To UL or not to UL

[Adrienne Boswell]

I am redoing a site and I have some data that was originally in list
form, and I am thinking that it should be for a table. Would someone
take a look
<http://atlas.nextblock.com/files/beta2/categories.asp?st=100&zip=18>?
I am interested in the categories, should these be in a table or three
UL columns? Is this really tabular data, or is it three lists?

Please note that almost none of the links work, and there are a lot of
testing notes strewn about - this is some serious beta going on.

TIA

 

[Beauregard T. Shagnasty]

I am redoing a site and I have some data that was originally in list
form, and I am thinking that it should be for a table. Would someone
take a look
<http://atlas.nextblock.com/files/beta2/categories.asp?st=100&zip=18>?
I am interested in the categories, should these be in a table or
three UL columns? Is this really tabular data, or is it three lists?

You're talking about: Stores Restaurants Services ?

Well, if "Banks" and "Bars & Pubs" are a directly related entity of
"Automotive & Service", then it would be a table.

This is three lists, not tabular data. Notice the three columns are not
even the same length. A display of tabular data would normally have all
columns the same length.

Tables are

Name Address City State/Province Postal Code Telephone ...

 

[Adrienne Boswell]

Beauregard T. Shagnasty wote:

You're talking about: Stores Restaurants Services ?

Well, if "Banks" and "Bars & Pubs" are a directly related entity of
"Automotive & Service", then it would be a table.

This is three lists, not tabular data. Notice the three columns are not
even the same length. A display of tabular data would normally have all
columns the same length.

Tables are

Name Address City State/Province Postal Code Telephone ...

Thank you. I knew that, but, I wanted to make sure. Four eyes are
better than two.

 

[dorayme]

Beauregard T. Shagnasty wote:

Thank you. I knew that, but, I wanted to make sure. Four eyes are
better than two.

How about 3 brains? The point about the example of name, address
etc is that these categories form a class of attributes that
apply to individuals. The individual can be understood to have
all the attributes on a row. There is nothing, for example, that
Accessories, American and Accounting applies to that is being
conveyed by these lists

 

[Adrienne Boswell]

dorayme wote:

How about 3 brains? The point about the example of name, address
etc is that these categories form a class of attributes that
apply to individuals. The individual can be understood to have
all the attributes on a row. There is nothing, for example, that
Accessories, American and Accounting applies to that is being
conveyed by these lists

Yes, three brains _are_ better than one... I think I've been reading
too much Winnie the Pooh to my two and a half year old... I found
myself today with Very Little Brain. Thanks for that great example.

 

[dorayme]

The point about the example of name, address
etc is that these categories form a class of attributes that
apply to individuals. The individual can be understood to have
all the attributes on a row. There is nothing, for example, that
Accessories, American and Accounting applies to that is being
conveyed by these lists

And it is really more general than this... there needs to be a
relationship between the cells which the structure of the table
conveys. The relationship can be internal: Product, size, colour,
cost etc. I call this internal because the first col might name
or describe the individual to which the other cols assign
attributes. But there are other more subtle things that tables
can do, the individual that ties a set of properties might be
understood rather than named as an item. Likely clusters of
symptoms that are life threatening, each row displaying the
cluster across the cols, in each of which is a type of attribute
(eg., blood type). The individual here is something more
abstract, true - folk at risk of something or other...

I doubt that there is any definition of what a table "really" is
that would capture all the things everyone would reasonably want.
But there are paradigm cases. And, of course, use of them for
layout is just too far from these paradigms for folks around here
to stomach...

 

[dorayme]

Winnie the Pooh to my two and a half year old... I found
myself today with Very Little Brain.

There is nothing more pleasurable than reading to little ones...
Among many things I would read Dr. Seuss books... and the rhythm
of the story about a dog still echoes in my brain, it was bought
in a pet store, never, to its shame, learnt to dig but then got
the knack and all hell then broke loose...

 

[Joe]

I doubt that there is any definition of what a table "really" is
that would capture all the things everyone would reasonably want.

Like I said once before,
...." if you can put meaningful headings on it, it's a table."

As a working definition, it'll do me.

 

[Jonathan N. Little]

Like I said once before,
..." if you can put meaningful headings on it, it's a table."

As a working definition, it'll do me.

No I would say if the data is arrange in a grid because the data has a
row|column relationship then its a table, only one column it's a list.

 

[dorayme]

No I would say if the data is arrange in a grid because the data has a
row|column relationship then its a table, only one column it's a list.

Joe, you can put meaningful heading on simple table layouts, left
navigation and right content. That does not make it a real table
as is commonly meant in alt.html. So as a definition, it would
need to be sharpened up.

Jonathan, it depends what is meant by relationship and data. Data
in a 2 col table with only one row, used for layout, would
satisfy this unless the definition was longer and more complex.

 

[Jonathan N. Little]

dorayme wrote:

Jonathan, it depends what is meant by relationship and data. Data
in a 2 col table with only one row, used for layout, would
satisfy this unless the definition was longer and more complex.

|Widget | Doohickeys |
---+-------+------------+
Qty| 4 | 3 |

Only one row, but rank and file relationship conveys some information
based on the intersection tells me it's a table...


Logo | Apple Pie |
-----+-----------+
Links| Toads |

Table where arragement has no meaning* other than placement is not a table.

*(unless your smoking something very strong whereby world events can be
predicted in the various shapes of discovered belly-button lint)

 

[dorayme]

dorayme wrote:


|Widget | Doohickeys |
---+-------+------------+
Qty| 4 | 3 |

Only one row, but rank and file relationship conveys some information
based on the intersection tells me it's a table...


Logo | Apple Pie |
-----+-----------+
Links| Toads |

Table where arragement has no meaning* other than placement is not a table.

*(unless your smoking something very strong whereby world events can be
predicted in the various shapes of discovered belly-button lint)

Your post with the diags and explanations do not meet the idea of
a strict general definition. I suspect there is just a whole
collection of posts like yours with little diags and sentences
and they all add up to the general idea.

 

[Jonathan N. Little]

Your post with the diags and explanations do not meet the idea of
a strict general definition. I suspect there is just a whole
collection of posts like yours with little diags and sentences
and they all add up to the general idea.

Didn't know the idea was to create a strict definition but communicate a
concept. Anyway I'm an artist, what you you expect!

 

[dorayme]

Your post with the diags and explanations do not meet the idea of
a strict general definition. I suspect there is just a whole
collection of posts like yours with little diags and sentences
and they all add up to the general idea.

Didn't know the idea was to create a strict definition but communicate a
concept. Anyway I'm an artist, what you you expect![/QUOTE]

You are a good man, Jonathan, you do as well as any of us to get
the idea across. I just made a remark a while back that it is
probably an illusion to think one can put the idea to cover all
we want to cover in a simple definition. I am not going to
suddenly give this up because there are a lots intelligent ways
of communicating various parts of the general idea. It is simple
enough to know when a table is being used for layout without need
of a definition. Giving one, opens one up to counter-examples. I
am saying thatthis process is probably never ending and not
because of anyone's stupidity.

 

[Andy Dingley]

Joe, you can put meaningful heading on simple table layouts, left
navigation and right content. That does not make it a real table
as is commonly meant in alt.html. So as a definition, it would
need to be sharpened up.

How about this
"If you can meaningful headings on your table, it ought to be a table."

Of course you can caption things that aren't tables, but then they're
_obviously_ not even trying to be tables, so there's no confusion. The
test is intended to filter out those "layout tables" where they're
constructed as a <table>, but ought not to be.


(I agree with Joe's rule of thumb BTW and have suggested it myself
before)

 

[mbstevens]

Like I said once before,
...." if you can put meaningful headings on it, it's a table."

As a working definition, it'll do me.

Joe and Andy,
I think your heading requirement is sufficient, but not necessary.

Consider matrix algebra. In many cases it would be difficult to
assign meaningful headings, but each cell of data is definitely
in a particular relationship to other cells in a way that the
position of the cell makes a difference. You could probably come
up with a case that even applied over a single dimension.

 

[Jonathan N. Little]

I just made a remark a while back that it is
probably an illusion to think one can put the idea to cover all
we want to cover in a simple definition. I am not going to
suddenly give this up because there are a lots intelligent ways
of communicating various parts of the general idea. It is simple
enough to know when a table is being used for layout without need
of a definition. Giving one, opens one up to counter-examples. I
am saying thatthis process is probably never ending and not
because of anyone's stupidity.

I think the root of the problem is we humans, Martians included, desire
to categorize the turbulent, fuzzy, swirly world in black and white and
delineated little boxes. Finding the precise point where that blur goes
from white to black will always a problem, it is the task for scientists
and artists!

 

[dorayme]

"Andy Dingley <[email protected]>"

How about this
"If you can meaningful headings on your table, it ought to be a table."

A table used for layout with 2 cols and only one row could have
meaningful headings ("navigation", "content"). So this does not
cut it. The idea of trying to tie the content of the cells in the
rows with respect to cols is more detailed. but even this has
trouble with a 2 col, 1 row table used for layout. I won't go
into it unless someone is interested...

 

[Joe]

Joe, you can put meaningful heading on simple table layouts, left
navigation and right content. That does not make it a real table
as is commonly meant in alt.html. So as a definition, it would
need to be sharpened up.

I never thought of that. Bugger.

 

[Joe]

No I would say if the data is arrange in a grid because the data has a
row|column relationship then its a table, only one column it's a list.

No such thing as a single-column table. That's why my def says
"headingS".