Re: Development with IE 7 for non-Windows users.


Tim Judd wrote:
Daz,

I've read your posts, and Rob's replies. I see both sides of the story.

I'm glad you do. I also accept that there's not just one side. :)

However, since Microsoft has obviously ignored or superceded the
standards that most other browsers accept, I rely only on the standards.

That would make sense, but IE still renders things differently.
Although I can't vouch for IE7. Aren't JavaScript and CSS/CSS2 in the
standard? I am pretty sure that CSS is, but IE doesn't even support the
:hover event. It's insane.

If you code the site to be W3C compliant with their validator, no amount
of code to get it to appear the same is worth the time. Validate the
code, and if MSIE renders it differently, it is not the web developer's
fault. It is only the rendering engine itself that is to blame. And in
this topic, the rendering engine is made by microsoft, so I blame
microsoft for the inability to render W3C standards properly.

Plausable deniability. I like that. It would be hard to live with, but
I think it's 'doable'. But shouldn't all pages display correctly from
within the most 'used' browser in the world?

I do web design as a hobbyist only. I suck at the design because I want
it to be lined up, but then I remember about the W3C standards and
abandon the website design.

The newer thing out there is to turn all scripting off -- well, it's
possible to do a complete site w/out client side scripting, but it can
make dynamic pages a pain... Where you're back to square one when the
web was "created" -- static HTML pages that never change. Don't confuse
this with dynamic pages that change on server-side scripting, which is
all technology.

I design all of my pages to work without script (where possible, of
course). Sometimes script can be very useful. If I wanted to make some
kind of validator for example. I see script as being a method for
allowing users to use programs that you've written. Be it to validate
data, or format data or whatever else the need might before. I also see
it as an excellent opportunity to save some bandwidth on my website.
Rather than having to render an entire page each time, the page is only
rendered once, and their computer does the rest of the work once we
send it a little extra data upon request. Obviously, this is using
AJAX, which I think is very useful. Obviously, if a users browser
doesn't support script, then they can do it the slightly longer way,
but I think that in general script is a marvelous thing that should be
taken advantage of. At the same time, we shouldn't be shutting out any
of the users who have browsers that don't support JavaScript, even
though about 99.9% of browsers these days will support it. As Rob said,
we should be catering for the majority, not the minority. I think we
should be catering for just about everyone. No one likes having
web-based doors slammed in their face:

"Sorry, to use this page you must enable JavaScript".
"Sorry, to use this page your browser must support frames".

I disagree with all of that kind of thing. I like to have a backup
plan, and try to divert the users browser using <noscript> tags between
my <head> tags, to an HTML version of the page, if javascript is not
available.


In short -- the long answer is: yes, there is always inconsistencies.
And it's the software developer's fault.
short answer: Code to W3C standards, leave the rendering alone.

I would like to know what your final decision is, once you come up with
it. :)

To be honest, I'm not sure if I will ever reach a 'final decision' as
such. It's human nature to want what we can't have, and as a developer
I will always have ideals which I will grumble about. It due time,
these annoyances will no doubt be resolved, and new annoyances will
replace them. I guess we can only do the best we can to please
everyone. (Which is more than I can say about Microsoft. Hehe!)

Thanks for your comments Tim. I really appreciate them. I understand
that all the decent developers just get on with it and make the most of
the tools they have to hand. I understand that I am in the minority as
an idealist. However, I am a firm believer that if anything is ever
going to change, now is the time to work toward it. I know I am not
very influencial, but there's no harm in trying. At the absolute worst,
nothing will change, but you never know unless you try.

I have written a nice email to Microsoft, expressing my concerns with
regards to this matter. I am not going to hound them, nor write abusive
or threatening emails to them as it's not my kind of thing, but I would
be interested to know what they suggest. It wouldn't surprise me if
they don't reply at all as I am not a 'paying customer'. :)

All the best.

Daz

.

Relevant Pages

  • Re: document.write and problems with how it overwrites existing code
    ... >>> Here's the problem with the property 'innerHTML'. ... the script writer risks (at a higher probability than ... > versions of a browser which strictly adhere to standards. ...
    (comp.lang.javascript)
  • Re: Why why why? (Was: Re: Vertical alignment of text within a DIV)
    ... It's also expected behaviour for standards bodies. ... The rendering *is not* necessarily part of the base web experience, ... parser is the trivial bit of writing a web browser. ... environment (display code for MS Windows probably won't work in X ...
    (comp.infosystems.www.authoring.stylesheets)
  • Re: missing ) after argument list
    ... >> and there is proof that at least one script engine, the JavaScript ... Give one browser, just one that produces a syntax error ... Are you arguing that standards are irrelevant and that we should just ...
    (comp.lang.javascript)
  • Re: PicForth 1.0 is released
    ... and the design has a nice clean CSS ... The most common HTML rendering tool doesn't follow the ... browser magically has become much more comfortable;-). ... terms of W3C standards). ...
    (comp.arch.embedded)
  • Re: fixing Google AdSense
    ... I could understand that page display is ... >wait with further rendering until it has the script and has it executed ... >browser continues without having the script. ... It does not "finish" rendering the page and give control back to you. ...
    (comp.lang.javascript)