[thelist] Site redirect check : old browser

aardvark roselli at earthlink.net
Tue Jun 5 07:24:17 CDT 2001


> From: "Mark Cheng" <mark.cheng at ranger.com.au>
> 
> I really appreciate the time you've spent.  couple of points and the
> beginning of a debate?? :)

hope so...

> >no there wasn't...were you using a document.write?
> 
> Actually there should be a button on the bottom of the nobrows page -
> "take me to the site anyway" or some such thing.

so, was that supposed to be written by a document.write, or did 
the page just crap out before rendering, or were you missing 
<form> tags?

oh yeah, look at that, there's no <form></form>... you really need 
to validate even your old browser page...  and while you're at it, it 
would be easier to validate if you had a DTD at the top of the page...

you need to fix that...

so why did you choose *not* to make that page standards-
compliant?  and don't use the browser as an excuse...

(http://validator.w3.org/check?uri=http%3A%2F%2Fsvc027.bne246d
.server-web.com%2Fnobrows.html&doctype=Inline)

> Fair point.  The wording should be clearer - browsers should include
> aural browsers etc.  thanks for the comments

overall, it needs to be more clear... the WaSP text is pretty good, 
although i still find it written from a haughty perspective... but at 
leas then you can blame them for poor wording...

> Actually, what I am doing is coding to standards designed to separate
> style from content (HTML 4.01, CSS (including CSSP), ECMA and DOM). 
> My objective is to fully separate design from content (that includes
> using no tables to achieve positioning of major design elements) to
> get the massive benefit that brings from a maintenance and redesign
> point of view.

maybe i'm the only one who thinks this way, but sites like evolt.org 
are *completely* separating style from content... the data is 
wrapped into a template, and the articles contain no HTML other 
than structural HTML to mark-up the article... we changed 
templates and the whole thing was a new design... and we could 
slap a new template on there and it would be new again...

so how is the style not separated from the content on evolt.org?

i understand your argument, but it parrots the ALA/WaSP 
arguments too much, without addressing the holes in their 
arguments... why did *you* choose to take such a radical 
approach?  do you understand that you can write to HTML 2.0 and 
still have content and style separate?  in fact, if it was truly 
separate, you should be able to slap a new template on that page 
and show us...

with something like evolt.org, i can go in and select the content of 
an article in its entirity without getting any layout or styling 
information... i can't select the entire content of your copy without 
getting the <div>s used to lay it out... how is that separate?

> as a newbie, from my reading of various articles (including evolt and
> A List Apart),  life is a lot simpler doing that than trying to get
> the exact same layout on NN4x and ie, let alone anything before that.

i'd argue that's just the easy way out... it's not as hard to do cross-
browser development as most supporters of this approach 
purport... if you're trying to do major DHTML trickery, yeah, but 
otherwise i don't seem to be having that problem... and i'm cranking 
out lots of sites...

and besides, if you are trying to get the *exact* same layout in NN 
and IE, you're missing the point... but now you've ensured you 
*can't* get the same layout, and shroud that failure under the 
argument of the WaSP initiative...  nah, sorry, that's too easy... i 
view it as a cop-out...

> I am coding beyond concern for older browsers (actually opera 511 has
> probs with some of the code in the site as well!).  However, I care
> about those older browsers - enough so that they don't try and render
> the "latest standards compliant" site.  I don't want them to get JS

how is that caring?  you've told them to bugger off... if you cared, 
you'd let *them* decide, not dictate to them...

> errors and I definitely don't want their browser to crash if they hit
> the site.  However, if they choose to try and render it - then they

heh, if the JS worked and was conditionalized based on capability, 
you should have no such concerns... and if it weren't this non-
tabled layout, would you even need any JS at all?  then wouldn't 
the older browsers be peachy, as well as new ones?  and couldn't 
it still be compliant?

> have made the choice.  If they don't want to upgrade their browsers
> for free - I'm not forcing them to - they can go to the text only site
> and get a subset of the info available.

actually, by the very points i've mentioned, they can't... think of all 
the time you're spending on this alternate page and the conditional 
code to handle it... think of the testing... wouldn't it have been 
easier to just create a simple tabled layout after all?  and since 
there's nothing terribly complex in this design, a simple table could 
have sufficed... (btw, i'm not altogether sure it's rendering in IE5 as 
you intended, the pictures are stacked in the upper left corner, and 
in content pages the headers are clipped)

btw, let's not forget that not everyone can upgrade their browsers... 
you can read a lot of good reasons why in the thread that came 
about on this list after the ALA switch...




More information about the thelist mailing list