W3C home > Mailing lists > Public > public-html@w3.org > April 2007

Re: Pandering to poor authorship (was Proposing <indent> vs. <blockquote>)

From: Mike Schinkel <w3c-lists@mikeschinkel.com>
Date: Sat, 28 Apr 2007 03:23:32 -0400
Message-ID: <4632F674.5020407@mikeschinkel.com>
To: public-html@w3.org

Sander Tekelenburg wrote:
 > My previous sentence was too unclear. Sorry. What I meant to convey 
was that
 > for your use case <indent> offers no more than <div
 > style="padding-left:2em"> does.

I agree it offers no more visually, but it offers a lot more in terms of 
1.) less learning required (HTML only and not CSS), 2.) less typing, and 
3.) less liklihood of mistyping.

 > Whoever is in control of the site is in
 > control. If they don't want to allow style attributes, they'll just strip
 > them.

That's idealistic.  Many people run software build by other people that 
makes the decisions for them (i.e. WordPress, vBulletin, etc.)

 > Adding <indent> to the spec will just make the same admin replace
 > <indent> with <div>.

Huh?


 >>> The situation in which people can only insert snippets and not 
affect their
 >>> presentation can exist for very good reasons: to ensure that they don't
 >>> create a mess.
 >> Again, specious. The case would be far more often than people would need
 >> to be given reasonable control of how their snippets would be
 >> formatted.
 > I have no numbers (do you?) but I can assure you that many site 
admins need
 > to ensure that content editors cannot affect presentation. This is no
 > different from companies/organisations requiring their content 
providers to
 > use consistent styling in printed material that goes out.

No disagreement, but your argument is tangential. It is your attempt to 
tie the need for website publishers to control submitted content to the 
validity of an <indent> element, and they are not related.  IOW, if 
<indent> does or does not exist it doesn't change website publishers 
need to control submitted content.  Can we please stay on topic?

 >> What site owner wants user contributions to be poorly
 >> visually formatted ?
 >
 > Exactly. Usually the only way to achieve that is to remove as many
 > possibilities from content providers as possible.

Which is draconian and is used simply because it is easier not because 
it is better. But that said, let's get back on topic.

 > If you don't, your content
 > editors will go nuts -- every paragraph will be in a different font with
 > different margins, etc. Or at the very least, your content editors 
will each
 > be suggesting different presentations, making the site look a mess to
 > visitors.

That's why it is so important to have simple and generic elements like 
<indent> instead of encouraging people to put explicit styling in. You 
are making my point.

 > And when the site admin *does* want to allow users to 'go nuts', he can
 > simply allow style attributes.

Currently either he has to allow style elements or he cannot give users 
a complete set of reasonable control w/o essentially endorsing misused 
semantics (i.e. <blockquote>)

 >> That said, how can <indent> create a mess where <div
 >> style="padding-left:2em"> cannot?
 >
 > I meant to make clear that for your use case there is no difference. 
<indent>
 > or <div style=blah> will be under the control of whoever is in control.

But <indent> is much easier to parse and thus will be more likely 
controlled effectively.

 >>> And the other way around: when an environment is too limiting,
 >>> the author (hopefully) has the freedom to use another environment 
instead.
 >>>
 >> Huh?
 >
 > If system or service x doesn't allow you to publish your content the 
way you
 > want, you switch to a system or service that does.

Funny, you continue to make my point for me...

 >>> [*] In fact, even <div style="padding-left:2em"> might simply be 
stripped
 >>> down to <p> by the authoring environment. When you're not in 
control you're
 >>> not in control.
 >>>
 >> And in that case it would be much easier o allow <indent> than to have
 >> to parse <div style="padding-left:2em">.
 >
 > What makes you think <indent> wouldn't be stripped, or replaced?

I don't; you need to parse something to know what to keep as much as to 
know what to strip.

 >> What I'm hearing is that you have a "religious" objection to <indent>
 >
 > I'm anti-religious.

Maybe with questions about god, but not it seems with questions about 
markup. :-)

 >> [...] So then should we do
 >> away with default formatted on <p>, <ol>, <ul>, etc. too?
 >
 > I've been arguing against some people's wishes to aim for sites to 
look the
 > same in every browsing environment by speccing default presentations, 
yes.

That is apples and oranges. I'm not asking for them to look the same in 
every browsing environment, I'm asking that there be an element that for 
a given browsing environment the element can have a default presentation 
and not any associated semantics.


Dão Gottwald wrote:
 > Mike Schinkel schrieb:
 >> Dão Gottwald wrote:
 >>> Mike Schinkel schrieb:
 >>>>> The situation in which people can only insert snippets and not 
affect their
 >>>>> presentation can exist for very good reasons: to ensure that they 
don't
 >>>>> create a mess.
 >>>> Again, specious. The case would be far more often than people 
would need to be given reasonable control of how their snippets would be 
formatted.  What site owner wants user contributions to be poorly 
visually formatted ?
 >>> It is the page author who rules the layout. If she wants code 
blocks to be indented, adding |pre { margin-left: 2em; }| to the 
stylesheet is the trivial solution.
 >> Once again you are mistakenly believing that all HTML authors author 
entire HTML pages.  That is far from reality, and increasingly less so 
each passing day.
 > Replace "page author" by "site owner". It doesn't matter. The point 
is, you usually want pages to look frankensteinesque, no matter how many 
people participated.
 >

As I just said to Sander, that issue is not relevant to this debate.

 >>>> That said, how can <indent> create a mess where <div 
style="padding-left:2em"> cannot?
 >>> Who says that <div style="padding-left:2em"> cannot? Using <div> 
instead of semantically descriptive elements should be considered 
equally bad.
 >> But you (or someone else) implied it was a better solution than 
<indent>.  I was pointing out for this reason it was not.
 >
 > Well, what makes it better is that <div> and the style attribute are 
already sped'ed.

See the three benefits I stated at the beginning of this email in reply 
to Sander.

 >>> Btw, that's also why the style attribute was deprecated in XHTML 1.1.
 >> I thought the reason we are working on HTML5 is because XHTML was a 
determined to not be such a good idea after all?
 >
 > That was mainly related to XHTML2. XHTML1 isn't necessarily a bad 
thing; and I'm not sure if modularization (XHTML1.1) is considered harmful.

I disagree, I think XHTML 1.0 is a bad thing and I think that's why TBL 
recognized recently. My discussions with Ian et. al. on the WHATWG where 
what brought me to this conclusion.

 >>> HTML4 says "For optimal flexibility, authors should define styles 
in external style sheets."
 >> And that was written in the days before there were many situations 
where people contribute HTML to be stored in a database and then 
composed with HTML other people wrote, as in blogs, wikis, forums, etc.  
Many HTML author simply do not have access to adding an external style 
sheet and we really should acknowledge and accommodate that reality.
 > That's by design. Site owners could give others access to style 
sheets, but the reality is that they want people to contribute content, 
not layout.

That is painful idealistic. You are assuming site owners are fully 
congizant of all these issues and that they can cause implementation 
changes merely by thinking about them. I'll content that, with the 
explosion of use of open-source content management systems (WordPress, 
Drupal, etc. etc.) that most sites on the web are developed by someone 
other than the site owner; if not, they soon will be.  And I think it is 
critical that the HTMLWG acknowledge this reality.

-- 
-Mike Schinkel
http://www.mikeschinkel.com/blogs/
http://www.welldesignedurls.org
http://atlanta-web.org - http://t.oolicio.us
Received on Saturday, 28 April 2007 07:23:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:15:53 GMT