Evil Datagrids! ( was Thoughts on AlanCooper's Keynote )

13 Feb 2008 - 4:36pm
6 years ago
7 replies
498 reads
Pierre Roberge
2005

Greg said:
Not being funny, just looking to learn something from folks with way more experience than myself!

My Response:

Sparklines come to mind or anything à la Tufte.

http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001OR

Pierre Roberge
User Experience Designer - Business Analyst
etfs

(819)-566-2901 #2193
1-800-465-8602 #2193
http://www.etfsinc.com/

----------
etfs inc. The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential and/or
privileged material. Any review, retransmission, dissemination or other
use of, or taking of any action in reliance upon this information by
persons or entities other than the intended recipient, is prohibited. If
you have received this in error, please contact the sender and delete
the material from any computer. Unless otherwise stated, opinions
expressed in this e-mail are those of the author and are not endorsed
by the author's employer.

etfs inc. L'information transmise ne s'adresse qu'au particulier ou à
l'organisme à qui il est dirigé. Il peut contenir des renseignements de
nature privilégiée et/ou confidentielle . Si le lecteur de ce message
n'est pas le destinataire visé, ni l'employé ou le mandataire chargé de
la livraison au destinataire visé, il est par la présente avisé que
toute dissémination, distribution ou transcription de cette
communication est strictement interdite. Si vous avez reçu la présente
communication par erreur, veuillez nous en aviser immédiatement par
courriel et détruire le document de tout ordinateur le contenant. À
moins d'avis contraire, toute opinion exprimée dans le présent courriel
est celle de son auteur et n'est pas endossée par l'employeur de la
personne qui l'exprime.

Comments

13 Feb 2008 - 4:41pm
SemanticWill
2007

Tufte is a good place to start - Envisioning Information, Visual Display of
Quantitative Data, Beautiful evidence. His examples are old, but the ideas
about how to think about the problem are par none!

2008/2/13 Pierre Roberge <Pierre.Roberge at etfsinc.com>:

> Greg said:
> Not being funny, just looking to learn something from folks with way more
> experience than myself!
>
> My Response:
>
> Sparklines come to mind or anything à la Tufte.
>
> http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001OR
>
>
> Pierre Roberge
> User Experience Designer - Business Analyst
> etfs
>
> (819)-566-2901 #2193
> 1-800-465-8602 #2193
> http://www.etfsinc.com/
>
>
> ----------
> etfs inc. The information transmitted is intended only for the person or
> entity to which it is addressed and may contain confidential and/or
> privileged material. Any review, retransmission, dissemination or other
> use of, or taking of any action in reliance upon this information by
> persons or entities other than the intended recipient, is prohibited. If
> you have received this in error, please contact the sender and delete
> the material from any computer. Unless otherwise stated, opinions
> expressed in this e-mail are those of the author and are not endorsed
> by the author's employer.
>
> etfs inc. L'information transmise ne s'adresse qu'au particulier ou à
> l'organisme à qui il est dirigé. Il peut contenir des renseignements de
> nature privilégiée et/ou confidentielle . Si le lecteur de ce message
> n'est pas le destinataire visé, ni l'employé ou le mandataire chargé de
> la livraison au destinataire visé, il est par la présente avisé que
> toute dissémination, distribution ou transcription de cette
> communication est strictement interdite. Si vous avez reçu la présente
> communication par erreur, veuillez nous en aviser immédiatement par
> courriel et détruire le document de tout ordinateur le contenant. À
> moins d'avis contraire, toute opinion exprimée dans le présent courriel
> est celle de son auteur et n'est pas endossée par l'employeur de la
> personne qui l'exprime.
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

--
~ will

"No matter how beautiful,
no matter how cool your interface,
it would be better if there were less of it."
Alan Cooper
-
"Where you innovate, how you innovate,
and what you innovate are design problems"
-------------------------------------------------------
will evans
user experience architect
wkevans4 at gmail.com
-------------------------------------------------------

13 Feb 2008 - 4:47pm
SemanticWill
2007

Another thing - start by not thinking about rows of data filled with
records. Instead start with the fundamental unit -
A person in a HR system
A book in a library
A car in used car lot.

Imagine them all as nodes on a graph. Now - instead of being simple circles
representing - say - honda accord - think of the shape as a dodecahedron.
Every place on the shape where vertices meet is called a faced - which maps
to an attribute - say - color, price, mpg, horsepower - now imagine all the
more powerful ways of dynamically visualizing things/sorting things (nodes -
cars) based on those facets, grouping them in sets of thinks - like playing
with math blocks as a kid - if you start to think about discreet database
records as objects like this thought experiement - you might come up with
whole new ways to display the data.

On Feb 13, 2008 4:41 PM, W Evans <wkevans4 at gmail.com> wrote:

> Tufte is a good place to start - Envisioning Information, Visual Display
> of Quantitative Data, Beautiful evidence. His examples are old, but the
> ideas about how to think about the problem are par none!
>
> 2008/2/13 Pierre Roberge <Pierre.Roberge at etfsinc.com>:
>
> Greg said:
> > Not being funny, just looking to learn something from folks with way
> > more experience than myself!
> >
> > My Response:
> >
> > Sparklines come to mind or anything à la Tufte.
> >
> > http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001OR
> >
> >
> > Pierre Roberge
> > User Experience Designer - Business Analyst
> > etfs
> >
> > (819)-566-2901 #2193
> > 1-800-465-8602 #2193
> > http://www.etfsinc.com/
> >
> >
> > ----------
> > etfs inc. The information transmitted is intended only for the person or
> > entity to which it is addressed and may contain confidential and/or
> > privileged material. Any review, retransmission, dissemination or other
> > use of, or taking of any action in reliance upon this information by
> > persons or entities other than the intended recipient, is prohibited. If
> > you have received this in error, please contact the sender and delete
> > the material from any computer. Unless otherwise stated, opinions
> > expressed in this e-mail are those of the author and are not endorsed
> > by the author's employer.
> >
> > etfs inc. L'information transmise ne s'adresse qu'au particulier ou à
> > l'organisme à qui il est dirigé. Il peut contenir des renseignements de
> > nature privilégiée et/ou confidentielle . Si le lecteur de ce message
> > n'est pas le destinataire visé, ni l'employé ou le mandataire chargé de
> > la livraison au destinataire visé, il est par la présente avisé que
> > toute dissémination, distribution ou transcription de cette
> > communication est strictement interdite. Si vous avez reçu la présente
> > communication par erreur, veuillez nous en aviser immédiatement par
> > courriel et détruire le document de tout ordinateur le contenant. À
> > moins d'avis contraire, toute opinion exprimée dans le présent courriel
> > est celle de son auteur et n'est pas endossée par l'employeur de la
> > personne qui l'exprime.
> >
> > ________________________________________________________________
> > Welcome to the Interaction Design Association (IxDA)!
> > To post to this list ....... discuss at ixda.org
> > Unsubscribe ................ http://www.ixda.org/unsubscribe
> > List Guidelines ............ http://www.ixda.org/guidelines
> > List Help .................. http://www.ixda.org/help
> >
>
>
>
> --
> ~ will
>
> "No matter how beautiful,
> no matter how cool your interface,
> it would be better if there were less of it."
> Alan Cooper
> -
> "Where you innovate, how you innovate,
> and what you innovate are design problems"
> -------------------------------------------------------
> will evans
> user experience architect
> wkevans4 at gmail.com
> -------------------------------------------------------

--
~ will

"No matter how beautiful,
no matter how cool your interface,
it would be better if there were less of it."
Alan Cooper
-
"Where you innovate, how you innovate,
and what you innovate are design problems"
-------------------------------------------------------
will evans
user experience architect
wkevans4 at gmail.com
-------------------------------------------------------

13 Feb 2008 - 5:51pm
Todd Roberts
2005

For a pragmatic take on data viz that draws heavily from Tufte, check out
Stephen Few's Information Dashboard Design. I've found it useful for lots of
data display scanarios, regardless of whether the data were on a dashboard.
He has another book too that I haven't checked out yet but am planning to.

Interesting thought experiment Will - I'll have to try that next time I have
a data display problem.

Todd

On Feb 13, 2008 4:41 PM, W Evans <wkevans4 at gmail.com> wrote:

> Tufte is a good place to start - Envisioning Information, Visual Display
> of
> Quantitative Data, Beautiful evidence. His examples are old, but the ideas
> about how to think about the problem are par none!
>
>

13 Feb 2008 - 6:25pm
Rob Adams
2004

There's nothing wrong with data grids per se. Tufte himself recommends
the use of data tables on numerous occasions. That's why we have one
in Flex.

The problem arises when people use data tables just because they are
(supposedly) easy to design and implement when another form of
representing the data would be more effective at answering the
questions the viewer has about her data. I find it is always valuable
to ask yourself what these questions are likely to be, and whether the
information visualization you chose is an effective way of answering
them.

There is a very nice set of graphical charting components in Flex as
well, btw. Be sure to consider them in addition to DataGrid, and let
me know off list if there are other ways we can help you visualize
your customers' data.

Rob
Design Researcher, Flex product line
Adobe

On Feb 13, 2008 4:47 PM, W Evans <wkevans4 at gmail.com> wrote:
> Another thing - start by not thinking about rows of data filled with
> records. Instead start with the fundamental unit -
> A person in a HR system
> A book in a library
> A car in used car lot.
>

--
Rob
http://usereccentric.com/

14 Feb 2008 - 1:46pm
Anonymous

On Wed, Feb 13, 2008 at 6:25 PM, Rob Adams <rob at lokislabs.org> wrote:

> There's nothing wrong with data grids per se.

Technology has tendencies which affect our behavior perniciously - the way
in which we converse and express. So I don't think a data-grid can
be right or wrong, but dangerous yes. Data grids are dangerous.

The danger with data grids is two-fold

a) They imply a transactional means of interaction. You enter input fields
into a search and a datagrid binds outputs. Input entry from user, grid
bound output from computer. Stimulus, response.

b) They imply a means of information/visualization. Namely, data cells in a
regular grid.

Want an example of where the datagrid might be dangerous, but nonetheless *the
default strategy*? A normal train schedule search. I want to find a train
to go from place A to place B at time X.

*Data grid Ixd: * Input: place A, place B, time X. Press Search.
Output: rows of matching train entries. Don't find what you are looking
for? Modify A, B, or X and repeat.

*Data grid Visualization:* Data are values placed in a grid cell.
Stations, times, and everything else.

*Problems with the Data grid approach:* User must find appropriate trains by
mental calculation from data. If user doesn't find what he wants, or is off
on the times, he begins the dialogue over again.

*Non data grid IxD:* Inputs are places on a map. Train time starts now, so
no input required.

*Non data grid Visualization:* A time schedule where the horizontal axis is
time and trains are placed on it.
Just look here: http://worrydream.com/bartwidget/ (thanks to Aza Raskin
for this example)

*Solutions inspired by alternative approach:* A seamless interactive
input-output dialogue of dragging stations around which inspires more "what
if" scenarios. This solution lets people *see time and compare trains*, it
is oriented around planning your route.

Therefore, the problem with data grids like any technology is that it
curtails our thoughts and limits our expression, and suggests that all
problems are readily solved by using it.

Navid Sadikali
Agfa Healthcare
Interaction Design

15 Feb 2008 - 4:32am
sajid saiyed
2005

Hi,
I particularly like the interactive versions of complex data structures.

One specific person I would like to mention here is Ben Fry
[www.benfry.com]. I like the way he has been experimenting with
processing [www.processing.org].

Some example of his work:
http://benfry.com/salaryper/
http://benfry.com/isometricblocks/
http://benfry.com/isometricblocks/

There are many more such examples where highly complex data structures
and relationships are visualized in very interactive formats.

-sajid
www.ssdesigninteractive.com/blog

On Fri, Feb 15, 2008 at 12:16 AM, Navid Sadikali
<navid.sadikali at gmail.com> wrote:
> On Wed, Feb 13, 2008 at 6:25 PM, Rob Adams <rob at lokislabs.org> wrote:
>
> > There's nothing wrong with data grids per se.
>
>
> Technology has tendencies which affect our behavior perniciously - the way
> in which we converse and express. So I don't think a data-grid can
> be right or wrong, but dangerous yes. Data grids are dangerous.
>
> The danger with data grids is two-fold
>
> a) They imply a transactional means of interaction. You enter input fields
> into a search and a datagrid binds outputs. Input entry from user, grid
> bound output from computer. Stimulus, response.
>
> b) They imply a means of information/visualization. Namely, data cells in a
> regular grid.
>
> Want an example of where the datagrid might be dangerous, but nonetheless *the
> default strategy*? A normal train schedule search. I want to find a train
> to go from place A to place B at time X.
>
> *Data grid Ixd: * Input: place A, place B, time X. Press Search.
> Output: rows of matching train entries. Don't find what you are looking
> for? Modify A, B, or X and repeat.
>
> *Data grid Visualization:* Data are values placed in a grid cell.
> Stations, times, and everything else.
>
> *Problems with the Data grid approach:* User must find appropriate trains by
> mental calculation from data. If user doesn't find what he wants, or is off
> on the times, he begins the dialogue over again.
>
> *Non data grid IxD:* Inputs are places on a map. Train time starts now, so
> no input required.
>
> *Non data grid Visualization:* A time schedule where the horizontal axis is
> time and trains are placed on it.
> Just look here: http://worrydream.com/bartwidget/ (thanks to Aza Raskin
> for this example)
>
> *Solutions inspired by alternative approach:* A seamless interactive
> input-output dialogue of dragging stations around which inspires more "what
> if" scenarios. This solution lets people *see time and compare trains*, it
> is oriented around planning your route.
>
> Therefore, the problem with data grids like any technology is that it
> curtails our thoughts and limits our expression, and suggests that all
> problems are readily solved by using it.
>
>
> Navid Sadikali
> Agfa Healthcare
> Interaction Design
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> Unsubscribe ................ http://www.ixda.org/unsubscribe
> List Guidelines ............ http://www.ixda.org/guidelines
> List Help .................. http://www.ixda.org/help
>

14 Feb 2008 - 10:13pm
Scott McDaniel
2007

On Wed, Feb 13, 2008 at 6:25 PM, Rob Adams <rob at lokislabs.org> wrote:
> There's nothing wrong with data grids per se. Tufte himself recommends
> the use of data tables on numerous occasions. That's why we have one
> in Flex.
>
> The problem arises when people use data tables just because they are
> (supposedly) easy to design and implement when another form of
> representing the data would be more effective at answering the
> questions the viewer has about her data. I find it is always valuable
> to ask yourself what these questions are likely to be, and whether the
> information visualization you chose is an effective way of answering
> them.

To comment on this - yes, exactly. They are a versatile tool, and
that's very much it - our IAs who are just stretching their arms into
RIA-related things such as Flex and various Ajax frameworks will often
reach for the spice on the rack which is most readily available,
easily identifiable with previous experience, and more easily
understood by developers who are newcomers to the framework as well.
In some ways this resembles Alan Cooper's "Dancing Bear" example -
hey, we can do it at all! In extended effect - at least in my limited
local experience - it does get overused in design,
but also honestly has provided more depth to desired application
goals, even if we should seek to explore better suited options.

Scott
--
'Life' plus 'significance' = magic. ~ Grant Morrison

Syndicate content Get the feed