[thelist] Site redirect check : old browser

[.jeff]

mark,

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: From: Mark Cheng
:
: > so how is the style not separated from the
: > content on evolt.org?
:
: I didn't say it wasn't.  What I said was that
: I was looking for the benefits of separating
: style from content.  As I have no backend skills
: whatsoever I didn't need the complication of a
: backend template and sourcing the content from
: a DB.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

nobody said you needed the complication of sourcing the content from the db.
you could just as easily use server side includes and have the content of
each page stored in text files.  you should be able to learn the basics of
how to do that in a matter of an hour or so.

here's a couple of articles to get you started:

http://www.evolt.org/article/thelist/17/1344/
http://httpd.apache.org/docs/howto/ssi.html

the concept you're missing when we use the term "separation of content and
style" is that the content is not contained in the same file as the rest of
the html for the design of the site.  you have a header file which contains
all the html from the top of your layout down to the point where the content
begins.  you also have a footer file that contains all the html from the
point where the content ends to the end of your layout.

what this gives you is exactly two templates that you have to change to
alter the design of the entire site instead of having to alter the html in
*every* page of the site.  that's separation of content and style -- or
separation of markup and content as aardvark called it.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: However, I achieved much the same effect with CSSP.
: Why not a table?  Because with CSSP I can move the
: design of all the pages just by changing the
: stylesheet. I don't need to change my HTML at all.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

you can *move* the items around, but you can't change the items that easily.
let's say you wanted to change one of the images used in the design -- you'd
have to go into the html on every page.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: Actually, you can slap a new "template" on
: it - just insert a color change (or a
: positioning change, or a font change) into
: a user stylesheet for #content and mark it
: !important.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

that's not truly a new template though.  that's just a variation on the
current design -- things in different places, different colors, and
different fonts.  that doesn't give you the ability to add/remove design
elements to achieve an entirely different look and feel.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: One of the first things was to select the
: browsers to support - That was simple -
: anything with getElementById.  Why? -
: because it made the javascript easier.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

no, it doesn't make the javascript that much simpler.  you still have to
perform object/method support checks before using those objects or methods.
for example, if you're going to use getElementById, then you'd better first
check that the browser supports.

if(document.getElementById)
{
  el = document.getElementById('foo');
}

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: But - the flexibility offered by CSS-P to
: change it is a key thing that we wanted to
: keep.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

but it doesn't give you the flexibility you think it does.  you can't make
changes to it.  you can only make variations of it.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: I don't consider using the full potential
: of the standards available a "failure"
: because I *can't* get the same layout in
: a browser that predates the writing of the
: standards.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

then don't *make* it a failure by preventing pre-standards browsers from
viewing the content from the beginning.  ditch the redirect and just let
them through.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: [...] it costs nothing to upgrade a browser.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

that may be true for some, but not for everyone.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: Basically, that is a user choice - I'm not
: shrouding my "failure" - I'm pointing out to
: users who don't know/realise a simple fact.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

another ridiculous assumption -- it's not always up to the user.  for some
that have the choice, they may not even care.  when it doesn't work they're
going to blame it on your site, not their browser.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: The only thing I have taken away from the user
: is the opportunity to hit this site with no
: warning about what may happen.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

what's the reason for the warning?  what is so complex about the site that
it's likely to explode in some ugly fashion for non-standards compliant
browsers (which, if this happens btw, you've dropped the ball by not doing
your best to prevent that)?

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: Actually this discussion has made me realise
: that the issue with this site is not the
: positioning - its the javascript.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

it's two things really.  it's the javascript, but more importantly it's the
priorities in place when you started building the site.  the most important
thing when building this site was to make it accessible -- regardless of the
user's browser.  then, after the entire thing worked, you go back in and
start adding the styling to the elements for newer browsers and then the
dhtml.  the entire time though you've gotta make sure that some piece of
additional functionality does not remove any of the original accessibility
you put in place.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: So - as long as I put in appropriate capability
: checking - and a warning that some things won't
: work - I don't need a redirect.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

mostly correct.  if you do capability checking then you don't need the
redirect -- skip the warning entirely so long as you've made sure that it's
still usable.

:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
: I think that I'll still keep it, but a wording
: change may be appropriate.
:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

i still think you're missing the fact that the redirect is not
user-friendly.

good luck,

.jeff

http://evolt.org/
jeff at members.evolt.org
http://members.evolt.org/jeff/