Re: VAST article on the Obvious Toolkit (remember?)

6 views

Skip to first unread message

Joshua O'Madadhain

unread,

Feb 14, 2011, 8:04:33 PM2/14/11

to Jean-Daniel Fekete, obvious...@googlegroups.com

Greetings, Jean-Daniel. Good to hear from you.

I've read your précis. I'm not really clear on what, exactly, you'd
like to cover in the paper, though. Perhaps adding an outline to the
doc might help to make that more generally clear.

Joshua

On Wed, Feb 9, 2011 at 3:11 AM, Jean-Daniel Fekete
<Jean-Dani...@inria.fr> wrote:
> Dear participants of the VisMaster Workshop on Software Infrastructures
> for Visual Analytics,
>
> Since the workshop, I have been busy recruiting an engineer (Pierre-Luc
> Hémery) and implementing the specifications of the Obvious Toolkit that
> we drafted in Dec. 2008.
> Now, Obvious has made good progress and will soon be released -
> although, technically, it can be used right now already.
>
> I would like to invite those of you who can devote some time to it to
> contribute to a paper we'll be writing about Obvious for the VAST
> conference. The deadline is March 31st
> (http://visweek.org/visweek/2011/info/call-participation/vast-papers).
>
> A kind-of summary of what I'd like to write is available in this google doc:
> https://docs.google.com/document/d/1NJLZ7GzOkIMWSdi_dtbL-3iZuAfUYfieEU8KlmBqp2c/edit?hl=fr&authkey=CJzKnIQK
>
> Please, tell me if you are interested in contributing to the paper and
> what level of involvement you are ready to commit to. I believe this
> paper should be co-authored by all of us since the toolkit has been
> specified with contributions from all of us. However, coordinating
> several people is difficult so I'd like to plan ahead.
>
> Hoping you'll be interested,
> Jean-Daniel
>
> --
> Jean-Daniel Fekete Jean-Dani...@inria.fr
> AVIZ Team Leader, INRIA www.aviz.fr
> INRIA Saclay - Île-de-France www.aviz.fr/~fekete
> Bat 490, Université Paris-Sud tel: +33 1 69156494
> F91405 ORSAY Cedex, France fax: +33 1 69154240
>
>

--
joshua.o...@gmail.com...................www.ics.uci.edu/~jmadden
Joshua O'Madadhain: Information Scientist, Musician, Philosopher-At-Tall
It's that moment of dawning comprehension that I live for. -- Bill Watterson
My opinions are too rational and insightful to be those of any organization.

Jeffrey Heer

unread,

Feb 17, 2011, 2:35:46 PM2/17/11

to obvious...@googlegroups.com, Joshua O'Madadhain, Jean-Daniel Fekete

Hi all, I just read through the document and am happy to help
refine/revise going forward. I also agree with Joshua regarding the
outline -- it would be great to see a bit more detail about the aim
and scope of the paper.

I'd also be interested in seeing what kinds of "chimera" applications
the framework has enabled! It would be great for a paper to include
multiple compelling examples of things you can't easily do with other
toolkits alone.

cheers,
-jeff

> --
> You received this message because you are subscribed to the Google Groups "obvious-discuss" group.
> To post to this group, send email to obvious...@googlegroups.com.
> To unsubscribe from this group, send email to obvious-discu...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/obvious-discuss?hl=en.
>
>

Jo Wood

unread,

Feb 17, 2011, 4:24:12 PM2/17/11

to obvious...@googlegroups.com

This is a copy of the mail I sent Jean-Daniel last week...

Jo.

Jean-Daniel,

Thanks for the invitation. It's good to see how much the interface has progressed. I'd fully support a paper to VAST and would like to be involved as a co-author. I think the summary you provide is a good one of how the toolkit came about and its motivation. There is clearly something useful to be shared on the implications for toolkit design and the assumptions built into existing toolkits.

What I think is currently missing is some demonstration that the integration of different toolkits with a common interface actually allows us to do some VA that would otherwise have been more difficult. I wonder if one possibility might be to use a previous VAST Challenge dataset and show how Obvious allows VA applications to be built with multiple toolkits with coupling between them. One candidate might be the 2009 challenge 'Social Network and Geospatial' as it contains some interesting network data amenable to use by Infovis, Jung and Prefuse but offers the possibility of a range of alternative representations by each of the toolkits. I also entered that challenge and built an application using Processing that used an earlier implementation of the Obvious interface. I note that there was also a winning entry from INRIA that year too so there may be some local expertise and experience with the data that could be called upon.

Jo.

Jean-Daniel Fekete

unread,

Feb 17, 2011, 6:12:58 PM2/17/11

to obvious...@googlegroups.com, Jean-Daniel Fekete

Hi All,

Thanks for the feedback. Indeed, I will complete the document and try to
propose a more complete outline.
As for Chimera, there are several ways to address it.
I think the challenge is in the separation of concern and possible
scalability.
I'll address that and we can see how convincing it is to you.

As for several compelling examples, it is a difficult issue. You can
always show examples but the difficult question is "what would be the
limits?". I'd like to show several small examples to stress the benefits
but probably stay at the principles level to explain the benefits.
If my point is not clear, you may think of an example in another domain:
what compelling examples could be done using SQL that was impossible
without? Well, almost none. But having a common vocabulary (language or
API) is the enabling technology.
I'll address that too.

Thanks for the initial comments. I'll work on it quickly.

Jean-Daniel

Le 2/17/2011 19:35, Jeffrey Heer a �crit :

> Hi all, I just read through the document and am happy to help
> refine/revise going forward. I also agree with Joshua regarding the
> outline -- it would be great to see a bit more detail about the aim
> and scope of the paper.
>
> I'd also be interested in seeing what kinds of "chimera" applications
> the framework has enabled! It would be great for a paper to include
> multiple compelling examples of things you can't easily do with other
> toolkits alone.
>
> cheers,
> -jeff
>
> On Mon, Feb 14, 2011 at 5:04 PM, Joshua O'Madadhain
> <joshua.o...@gmail.com> wrote:

>> Greetings, Jean-Daniel. Good to hear from you.
>>

>> I've read your pr�cis. I'm not really clear on what, exactly, you'd

>> like to cover in the paper, though. Perhaps adding an outline to the
>> doc might help to make that more generally clear.
>>
>> Joshua
>>
>> On Wed, Feb 9, 2011 at 3:11 AM, Jean-Daniel Fekete
>> <Jean-Dani...@inria.fr> wrote:
>>> Dear participants of the VisMaster Workshop on Software Infrastructures
>>> for Visual Analytics,
>>>
>>> Since the workshop, I have been busy recruiting an engineer (Pierre-Luc

>>> H�mery) and implementing the specifications of the Obvious Toolkit that

>>> we drafted in Dec. 2008.
>>> Now, Obvious has made good progress and will soon be released -
>>> although, technically, it can be used right now already.
>>>
>>> I would like to invite those of you who can devote some time to it to
>>> contribute to a paper we'll be writing about Obvious for the VAST
>>> conference. The deadline is March 31st
>>> (http://visweek.org/visweek/2011/info/call-participation/vast-papers).
>>>
>>> A kind-of summary of what I'd like to write is available in this google doc:
>>> https://docs.google.com/document/d/1NJLZ7GzOkIMWSdi_dtbL-3iZuAfUYfieEU8KlmBqp2c/edit?hl=fr&authkey=CJzKnIQK
>>>
>>> Please, tell me if you are interested in contributing to the paper and
>>> what level of involvement you are ready to commit to. I believe this
>>> paper should be co-authored by all of us since the toolkit has been
>>> specified with contributions from all of us. However, coordinating
>>> several people is difficult so I'd like to plan ahead.
>>>
>>> Hoping you'll be interested,
>>> Jean-Daniel
>>>
>>> --
>>> Jean-Daniel Fekete Jean-Dani...@inria.fr
>>> AVIZ Team Leader, INRIA www.aviz.fr

>>> INRIA Saclay - �le-de-France www.aviz.fr/~fekete
>>> Bat 490, Universit� Paris-Sud tel: +33 1 69156494

>>> F91405 ORSAY Cedex, France fax: +33 1 69154240
>>>
>>>
>>
>>
>> --
>> joshua.o...@gmail.com...................www.ics.uci.edu/~jmadden
>> Joshua O'Madadhain: Information Scientist, Musician, Philosopher-At-Tall
>> It's that moment of dawning comprehension that I live for. -- Bill Watterson
>> My opinions are too rational and insightful to be those of any organization.
>>

>> --
>> You received this message because you are subscribed to the Google Groups "obvious-discuss" group.
>> To post to this group, send email to obvious...@googlegroups.com.
>> To unsubscribe from this group, send email to obvious-discu...@googlegroups.com.
>> For more options, visit this group at http://groups.google.com/group/obvious-discuss?hl=en.
>>
>>

Jean-Daniel Fekete Jean-Dani...@inria.fr
AVIZ Team Leader, INRIA www.aviz.fr

INRIA Saclay - �le-de-France www.aviz.fr/~fekete
Bat 490, Universit� Paris-Sud tel: +33 1 69156494

Jean-Daniel Fekete

unread,

Mar 11, 2011, 10:47:43 AM3/11/11

to Jeffrey Heer, obvious...@googlegroups.com, Joshua O'Madadhain

Hi Everybody,

So, after coming back from PacificVis, we met with Thomas Baudel and
settled on an outline that we are quickly completing now.

You can take a look at the same place:
https://docs.google.com/document/d/1NJLZ7GzOkIMWSdi_dtbL-3iZuAfUYfieEU8KlmBqp2c/edit?hl=en_GB&authkey=CJzKnIQK

Pierre-Luc H�mery - who has done most of the coding - is currently
adding descriptions and code that will need some rewriting but describe
factually the work.

Please, don't hesitate to provide feedback, by email or in the text
itself since it is still in flux.
As a simple rule, everyone that participate to the text is entitled to
have his name in the author list. The others will be acknowledged in the
proper section anyway. If you have any issue with this decision, let me
know. Alternatively, we can use a collective name such as "The Obvious
Group" that will prevent our h-index to grow form that article but I
don't mind too much.

Still, the validation part is the difficult one. We have several
examples but nothing "formal". Applying Obvious to more examples might
help a bit the validation but will also take a lot of space.
I asked Pierre-Luc to try to wrap Weka in Obvious to be able to apply it
to any of the existing toolkits. This would be cool but will that be enough?
I am open to any idea at this point. Lack of better ideas, I'd go for
examples, like in the Prefuse paper.

Waiting for your input,
Jean-Daniel

Le 2/17/2011 19:35, Jeffrey Heer a �crit :
> Hi all, I just read through the document and am happy to help
> refine/revise going forward. I also agree with Joshua regarding the
> outline -- it would be great to see a bit more detail about the aim
> and scope of the paper.
>
> I'd also be interested in seeing what kinds of "chimera" applications
> the framework has enabled! It would be great for a paper to include
> multiple compelling examples of things you can't easily do with other
> toolkits alone.
>
> cheers,
> -jeff
>
> On Mon, Feb 14, 2011 at 5:04 PM, Joshua O'Madadhain
> <joshua.o...@gmail.com> wrote:

>> Greetings, Jean-Daniel. Good to hear from you.
>>

>> I've read your pr�cis. I'm not really clear on what, exactly, you'd

>>> H�mery) and implementing the specifications of the Obvious Toolkit that

>>> INRIA Saclay - �le-de-France www.aviz.fr/~fekete

>>> Bat 490, Universit� Paris-Sud tel: +33 1 69156494

Jean-Daniel Fekete Jean-Dani...@inria.fr
AVIZ Team Leader, INRIA www.aviz.fr

INRIA Saclay - �le-de-France www.aviz.fr/~fekete

Bat 490, Universit� Paris-Sud tel: +33 1 69156494

Reply all

Reply to author

Forward

0 new messages