Browser standards

At 4:13 AM -0700 3/23/08, Denis or Janet Jarvis wrote:
>Jincarr gave a reference to the browser standards issue on Mar 22, 2008:
>>The following link nicely explains IE8, W3C standards and all that.
>>The headphone discussion is just an intro to the main topic.
>>
>>http://www.joelonsoftware.com/items/2008/03/17.html

>I have no direct insight into the browser market but am just a
>cynical consumer.
(moved for context)

Which, IMO, gives you the best insight of all. Too often developers
lose site of basic usability, human factors, and all that rot. They
created this mess and now they're playing games, deciding how to
maybe sortof fix it, with the least pain for *them*.

>This article may be a good summary of the present dilemma but IMHO
>it does not give enough emphasis to the commercial strategies that
>caused the problem.

Power and Profit. heh.

This past year, I've seen a big change in my clients. Their new
attitude is that people can just use Firefox or Safari, on Mac or PC.
They have NO interest in investing developer time to make their sites
work with any version of IE anymore.

IMO, this is the right attitude:

We need to simply stop paying blackmail - be it cash or our sweat -
to Microsoft.

And in the business & edu worlds, we need to stop accepting MS'
bribes -- those almost free desktops, laptops, servers, and low-cost
copies of Windoze, that suddenly become available when we start to
switch to all-Mac/Unix/Linux.

>The above article says a solution is difficult. Naively, I would
>expect that if all parties honestly worked to form and maintain a
>common interface standard then this would be possible

Except that it seems that MS has no interest in being part of the
standards development. The references recently posted in the "Safari
3.1 & Firefox 3 trumped" thread, about their poor participation in
the HTML 5 effort, are telling. SSDD.

>For me this is not just a hypothetical problem. I am currently using
>iWeb to create a web site for our family and there are two main
>problems. Firstly, the Navigation Bar produced by iWeb is not
>displayed by IE7, even though all other modern browsers do.

Have you reported this to Apple? I don't mean posted a thing on
their discussion boards. I mean by filing a real trackable bug
report... I've found they're very good about following up with
problems like this.

<http://bugreport.apple.com/>

>The second problem occurs when I need to upload the web site for
>hosting on Comcast. The Comcast FTP
>interface will only work with IE7. No other browser and not even IE5
>on a Mac will work.

Comcast is running a standard ftp service. You don't need to use
their cruddy web-based interface; just use a normal FTP client, such
as Cyberduck.

ObQuip: Comcast's web services are soooo unreliable. Might be a
good plan to find a better host.

FWIW,
- Dan.
--
- Psychoceramic Emeritus; South Jersey, USA, Earth

Thanks for the link to Joel Spolsky's article on web standards. It's written in an amusing style, on a subject I fret about. I like to blame Microsoft for making web design so difficult that I don't enjoy it and don't want to do it.

I'm in the idealist camp, that wants to enforce standards. Joel Spolsky criticizes both sides. He explains the problem and its history well. Makes some good points. (The standards are unreadable, and there are no reference implementations.) Even makes me reconsider my position.

I think the fault in his logic is that he doesn't follow his predictions far enough. If we don't take a hardline stance of enforcing standards, then the mess will continue forever. If we do, there will be lots of incompatibilities, broken websites, and frustration -- but in a few months, they will be fixed, and the benefits of strict standards will be realized. Websites will work better, and it will cost less to develop them.

That lesson will then be applied elsewhere in the industry, and software development will be more productive and more fun. And maybe more end users will appreciate what standards are for, and not support companies that don't cooperate in setting open standards.

--Davi

> I think the fault in his logic is that he doesn't follow his
> predictions far enough. If we don't take a hardline stance of
> enforcing standards, then the mess will continue forever. If we do,
> there will be lots of incompatibilities, broken websites, and
> frustration -- but in a few months, they will be fixed, and the
> benefits of strict standards will be realized. Websites will work
> better, and it will cost less to develop them.

Another part of the picture, Davi, is that 'The West' is in a
minority when it comes to logic. About a billion of us and five
billion of them. What is especially difficult for enforcing logic is
that it is the 5 billion who are hosting the most development of
business and industry, right now, while we are dying off.

If there is a future it is in combining western and eastern logic:
and that is what many leaders of the five billion are doing. Taking
the hard edges off it.

George

At 05:59 -0700 UTC, on 2008-04-20, post wrote:

[...]

> I'm in the idealist camp, that wants to enforce standards. Joel Spolsky
>criticizes both sides. He explains the problem and its history well. Makes
>some good points. (The standards are unreadable, and there are no reference
>implementations.) Even makes me reconsider my position.

One big problem with the HTML standards so far as that they don't say what a
user agent is to do with a non-conforming document. The result is behavioral
differences between browsers when confronted with non-conforming documents.
(Some 95% of the Web is non-conformin, btw.)

XHTML being XML, and thus not allowing non-conformance, was to be a solution.
But because Internet Explorer doesn't support XHTML, 99% of authors don't use
XHTML[*].

HTML 5 takes another approach: it doesn't require browsers to ignore
non-conforming documents, but instead describes, in detail, how browsers are
to handle non-conformity. Assuming browsers will become HTML 5-compliant,
this should result at least in non-conforming documents being interpreted the
same across HTML 5 browsers. Thus a document checked in one (HTML 5) browser
only will be interpreted the same in other HTML 5 browsers.

Still, as long as Microsoft maintains its monopoly, and doesn't conform to
HTML 5, none of this will mean very much I'm afraid. The only way there will
ever be a real change is when users start demanding better quality browsers.
Be it by upgrading to non-IE, or by breaking Microsoft's monopoly through
legal means.

[*] Well, authors do use XHTML, but they serve it as HTML because IE still
doesn't accept XHTML. So for all practical purposes those authors give up on
exactly that benefit of XHTML.

> I think the fault in his logic is that he doesn't follow his predictions
>far enough. If we don't take a hardline stance of enforcing standards, then
>the mess will continue forever. If we do, there will be lots of
>incompatibilities, broken websites, and frustration -- but in a few months,
>they will be fixed, and the benefits of strict standards will be realized.
>Websites will work better, and it will cost less to develop them.

Agreed on the theory. However, in practice there is no way to force browser
vendors to have their browser refuse to display non-conforming documents.


--
Sander Tekelenburg, <http://www.euronet.nl/~tekelenb/>