RE: Another question on the Tagging use case

Thanks Susanne for mentioning Zonetag (there are Motorola clients for
Zonetag).  Zonetag is good for mobile phone users, but some users may
prefer a simpler user interaction mechanism for entering tags.
 
Voice dialling is available on many mobile phones and voice driven
mobile services have been launched by mobile operators e.g. Cingular in
the US, so I think it could be realistic to think about voice tags for
future applications.  
 
Some digital cameras allow you to add voice annotations to images as
.wav files.  Is this an allowed annotation within the scope of the MMSEM
XG?
 
Paola
 
 

________________________________

From: public-xg-mmsem-request@w3.org
[mailto:public-xg-mmsem-request@w3.org] On Behalf Of Susanne Boll
Sent: 30 November 2006 14:51
To: Hobson Paola-BPH001
Cc: MMSem-XG Public List
Subject: Re: Another question on the Tagging use case



Dear all,

just a remark.

Yahoo! Research is working on ZoneTag. Over a server the mobile 
phone gets good suggestions for tags which other users have added
to photos taken at the same place / the same cell id.
http://research.yahoo.com/zonetag/

I would not suggest voice tags on mobile phones not bo be practical
these days as speech recognition in typically noisy mobile 
enviroments is a still unsolved problem.

Kind regards,
Susanne

Am 30.11.2006 um 15:42 schrieb Hobson Paola-BPH001:


	Dear George, Susanne, Thomas
	 
	Relating to the practical situation in the Tagging use case, it
seems that the use case assumes that users will tag their content after
they have uploaded it, and that they will have access to terminals that
support keyboards.  However, users may want to real-time tag their
content when they acquire it.  They may be mobile and therefore using
limited interaction devices, and typing textual tags on a mobile phone
(or a wi-fi enabled camera) is not easy.  Access to suggested tags such
as proposed by Automatic Linguistic Indexing of Pictures - Real Time (
www.alipr.com) is a partial solution but this would only work for simple
content where appropriate tags already exist.
	 
	Another possibility would be to use voice tags which can be
added when the user captures the content, which leads to my question on
interoperability.  Does the use case imply textual only tags?  Does it
make any difference if voice tags are applied?  Does SKOS support
multi-lingual tags?
	 
	Paola
	 
	Dr P M Hobson
	Director, Personalization & Knowledge 
	Motorola Labs
	Jays Close
	Basingstoke
	RG22 4PD
	EMail : Paola.Hobson@motorola.com
<mailto:Paola.Hobson@motorola.com> 


Professor Dr. Susanne Boll
University of Oldenburg
Department of Computing Science
Media Informatics and Multimedia Systems
Escherweg 2
D-26121 Oldenburg
Germany
Tel: +49-441-9722 213
Fax: +49-441-9722 202
WWW: mmit.informatik.uni-oldenburg.de
eMail: susanne.boll@informatik.uni-oldenburg.de