website: http://calder.ncsa.uiuc.edu/ACE-grid
elist: ace-grid@gridforum.org

attendees:
    Jason Leigh             spiff@evl.uic.edu
    Tom Coffin          tcoffin@ncsa.uiuc.edu
    Barbara Kucera      bkucera@ncsa.uiuc.edu
    Micheal Papka           papka@mcs.anl.gov
    Lisa Childers        childers@mcs.anl.gov
    Glenn Bresnahan              glenn@bu.edu
    Arsi Vazir              arsi@nas.nasa.gov
    Adam Maddock            amaddoc1@ford.com
    Manish Parashar parashar@caip.rutgers.edu
    Robert Hood            rhood@nas.nasa.gov
    Brett Didier         brett.Didier@pnl.gov
    John Brooke          j.m.brooke@man.ac.uk
    Tom Eidson              teidson@lcase.edu
    Ted Hanss               ted@internet2.edu
    Craig Lee                    lee@aero.org
    Ed Seidell             eseidel@aei.mpg.de
    Gabrielle Allen          allen@aei.mpg.de
    Kazuyuki Shudo        shudo@ni.aist.go.jp
    Mark Trebian mtrebian@eziiga.fdl.cc.mu.uk
    Al Gilman           asgilman@iamdigex.net
    Emilee Patrick      epatrick@labs.mot.com
    Terry Disz               disz@mcs.anl.gov
    Michael Noble      mnoble@cfa.harvard.edu
    Ken Bishop                 kbishop@ku.edu
    Kyoung Park
    Karen Camarda kcamarda@tal.engr.ukans.edu
    Erick Von Schweber        evs@cacheon.com
    Deb Agarwal             daagarwal@lbl.gov
    Jay Alameda        jalameda@ncsa.uiuc.edu
    Xuebin Chi          chi@juptier.cnc.ac.cn
    Mary Anne Scott          scott@er.doe.gov


---
Agenda and Notes from the Meeting:

1. Reintroduction of draft charter to new attendees and 
definition of new GGF structure.
"ACE Grid GGF2 Working Group Presentation"

    Research vs Working Group vs Areas
    
    ACE-grid will turn into a research group - which will spawn off working
    groups
    Draft Charter from GGF1: (see slide)
    Human factors techniques
    Maybe provide a central launching point for papers contacts tools and to
    identify future areas of research.
    Domains are to be as broad as possible - all inclusive of displays and
    computing (ie pda's) other human in the loop Grid applications
    Documents: (see slide)
    Working drafts
    Groups and structure
    Working groups - solve the problem and the group ends - short lifespan

2. Recap on results of previous meeting.

    Feedback from GGF1 meeting
    Begin with data gathering
    Use events and campaigns (SC Global) to demonstrate ACE products
    ACE Survey's
    Systems Survey to gather info about tools
    Expertise Survey - what do you do? - what do you want to get out of this
    group?
    Questions:

3. Define available expertise:

    Go through survey questions with attendees to find expertise 
    that's out there.
    Here survey also tries to identify Grid resources needed.

    Do you see the ACE grid working group creating visualization tools for the
    use of the GRID? Yes, if that is determined as priority.

    ---
    How do you bring together ACE with the applications Group? We want to
    gather info from the apps group but we are interested in building human
    factors into the applications group. Intersted in helping to find solutions
    for the Apps group
    
    Mike Nobel-
    Data streams are not unidirectional - but are conglomerate - are there
    tools to address these concerns? - checklists? Possible new survey
    participant. Not only single source
    But conglomerative - combining multiple sources of data to do visualization.
    What is the definition of visualization - the little guy or the fire hose
    of data guy?
    What are the characteristics of visualization do you need
    Does it have to plug into existing framework
    What tools have you tried?
    
    Don't forget to focus on visualization tool which focus on grid services
    and work on the grid - grid services. Need more resolution on what the grid
    services are.
    
    Al Gilman - visualization for scientists which needs to work for a broader
    market - need to understand the programming model - how can they pickup
    what they need how can we identify interfaces to make this happen - how
    much is done locally or on the data archive - data space vs. the
    researchers use - need to build the relationships between. Missing pieces -
    
    Mike Papka - build grid enabled tools for the community. What are the needs
    of the other groups - What grid capabilities do we need.
    
    How do the tools themselves talk to each other - everyone wants to build
    their own system - Here ias a framework for which we can all build our own
    plugins - The issue is integrating systems like AVS and how do you add
    operability to those types of applications
    
    
    Deb A. - find some core things that we share so we can focus on that -
    Security for example - Data management performance stuff - find 1 or 2
    thins that we want to focus on.
    
    Mike - yes this is something we have to do - and so that is why we will
    move towards a research group spawning these working groups
    
    What about funding - there is a large enough need for these tools that they
    can be integrated into other funding mechanisms. 
    
    The better we can articulate our needs the better we will be able to get
    funding - Once you develop something then NSF for example thinks that it
    shoul dthen be carried on by a company.
    
    One of the item which could unify the ACE-grid - is to look at the
    architecture of collaborating environments and how it sits on top of the
    grid - half will be inside the applications box the other half will be on
    the grid so we need to api within the system embedded in the application
    and then as seperate function which works with the grid.
    
    A problem right now is that the collaborative tools are using ad hoc ways
    to handle services. There are many collaborability issues. How do we take
    advantage of the grid services. A
    
    A collaboration of running tasks is the method that the computational
    services are working - so these tools may be instrumental in building the
    architecture of the grid
    
    From the tool builders point of view - there is an emphasis on
    synchionisation of  events - so we need to find collaborative researchers
    and ask them if we can help them - finding these communities is very
    difficult.
    
    This may be where communities outside the research community may be helpful
    for example in Industry (Mobil) they are heavy  on collaboration
    
    Do all paths lead to Globus? Do I have to download a 20meg globus to get
    Cavensoft to work? So Cavernsoft is built to be with or without globus -
    this can apply to condor and legion aswell.
    
    The component approach seperates the framework fro m the app. - you could
    have a whole set of tools which address the framework - components which
    can be addressed by multiple frameworks.
    
    One thing we might want to consider is figure about a way to leverage this 
    
    Advanced Programing Models - you need data you need code and you need a way
    of mixing and matching them - Dennis Gannon
    
    Jay Alameda CCAT as a common component architecture
    
    Wrapper strategy is currently employed to achieve migration - 
    
    Comon Componetn Architecture CCA Forum

4. Define needs: Discussion of applications that need collaboration or
   advanced viz capabilities.

    10-15 minute presentations by:
	Gabrielle Allen - Cactus group
	slides
	What are the top 3 things that we should focus on? All the ones involved in
	Data
	If they could be prioritized then that would give us a starting place
	
	John Brooke - Manchester Research Center for Computational Science
	slides
	What do you see as an outcome of ACE which will help in your work -
	everyone has simultaneous needs - which should we attack first.
	Networking cue
	Component-isation - interact in a much more efficient way and also persistant
	Interoperability of display systems

5. Discussion and prioritization of item 4. or research items in general.

6. Announcements:

    8/6 San FranWorkshop on ACE
    10/7-10 Rome GGF3
    IGrid September 2002 Amsterdam
    
    List of categories as possible white papers:
    Grid related
    Vtk avs open gl - interaction with grid components
    Within the group
    
    Need to get Utah people involved