Conrad Taylor | 10 Dec 14:00 2004
Picon
Picon

InfoD-Cafe: Table markup: style or structure?

Hello all,

I'm starting into a project in which a magazine publishing
company wants to insert in its pages tables of data that
have been generated from a database.  They also are of
the opinion that they want the data to be encoded in XML
so that it can be re-used in other publishing outlets,
primarily online.

Don't worry, I'm on top of this and exploring several
alternative methods for the process. My purpose in writing
today is much more idle and speculative!

One of the ideas I've been batting around with the other
consultants is that we get the client's database to write
a "canonical" XML version of the data that is as light as
possible on formatting information, and then transform
this for print production into a version that is rich
in procedural/layout-oriented mark-up for import into
the page make-up application.  (If you're interested,
the target application is InDesign and I'm favouring
Indesign Tagged Text Format for the import format,
rather than XML.)

Musing about what is "style" and what is "structure"
prompted me to raise the following question with my
colleagues -- to tease their brains as much as for
any other reason:

>    [Tables] are a paradoxical thing:  is the gridlike
(Continue reading)

Conrad Taylor | 11 Dec 02:17 2004
Picon
Picon

InfoD-Cafe: Re:InfoD Invisibility

Damien Braniff wrote:

>    The problem with Œgood¹ design is that, generally, it is
>    invisible ­ people don¹t talk about it, simply take it
>    for granted. It¹s like going to a restaurant - you may well
>    tell friends if the food/service was brilliant/terrible but
>    otherwise it¹s usually OK/OK+.  We expect a certain level
>    of service (and a certain standard of documentation) and
>    there¹s a band that, if it falls within that band, it¹s
>    accepted - outside that band we praise or condemn.

That suggests a certain clinical quality of efficiency both
to the restaurant service and the documentation.  I just came
back from dinner with Jane Teather (Cafe member) in a place
that was always quite idiosyncratic in its decor and cuisine
and has been given an additional pleasant twist through a kind
of Moroccanisation of both, with carpets on the ceiling and
ostrich on the mashed potato.  When that happens, you sit up
and pay attention!

It's not often that documentation dares to be witty or funny,
but when it does I likewise pay attention.  For example, Donald
Knuth's original TeX and METAFONT documentation.  You keep reading
stuff like "METAFONT is like Sesame Street in reverse: not that the
program was brought to you by the letter A, but the letter A was
brought to you by a program..."  because you don't want to miss
the next joke, and in the process you learn something.  Most
companies shy away from humour and flavour (or even humor and
flavor), and then it is up to authors like Robin Williams and
Sandee Cohen and Elisabeth Castro to fill the gap with highly
(Continue reading)

Curtis Clark | 11 Dec 05:34 2004
Picon
Picon

Re: InfoD-Cafe: Table markup: style or structure?

on 2004-12-10 05:00 Conrad Taylor wrote:
>>    Therefore, I argue that it is possible to ensure the
>>    data integrity of all of the data in a table without
>>    using one shred of table cell/row/column/header markup.

Interesting viewpoint: the semantics of a table cell is its contents, 
its x-category (column), and its y-category (row) *only*. And of course 
tables can be and often are productively transposed. It would be 
interesting to imagine a style language that would specify everything else.

-- 
Curtis Clark                  http://www.csupomona.edu/~jcclark/
Web Coordinator, Cal Poly Pomona                 +1 909 979 6371
Professor, Biological Sciences                   +1 909 869 4062
___________________________________________________________________

Use the following address to post a message to all subscribers: 
 infodesign-cafe <at> list.informationdesign.org

To subscribe, unsubscribe or change your options, visit:
 http://list.InformationDesign.org/mailman/listinfo/infodesign-cafe

For all Information Design matters:
 http://InformationDesign.org

Problems? Write to:
 InfoDesign-Cafe-Admin <at> list.InformationDesign.org
___________________________________________________________________

(Continue reading)

Conrad Taylor | 11 Dec 07:32 2004
Picon
Picon

InfoD-Cafe: Accessible Web development: jobs in London

Hi, I've been asked to publicize a couple of jobs at a
Web-site development co-operative in London.  From the
contacts I've had with Poptel, I'd say it would be a
nice place to work.

Conrad

>    Accessible Web Developer Post, Poptel Technology, London
>
>    Poptel Technology is a small, ambitious employee owned company
>    with a strong track record in a competitive marketplace.
>    Our clients include several major trade unions, charities
>    and NGOs.
>
>    As a UK leader in the field of web accessibility, we have
>    developed several award winning sites. We believe that
>    web accessibility must become mainstream, therefore we
>    build all our websites to be accessible.
>
>    We are seeking to recruit a dynamic team player who has
>    excellent attention to detail and a commitment to web
>    accessibility. This is a pivotal role where you will
>    ensure high standards are maintained and have a chance
>    to build future proof websites.
>
>    Please visit our website at 
><http://www.popteltechnology.coop/>http://www.popteltechnology.coop
>    for more details.  We are also recruiting a Project Manager
>    who will be responsible for overseeing accessible website
>    projects. If you have any queries please contact
(Continue reading)

Richard | 11 Dec 21:00 2004
Picon

Re: InfoD-Cafe: Table markup: style or structure?

Conrad's is an interesting question.

>    [Tables] are a paradoxical thing:  is the gridlike
>    structure of the table truly semantic, or is it
>    just a presentation format?
>    Or is it something in between?
>

On one hand, I'd speculate that the tabular presentation is is just 
that, a choice of layout. One can get similar meaning out of simple 
paragraph headings with text. Tables are more dense and support 
different scanning strategies. The choice of layout responds to scales 
such as length of text under each heading, prevalence of numerical or 
other coded values, a need to quickly locate a given value as opposed 
to a need to form discourse.

On the other hand, a table might seem to exploit the column-row 
coordinates to such a degree that there really is no other way to 
communicate the meaning of a given cell value in its relation to 
others. Not only are the cell coordinates intrinsic to the meaning of 
the value in the cell, but the representation of the intersection of 
the singular (cell) and the common or repetitious (column and/or row) 
is unique and powerful. The varying choice of categories for the axes 
(interval, set, order) can have a great effect on the interpretation of 
the values. The very density of presentation favors a play of readings 
that more extended layouts might hinder. I would guess that Gestalt 
principles are at work, as well as rhythms of perception.

So my answer is both or either, depending.
:)
(Continue reading)

Kimball, Miles | 11 Dec 21:27 2004
Picon

RE: InfoD-Cafe: Table markup: style or structure?

There's a new book with a very good chapter on the semiotics of table
design: Lee Brasseur's Visualizing Technical Information: A Cultural
Critique (Baywood Publishing, ISBN 0-89503-240-6). Stimulating reading
for anyone interested in information graphics, in particular. 

Miles Kimball

-----Original Message-----
From: infodesign-cafe-bounces <at> list.informationdesign.org
[mailto:infodesign-cafe-bounces <at> list.informationdesign.org] On Behalf Of
Conrad Taylor
Sent: Friday, December 10, 2004 7:01 AM
To: InfoDesign-Cafe
Subject: InfoD-Cafe: Table markup: style or structure?

Hello all,

I'm starting into a project in which a magazine publishing
company wants to insert in its pages tables of data that
have been generated from a database.  They also are of
the opinion that they want the data to be encoded in XML
so that it can be re-used in other publishing outlets,
primarily online.

Don't worry, I'm on top of this and exploring several
alternative methods for the process. My purpose in writing
today is much more idle and speculative!

One of the ideas I've been batting around with the other
consultants is that we get the client's database to write
(Continue reading)

Mark Barratt | 11 Dec 22:09 2004

Re: InfoD-Cafe: Table markup: style or structure?

Conrad Taylor wrote:

>>    [Tables] are a paradoxical thing:  is the gridlike
>>    structure of the table truly semantic, or is it
>>    just a presentation format?
>>    Or is it something in between?
>
>
Like Richard my answer is 'both or either, depending'. I work a lot with 
XML and other kinds of structured text and often hear the mantra about 
XML enabling the separation of semantics and presentation. This tends to 
be an item of faith among its adherents, not susceptible to rational 
enquiry.

Tables are of two kinds: the first is actually a list which it happens 
to be more convenient to present visually as a table, and there are 
plenty of these. The second derives meaning from the fact that data is 
presented at row/column intersections.

I'm not sure why some of the ML crew try to pretend that *nothing* which 
implies the visual appearance of a text should be embedded in markup - 
it's a much bolder claim than was made by Charles Goldfarb, who 
developed SGML. He talked about logical units and generalized markup in 
which processing is separated from description, and in the examples in 
his book uses what he calls 'logical units' such as lists, paragraphs 
and headings which are strongly tied to written (rather than spoken) 
language. In written language, as in mathematics and data processing, 
the table is an important archetype.

I have heard it claimed that markup should be 'pure' enough to allow the 
(Continue reading)

Mick McAllister | 12 Dec 14:33 2004

Re: InfoD-Cafe: Table markup: style or structure?


>This tends to be an item of faith among its adherents, not susceptible to 
>rational enquiry.
Indeed. Having been dragged, half unwillingly, into "form" from a lifetime 
interest in "content," I've found this particular bit of dogma very 
frustrating to deal with. I have countered sometimes with examples of 
street signs, such the perennial "Slow children playing" and my own 
favorite, "School CAUTION Zone."

If Mindy Sue dots her 'I's with plump little valentine hearts, this is 
content, not "form." If my boss leaves me a note saying "SEE ME FIRST THING 
IN THE MORNING, YOU NITWIT!" the caps are content, not form. And the word 
"dear" at the beginning of almost any letter is, like it or not, form. Not 
content.

M

___________________________________________________________________

Use the following address to post a message to all subscribers: 
 infodesign-cafe <at> list.informationdesign.org

To subscribe, unsubscribe or change your options, visit:
 http://list.InformationDesign.org/mailman/listinfo/infodesign-cafe

For all Information Design matters:
 http://InformationDesign.org

Problems? Write to:
 InfoDesign-Cafe-Admin <at> list.InformationDesign.org
(Continue reading)

Curtis Clark | 12 Dec 19:52 2004
Picon
Picon

Re: InfoD-Cafe: Table markup: style or structure?

on 2004-12-11 13:09 Mark Barratt wrote:
> Like Richard my answer is 'both or either, depending'. I work a lot with 
> XML and other kinds of structured text and often hear the mantra about 
> XML enabling the separation of semantics and presentation. This tends to 
> be an item of faith among its adherents, not susceptible to rational 
> enquiry.

Perhaps better put as structure vs. presentation, and this is 
emphatically not the same thing as separating content and form, even if 
some ML people get it wrong.

To me, the best measure comes from repurposing a document. What parts 
need to be conveyed in any format, and what parts are specific to one 
format? Because one of my tasks is to make accessible web pages, I'll 
sometimes ask "what does a blind person need to hear?" In the case of 
Mick's boss's capitalization, that is already well-established by 
convention--all-caps in email is often called SHOUTING. If 
capitalization were key to the meaning, a blind person would be out of 
luck, and if an elevated voice volume with strain harmonics were 
required, it couldn't be sent as plain text, but fortunately a markup of 
<shouting>could be interpreted as necessary in different presentational 
contexts</shouting>.

--

-- 
Curtis Clark                  http://www.csupomona.edu/~jcclark/
Web Coordinator, Cal Poly Pomona                 +1 909 979 6371
Professor, Biological Sciences                   +1 909 869 4062
___________________________________________________________________

Use the following address to post a message to all subscribers: 
(Continue reading)

Mick McAllister | 12 Dec 20:10 2004

Re: InfoD-Cafe: Table markup: style or structure?


>Perhaps better put as structure vs. presentation, and this is emphatically 
>not the same thing as separating content and form, even if some ML people 
>get it wrong.
I'd say definitely better put, but I still balk at the notion that they are 
two things rather than aspects of "one thing." One reason "structure vs 
presentation" works better is that neither is the "container." In the 
form/content model, the content is (obviously) contained, and the form 
"contains" it. The idea somehow makes me think of a Maria Martinez pot, 
which has no "content," just form. Or a Bernini sculpture, which has 
meaningless content (the marble under the surface), or a Blake text....

In the structure/presentation model, neither the structure nor the 
presentation "contains" the "content," which is some sort of other entity 
"structured" and "presented." That works for Blake, and for Bernini, and 
for Martinez. Except... for Bernini and Martinez there is still no 
meaningful "content," just "structure" and "presentation." So, can you have 
structure and presentation without content... in texts, I mean? How about 
concrete poetry?

Too much coffee.
M

___________________________________________________________________

Use the following address to post a message to all subscribers: 
 infodesign-cafe <at> list.informationdesign.org

To subscribe, unsubscribe or change your options, visit:
 http://list.InformationDesign.org/mailman/listinfo/infodesign-cafe
(Continue reading)


Gmane