the audio file streaming into a player object and it's 
associated slide show is via a revolution image that is
being updated from a separate script. I mean, it works great, I've used 
that update image object with
remote URL's for years, it's solid (even from the Kiev in the Ukraine on 
36 K modems!) and I would use this for many things as is where the 
entire context is some Rev app, game or eduTainment product.

But it means there is only a single viable delivery context. If you want 
to look to the future at a big library delivery system you would prefer 
to have a single URL that points to a singlecontainer that equals your 
show. Which might
want to run from another frame work (web, real player, QT player) later 
in its life cycle.

And  from a production flow angle (and that's the seat I sit in) it puts 
half the job on the desk of  the rev scriptor and half the job on the 
desk desk of the media team. With SMIL, the Delivery Team (web-rev app 
building team)  just asks the Media Team( sound, photoshop, video guys) 
  "build it and give us a SMIL File" ( I may changes hats and do all 
that, but the point is the same...)

That's the beauty (and future promise, if we kick enough life back into 
it) of SMIL...
but, if it breaks QT on Windows...  :-(

I would be very interested in your thoughts here.

And before i run of to look at the podcast formats.
Will the podcasts format run from inside a QT player?
I'm not aware that you can control screen regions and run
  multiple, concurrently playing streaming media from within via RSS 2.0...

I'm not sure what the productivity gains would be: we're cutting our 
teeth here on
Final Cut Pro, Shake and that whole suite and just to put the whole 
thing together as a one movie is not that hard.

but, we are into podcasts now
and since I'm already writing rev apps to build RSS 2.0 XML and rev 
CGI's and crons and stuff
for podcasts, if i could  do the slide show thing in RSS 2.0 that would 
be very sweet.
I will look at that URL in a few days.

And, then the question becomes(as always!) is

Will it work on Windows? (In a Rev player object with QT?)

Sivakatirswami


David Bovill wrote:
> I"ve been experimenting a bit with using SMIL for slide shows... what
> do you mean by:
> 
> On 04/09/06, Sivakatirswami <katir at hindu.org> wrote:
> 
>> 1) don't use SMIL with multi-seq's with durations
> 
> Which sequences are OK - do you mean layout with multiple tracks - so
> no sound track and slide?
> 
> Also my guess is that a better direction to look at is the use of
> podcast formats: take a look at this:
> 
>  http://www.voxmedia.org/wiki/PodcastChapterTool
> _______________________________________________
> use-revolution mailing list
> use-revolution at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your 
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-revolution
> 




More information about the Use-livecode mailing list