RE: invitation to try ImageSnippets

Hi Daniel,

Thank you for logging in and taking a look and for your
feedback. As you know, I'm a big fan of your work!

At the moment, we already do import (and save back to the image header
itself) a number of key attributes that can be found in the image header as
in IPTC info and existing keywords. 

We have not yet added the EXIF or Flickr wrapper data; but we have already
set the stage
for this and will do so in the future (a provision for Geo-Tag data is
already on the Details page). The Flickr wrapper data will be immensely
useful of course. 

I have always been keen on preserving as much original pre-existing metadata
as possible and
I am a big supporter of existing metadata standards and practices that have
been used for years (as in IPTC, DC and XMP, etc). 

I am not sure if you had a chance to check out the help files in the 
application regarding 'Ontology Use for Advanced Users' 
and the Details page. These both outline what we currently import and also
save back into 
the image header - if the image has been uploaded into ImageSnippets. Since
we also allow linking images directly from other URL's (including Flickr),
obviously we cannot/would not save the data back in the image header in that
case. 

As you are probably aware - many of the social sharing sites (including
Flickr) do 
not save any data back into the header of the image, while we do. Here is an
interesting
recent (March 2013) article about this: 

http://www.iptc.org//site/Home/Media_Releases/IPTC_study_shows_some_social_m
edia_networks_remove_rights_information_from_photos


Now that I have opened this first beta release for testing, my own first
'dog-food' experience was just realizing how easy it is for me to upload my
image into ImageSnippets first before sharing it in Facebook for example -
then I just paste the URL in FB as a post which is displayed with thumbnail
and I can feel completely comfortable that the image is being displayed with
metadata preserved both in the file on the IS site and also in the header of
the image itself. Then I can always go back into ImageSnippets and add more
metadata later without having to re-upload or share the image again. 
 
Thank you again very much for taking the time to work with the system a
little.   

I am happy to get this kind of feedback to find out which of our
desired/future features should
move to the 'top of the list' and I'm excited that you feel this is
something you would potentially
use! 

Best, 
Margaret

 

-----Original Message-----
From: Daniel Schwabe [mailto:dschwabe@inf.puc-rio.br] 
Sent: Monday, April 29, 2013 11:30 AM
To: Margaret Warren
Cc: public-lod@w3.org
Subject: Re: invitation to try ImageSnippets

Margareth (and all),
congratulations on the initiative!
I registered to experiment with the service.
As a serious non-professional photographer, besides a researcher in SW, I
feel  it is still too "basic".
As you know, I have thousands of photos online, mostly on Flickr. I've spent
a large amount of precious time tagging and geo-tagging these images.
It would be simply impractical for me to enter again all this information...
the minimum I would expect is that you are able to import the EXIF data
(Ivan Herman has written a program for this [1], for example; there's been a
flickr wrapper around for awhile [2]). In addition, for services like Flickr
and others that have an API, I would want the additional data imported as
well, if only as a starting point to begin refining and adding semantics
(e.g., people tags would be very useful).
Until such a function is available, I'm afraid I  won't have the time to add
any photos, sorry.
I hope you have already realized this and it is in the works ;-)!
Best
D
[1] http://ivan-herman.name/xmp-extraction-service/
[2] http://wifo5-03.informatik.uni-mannheim.de/flickrwrappr/

On Apr 28, 2013, at 13:27  - 28/04/13, Margaret Warren <mm@carmapro.com>
wrote:

> Hello All,
> 
> Metadata Authoring Systems, LLC has released a completely new kind of 
> application for images that uses linked data for metadata creation and 
> saves the data in html+RDFa.
> 
> ImageSnippets can be used both for digital asset management within the 
> application and/or as an image publishing platform - with images 
> shared/embedded in metadata wrappers.
> 
> The application has a lot of features for linked data professionals, 
> but it is designed to be very easy to use by people who have little or 
> no understanding of the semantic web. In preliminary testing, most 
> users who had no knowledge of linked data learned how to use the 
> system very rapidly.
> 
> But the system also has a number of features for advanced users, 
> oranizations or specialized communities.
> Custom datasets can be built on the fly and by request, custom 
> releases of ImageSnippets can be loaded with specialized vocabularies.
> 
> The system is in very early beta testing and works best in Chrome so 
> we urge you to test it that way. Firefox has some known bugs and other 
> browsers have not yet been tested at all.
> Also - at this time, any image you link or upload into the system will 
> be publically searchable.
> 
> We welcome your feedback!
> 
> http://www.imagesnippets.com
> 
> Thank you,
> Margaret Warren
> 
> 

Daniel Schwabe                      Dept. de Informatica, PUC-Rio
Tel:+55-21-3527 1500 r. 4356        R. M. de S. Vicente, 225
Fax: +55-21-3527 1530               Rio de Janeiro, RJ 22453-900, Brasil
http://www.inf.puc-rio.br/~dschwabe

Received on Monday, 29 April 2013 22:06:20 UTC