I'm glad that our paper has been referenced here again. Good to know that 
someone read it ;-).

This hypothesis of slowing down the take-up spool caused by the increasing
mass of the paper is not our idea, I took it from a dissertation 
about the Welte-Mignon reproduction system by P. Hagmann (the exact title can 
be found in our paper). The debate of constant speed vs. increasing speed has
not been decided at all. There are arguments and supporters on both side.

However, I would be interested in seeing some measured data. Unfortunately we don't
have any working player here.

So please, could someone (or maybe several of you) measure the change of the 
tape speed during the playback? This would at least give us some ideas.

I think it would be fairly easy to carry out such a measurement: take a roll, 
and during playback mark it with a soft pencil at regular time intervals (let's 
say once per second). After the replay has finished measure the distances between
the markers at several places (let's say once or twice per meter - or feet).

When you post your results, please indicate:

o What system (player) did you measure
o Identify the roll you used (by catalog no., title, etc.)
o The initial diameter of the roll
o The total length of the roll
o The final diameter of the roll
o The playback speed setting of your player (if applicable)
o Anything else you find relevant

And of course the measured data.

If you are very much interested, please try different systems, rolls, settings.

It would be highly interesting to have this experiment done. I believe it is a 
novel use of the Internet!

Please send your results to me <janosy@tsys.hit.bme.hu> or to the mailing list. 
I will post a summary of the results I get.

Thank you for participating,

Zoli

-------- (Mr.) Zoltan Janosy <janosy@tsys.hit.bme.hu> ------ #8^) ---------¶
Technical University of Budapest, Hungary, Department of Telecommunications¶
Sztoczek u. 2. 308., H-1111 Budapest,  t: +36-1-463-2093, f: +36-1-463-3266