Re: [WSG] Divs for tabular data

2011-12-05 Thread Grant Bailey

David,

I'm not sure about DOM issues but in my opinion, use of tables is 
consistent with accessibility and semantic requirements if an external 
style sheet is used to style the tabular data. Usually, some use of divs 
is required if you want to present the tabular data in a particular way.


Regards,

Grant Bailey

On 5/12/2011 6:22 PM, David McKinnon wrote:
OK, so I'm working on a project in which the developers are laying out 
tabular data using divs.
The site is using the 960 CSS grid system so making the 'tables' work 
just means applying the appropriate class to align each div/table cell 
to the grid.

They say this is good because:

 1. It's fast
 2. They can manipulate the resulting DOM much more easily than they
could with a table
 3. Developers find it easier to, say, add or remove columns from the
tables, without having to edit the code all the way down the table
(no wysiwyg editors here!)

To me this doesn't seem very good because:

 1. It's not very semantic (although they've used micro data in the
class names for some divs)
 2. It doesn't seem very accessible -- I might be wrong about this,
but to me good semantics is foundational to accessibility
 3. There's a lot of markup -- I know tables aren't exactly light on
code, but they seem quite light and efficient in comparison
 4. It doesn't seem to me like the code will be very easy to maintain
for anyone but the developers.


The lead developers assure me that this is good practice for speed and 
efficiency, but I'm not convinced.
Nevertheless, I don't want to be advocating tables as best practice if 
they aren't.


What do you think? Are tables too hard for the real world in large 
sites or web apps where large amounts of DOM manipulation is required? 
Or have these guys taken the 'Tables are bad' thing a bit too far?


Kind regards,
David

***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
*** 



***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
***

Re: [WSG] Divs for tabular data

2011-12-05 Thread Patrick H. Lauke
Tabular data should be marked up as actual tables. Anything else is a 
perversion of standards. Screenreaders, to take the technical extreme of the 
spectrum, have special controls and functionalities for users to navigate 
tables (moving between rows/columns directly, getting info on associated 
headers, getting a count of rows/cells and where the current focus is) that are 
impossible to replicate with anything other than table markup. So, as much as I 
sympathise with the developers' it's easier take, it won't cut it I'm afraid.

P

--
Patrick H. Lauke


On 5 Dec 2011, at 07:22, David McKinnon david...@mac.com wrote:

 OK, so I'm working on a project in which the developers are laying out 
 tabular data using divs.
 The site is using the 960 CSS grid system so making the 'tables' work just 
 means applying the appropriate class to align each div/table cell to the grid.
 They say this is good because:
 It's fast
 They can manipulate the resulting DOM much more easily than they could with a 
 table
 Developers find it easier to, say, add or remove columns from the tables, 
 without having to edit the code all the way down the table (no wysiwyg 
 editors here!)
 To me this doesn't seem very good because:
 It's not very semantic (although they've used micro data in the class names 
 for some divs)
 It doesn't seem very accessible -- I might be wrong about this, but to me 
 good semantics is foundational to accessibility
 There's a lot of markup -- I know tables aren't exactly light on code, but 
 they seem quite light and efficient in comparison
 It doesn't seem to me like the code will be very easy to maintain for anyone 
 but the developers.
 
 The lead developers assure me that this is good practice for speed and 
 efficiency, but I'm not convinced.
 Nevertheless, I don't want to be advocating tables as best practice if they 
 aren't.
 
 What do you think? Are tables too hard for the real world in large sites or 
 web apps where large amounts of DOM manipulation is required? Or have these 
 guys taken the 'Tables are bad' thing a bit too far?
 
 Kind regards,
 David
  
 
 ***
 List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
 Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
 Help: memberh...@webstandardsgroup.org
 ***


***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
***


Re: [WSG] Divs for tabular data

2011-12-05 Thread Hassan Schroeder

On 12/4/11 11:22 PM, David McKinnon wrote:

OK, so I'm working on a project in which the developers are laying out tabular 
data using divs.



They say this is good because:

 1. It's fast


Compared to what?


 2. They can manipulate the resulting DOM much more easily than they could with 
a table


Would love to see examples.


 3. Developers find it easier to, say, add or remove columns from the tables, 
without having to
edit the code all the way down the table (no wysiwyg editors here!)


?all the way down the table? So, a hard-coded data table, rather
than a set of rows populated by looping through a data source?

That's just sad. Perhaps you can gently hint that it's now the 21st
Century. And these people call themselves developers?

 What do you think? Are tables too hard for the real world ...

Stop! My sides!! :-)

FWIW,
--
Hassan Schroeder - has...@webtuitive.com
webtuitive design ===  (+1) 408-621-3445   === http://webtuitive.com
http://about.me/hassanschroeder
twitter: @hassan
  dream.  code.


***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
***



Re: [WSG] Divs for tabular data

2011-12-05 Thread Ted Drake
I remember seeing this for the first time. I was asked by a backend engineer
to help them fix a layout issue in a data table. When I looked at the source
code, the page was a jumble of absolutely positioned cells to look like a
data table. I shook my head and said he was on his own. I wasn't about to
wade into that mess.

Positioned divs would be completely inaccessible. It would be a string of
numbers with no context to what they mean.

YUI makes a great data table module.
http://yuilibrary.com/yui/docs/datatable/
You can easily make your data table sortable, generate charts, make it super
accessible and more. You can't do the same with positioned divs.

ted


On 12/5/11 8:44 AM, Hassan Schroeder has...@webtuitive.com wrote:

 On 12/4/11 11:22 PM, David McKinnon wrote:
 OK, so I'm working on a project in which the developers are laying out
 tabular data using divs.
 
 They say this is good because:
 
  1. It's fast
 
 Compared to what?
 
  2. They can manipulate the resulting DOM much more easily than they could
 with a table
 
 Would love to see examples.
 
  3. Developers find it easier to, say, add or remove columns from the tables,
 without having to
 edit the code all the way down the table (no wysiwyg editors here!)
 
 ?all the way down the table? So, a hard-coded data table, rather
 than a set of rows populated by looping through a data source?
 
 That's just sad. Perhaps you can gently hint that it's now the 21st
 Century. And these people call themselves developers?
 
 What do you think? Are tables too hard for the real world ...
 
 Stop! My sides!! :-)
 
 FWIW,



***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
***



Re: [WSG] Divs for tabular data

2011-12-05 Thread David McKinnon
Thanks for the sanity check everyone :)
I was worried that I'd missed something.

I'd seen the YUI table module, thanks Ted, and I know of at least two jQuery 
plugins that do the same thing so yeah, I'm not sure why they wouldn't use one 
of those.
They're saying it's faster to use divs compared to just tables, Hassan, but 
with the amount of extra code required it seems counter-productive. I'll see if 
I can get you an example. The grid is mostly hard-coded, but each row/cell is 
created dynamically so i'm not sure what the advantage is -- more investigation 
required!

And no laughing please, this is serious giggle /



On 06/12/2011, at 4:14 AM, Ted Drake wrote:

 I remember seeing this for the first time. I was asked by a backend engineer
 to help them fix a layout issue in a data table. When I looked at the source
 code, the page was a jumble of absolutely positioned cells to look like a
 data table. I shook my head and said he was on his own. I wasn't about to
 wade into that mess.
 
 Positioned divs would be completely inaccessible. It would be a string of
 numbers with no context to what they mean.
 
 YUI makes a great data table module.
 http://yuilibrary.com/yui/docs/datatable/
 You can easily make your data table sortable, generate charts, make it super
 accessible and more. You can't do the same with positioned divs.
 
 ted
 
 
 On 12/5/11 8:44 AM, Hassan Schroeder has...@webtuitive.com wrote:
 
 On 12/4/11 11:22 PM, David McKinnon wrote:
 OK, so I'm working on a project in which the developers are laying out
 tabular data using divs.
 
 They say this is good because:
 
 1. It's fast
 
 Compared to what?
 
 2. They can manipulate the resulting DOM much more easily than they could
 with a table
 
 Would love to see examples.
 
 3. Developers find it easier to, say, add or remove columns from the tables,
 without having to
edit the code all the way down the table (no wysiwyg editors here!)
 
 ?all the way down the table? So, a hard-coded data table, rather
 than a set of rows populated by looping through a data source?
 
 That's just sad. Perhaps you can gently hint that it's now the 21st
 Century. And these people call themselves developers?
 
 What do you think? Are tables too hard for the real world ...
 
 Stop! My sides!! :-)
 
 FWIW,
 
 
 
 ***
 List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
 Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
 Help: memberh...@webstandardsgroup.org
 ***
 



***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
***



[WSG] Divs for tabular data

2011-12-04 Thread David McKinnon
OK, so I'm working on a project in which the developers are laying out tabular 
data using divs.
The site is using the 960 CSS grid system so making the 'tables' work just 
means applying the appropriate class to align each div/table cell to the grid.
They say this is good because:
It's fast
They can manipulate the resulting DOM much more easily than they could with a 
table
Developers find it easier to, say, add or remove columns from the tables, 
without having to edit the code all the way down the table (no wysiwyg editors 
here!)
To me this doesn't seem very good because:
It's not very semantic (although they've used micro data in the class names for 
some divs)
It doesn't seem very accessible -- I might be wrong about this, but to me good 
semantics is foundational to accessibility
There's a lot of markup -- I know tables aren't exactly light on code, but they 
seem quite light and efficient in comparison
It doesn't seem to me like the code will be very easy to maintain for anyone 
but the developers.

The lead developers assure me that this is good practice for speed and 
efficiency, but I'm not convinced.
Nevertheless, I don't want to be advocating tables as best practice if they 
aren't.

What do you think? Are tables too hard for the real world in large sites or web 
apps where large amounts of DOM manipulation is required? Or have these guys 
taken the 'Tables are bad' thing a bit too far?

Kind regards,
David
 


***
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
***