[AG-TECH] AG 2.0 - easier operation

Allan Spale aspale at evl.uic.edu
Sun Oct 6 18:57:50 CDT 2002


Hello,

To comment on some things...

On Sat, 5 Oct 2002, Ivan R. Judson wrote:

> 
> Hey All,
> 
> Brian is right on the mark here. We envision interactions much like he's
> describing below. To be clear, 
> 
> Auxillary materials will be available implicitly if they are in the
> venue, the client will get the information about the materials when the
> user enters the venue and present the information. The user can then
> interact with the data in (if we get the UI right) trivial ways, e.g.
> drag it to powerpoint, excel, etc.
> 

I liked the idea of links.  It is a good way to distribute slides as well
as links to other files.

> Now, if we consider applications there are two classes that we
> immediately discover, in the first class a user drops the ppt slides
> onto their local powerpoint and powerpoint launches and they view the
> slides. This is the trival case that takes very little effort to
> support.
> 
> The second case, is when a user wants *everyone* to view the slides.
> This involves authoring a shared powerpoint viewer that can plug in to
> the Venues Client Infrastructure, so that when the user indicates they
> want to do a shared viewing, the shared viewer can contact the other
> users, point them at the software if they don't have it (so they can
> install it), point all the viewers at the data (which is available "in"
> the venue, they launch the shared viewing session.
> 

I think it is important to remember that platform-independence of the AG
in reference to this scenario.  PowerPoint files are tied to Windows
(although, Mac is included too).  Conversely, rendering video and audio
streams requires VIC which is limited to Linux, Windows, and SGI (no Mac).
I think it is crucial for viewers to be platform independent.  Also, I
would like to add that viewers should not be limited to visual displaying
of data.  In this case, renderer might be more appropriate.  It is very
important to make sure that application data is application-independent.
For instance, maybe I use Linux and not Windows, so I do not have MS
PowerPoint.  How could I participate in the shared slide presentation by
only using a viewer/renderer that is not dependent upon the application?


> This second class of applications is where I think most people are going
> to do very very interesting work. We will have to write some small set
> of applications to provide examples of how this functionality will work,
> but then I think others will have a much better set of tools for just
> importing or creating the tools they use regularly.
> 

I also think that platform-independence of applications should be
considered.  For this reason, I think that besides having WSDL for
network-based services there should also be APIs for common task domains
like numerical computing and visualization.  It is not so much about
thinking about what would go into the APIs because we are not the domain
experts.  We are the architecture experts, so providing some way for
people to add domain-specific APIs into the AG 2.0 framework might be a
good idea.  On a basic level, having an AG API that handles tying the
accessing of venues services into an application would probably be
important.  That could be part of some core AG API.  This may not be
coming out right, but hopefully you are understanding what I am trying to
convey. 

> Usability, another of Chris' points, is a big issue. I don't claim to be
> a User Interface Guru, and I suspect noone else at ANL does either.
> Rather than "taking our UI medicine", I'd prefer if people could make
> very concrete suggestions now, about what's not usable in 1.0, perhaps
> even sketch what you think the 2.0 UI should look like and mail it to
> ag-tech. These issues are things we need to address in 2.0, and we are
> planning on it, but I don't want us to miss any because we didn't ask
> enough, or you folks didn't think you should offer.
>

I do not think that people will all agree on one UI to use.  Of course,
you would not want splintering so as to confuse people either.  I think
that allowing people to customize their own AG experience would be
best...like configuring your desktop in Windows.  This would apply to a
person's AG space (full, reduced, mobile such as PDA).  It would also
apply to a person's abilities (hearing imparied, vision impaired, motor
impaired, etc.).

I also think that this is important enough to split this group from the AG
2.0 architecture group.  People that understand user interface design need
to know the purpose of the interface so that they can design an interface
or provide interface parts that people can use to build their own
interface in the best manner possible.


 
> Please speak up! I'm glad Chris started this thread :-)
> 
> --Ivan
> 
> PS -- For those who knew, baby is fine, born Thursday 2:47PM 8lbs 4oz,
> boy. I'm tired.
>

Congratulations!  What's the boy's name?


Allan
EVL at UIC
node-op
 
> ..........
> Ivan R. Judson .~. http://www.mcs.anl.gov/~judson
> Futures Laboratory .~.  630 252 0920
> Argonne National Laboratory .~. 630 252 6424 Fax
>  
> 
> > -----Original Message-----
> > From: owner-ag-tech at mcs.anl.gov 
> > [mailto:owner-ag-tech at mcs.anl.gov] On Behalf Of Brian Corrie
> > Sent: Friday, October 04, 2002 4:49 PM
> > To: Osland, CD (Chris) ; ag-tech at mcs.anl.gov
> > Subject: RE: [AG-TECH] AG 2.0 - easier operation
> > 
> > 
> > Hello all,
> > 
> > > Can I pick up on a point made on the first day of Ivan's 
> > presentations 
> > > - about ease of access to auxiliary materials (e.g. .ppt files).  I 
> > > asked whether information about these resources (master name, file 
> > > location, suggested application) could be included as part of the 
> > > session's description.  In his reply, Ivan indicated that 
> > the general 
> > > feeling was that was this sort of facility was beside, or 
> > beyond, what 
> > > AG 2.0 was addressing.
> > 
> > Is this what Ivan said (I don't remember the comment 
> > directly)? Is this what Ivan meant?
> > 
> > I see distributed powerpoint as being a service (which 
> > someone would have to create and provide for the AG community 
> > of course) that one could add to a venue. The service would 
> > have details on the master host/port, file location, client 
> > application, etc.). When one enters the room, the service 
> > downloads the file to the local disk, runs the client 
> > application, connecting it to the master as appropirate. 
> > Seems like the perfect candidate for service to me???
> > 
> > > What interests me is that the idea of an agenda with the resources 
> > > included is exactly what is happening at present, in an 
> > unformalized 
> > > way.  The meeting notice for the sessions had a URL to a 
> > web page that 
> > > was an agenda that contained URLs which were the .ppt files, and a 
> > > note of the master location!
> > 
> > I agree...
> > 
> > > I'm past my days as a programmer, but it seems to me that 
> > an interface 
> > > that, instead of going to a virtual venue, goes to a 
> > meeting, is not a 
> > > radical change (in terms of implementation!) In going to a meeting 
> > > (and the associated virtual venue becoming 'where you are' as a 
> > > result) it is also not difficult for the agenda/resource 
> > files also to 
> > > be displayed and the master server location set.  By 
> > clicking on each 
> > > resource as each agenda item comes up, it shouldn't be 
> > difficult for 
> > > the appropriate application to be launched, should it?
> > 
> > An interesting idea. How about this for a thought... 8-) Lets 
> > create an agenda service, which has links to appropriate 
> > other services for each agenda item.
> > 
> > Introduction (shared powerpoint)
> > Project budget (shared excel)
> > Project plans (shared powerpoint)
> > Project results (shared visualization)
> > Discussion
> > 
> > powerpoint, excel, and visualization are services. Agenda is 
> > a meta-service
> > (ouch) that invokes the other services as appropriate. There 
> > is an agenda server that controls the agenda services at each 
> > remote site. Someone controls the agenda just like someone 
> > controls powerpoint and moves the agenda from one item to the 
> > next. As agenda items are invoked the appropraite service is 
> > started and therefore the appropriate collaborative 
> > application would run.
> > 
> > > By this means, the majority of small meetings could be run by the 
> > > participants, without the mandatory use of an operator who 
> > is the only 
> > > one who knows the magic runes required to invoke DPPT!
> > 
> > I think I could see how this migth work in the AG 2.0 
> > architecture. Am I totally off base here? This sounds like a 
> > very powerful model to me and I like that. Is this beyond 
> > what the Argonne folks are imagining or is this exactly what 
> > they are imagining (assuming someone comes up with all these 
> > services of course???).
> > 
> > > I'm a passionate advocate of AG, and the three things that are 
> > > preventing its wider acceptance are:
> > > 
> > > 	cost - this is being addressed by any number of initiatives
> > > 
> > > 	network bandwidth - most UK academic institutions don't
> > > 		have (= can't afford) anything above 2 Mb/sec);
> > > 		many of those with better links (100 Mb or above)
> > > 		have or are building AG nodes
> > > 
> > > 	operator (running) costs
> > > 
> > > I feel that the above idea would ENORMOUSLY reassure 
> > potential users 
> > > that they are not creating a new sink for manpower 
> > (=money). naturally 
> > > it would also make life about 10 times easier for operators as well!
> > > 
> > > Whether this is considered as part of AG 2.0 or is an initiative 
> > > implemented in parallel and released together with (or even
> > > before!) AG 2.0, I don't mind, but I do feel it is important.
> > 
> > I agree, but I hope the AG 2.0 architecture can support this directly.
> > 
> > Cheers,
> > 
> > 	Brian
> > 
> 




More information about the ag-tech mailing list