design question - what makes a page of options clear for the user?

L

lawrence

Isofarro said:
Collections of images linking to pages - still a list of links


Collections of text linking to pages - still a list of links.


You are using a political quote as a form of navigation? Nuts.


I hope you realize how agressive your tone is. I really have to wonder
if you'd talk like this if we were face to face having a chat at some
restaurant. I'm always a bit saddened by the agressive tone so many
people take when they talk online.

The side bar is there for people to put stuff into it. What they put
into it I cannot know a head of time. How they wish to order their
websites I can not know ahead of time. What they want to do with their
web sites I can not know ahead of time. I have to program for maximum
flexibility.

A link, followed by a political quote, followed by a link, followed by
an image of themselves, followed by a PHP script that gives the date,
followed by another link, seems to me like a heterogenous collection.
I'm not sure that such a collection should be each preceded by <li> .
 
L

lawrence

Toby A Inkster said:
No -- things happen to work because pages are close enough to the
standards for browsers to guess what the authors meant.

If it weren't for the standards, then this wouldn't be the case.

I took nearly all of your posts and sent them to the two graphic
designers I work with. There reaction was basically "What is all
this?" They couldn't make sense of it. They had no idea what a
structured document was, nor why it might be important. They are
coming from a graphic design background, with a strong emphasis on the
word "graphic". They know how to make pages that look pretty. It was
interesting to watch their reaction. They felt like they were reading
a foreign language. They were wary. Using tags to offer logical
structure to a document sounded to them like a bit of programming, and
they both reminded me that they weren't programmers.

We've planned a meeting for Monday during which we're going to go over
this thread and I'll explain to them what a structured document is.
I'm not doing it to convert them, but I want them at least have a
sense about this point of view, which you do a fairly good job of
detailing.

Still, when they ask me why a structured document is important, I
admit I stutter. To my mind, the only reason that "structure" is
important is so that the document can be processed by software. But
HTML doesn't give me enough information to do anything useful. I can
not think of any interesting code I might write to process something
like <ul>. If I don't know what the content is, how might I process
it? So I'm not inclined to push it. Processing a structured document,
to my mind, is something one does with XML, because there the info is
so much richer. We've built Flash sites where the Flash reads info out
of an XML file to know what to show on screen. Also, I've a friend
who's in charge of the effort at Lexis/Nexus to convert their
datastore to an XML format. We've had some fun talks. Of course,
Lexis/Nexus doesn't dream of using HTML tags as their database,
because HTML tags aren't descriptive enough.

Still, I've come to see the value of using header tags and Blockquote
and Cite. Google looks at header tags, and I can write cool code to do
something with blockquote and cite.

Thanks for all the feedback.
 
L

lawrence

Toby A Inkster said:
Quite the opposite actually. How could anyone interpret the semantics of
the following well-formed XML snippet?

<flibble>
<foo f="John" x="Steve" />
<foo d="Chas" f="Dave" />
<bar zibble="zonk">
Potato<baz />Celery<baz />&russian;
</bar>
</flibble>

There is simply no way to make any sense of it, unless you know what the
elements flibble, foo, bar and baz mean, and what their attributes mean.

If it was my XML, then I'd know what it would mean, and I could write
the software to process it. If its not my XML, but it is a well known
XML standard, like some of Google's web services, or weblogs.com
services, then I can write code to process it. But I can't think of
any interesting code to write for HTML. The only software I can think
of writing for it is rendering software, that is, a webbrowser, which
I won't do because there are already plenty of web browsers out there.
But if I were writing such software, it seems like making the thing
look right would be my principal concern.
 
T

Toby A Inkster

lawrence said:
Toby A Inkster said:
lawrence said:
XML has semantics, HTML doesn't.

Quite the opposite actually. How could anyone interpret the semantics of
the following well-formed XML snippet? [snip]
There is simply no way to make any sense of it, unless you know what the
elements flibble, foo, bar and baz mean, and what their attributes mean.

If it was my XML, then I'd know what it would mean, and I could write
the software to process it. [snip]
But I can't think of any interesting code to write for HTML. The only
software I can think of writing for it is rendering software, that is, a
webbrowser

Which only goes to show that you have a very limited imagination.

How about a spreadsheet that uses HTML tables?

Or a document display system that can generate automatic summaries of
documents, based on their <hX> headings?

A web crawler that searches documents for <dt>/<dd> pairs to populate a
dictionary?

A C++ compiler with a built in browser, so you can surf the web and then
press a button to compile the C++ code between <code> and </code>?

Presentation software (like Powerpoint)? (This has already been done by
Opera).

A word processor that can compile a bibliography at the end of the
document automatically, using links and cite attributes?

A crawler that uses the rel attribute of HTML links to build a site map
for an entire web site?

There are tonnes of things that can be done with HTML, but only if you use
the correct elements for everything.

This is because HTML has semantics. XML has no inherant semantics.
 
I

Isofarro

lawrence said:
The side bar is there for people to put stuff into it.

Right, now its a side bar, and not a navigational element. Serious
difference.
What they put
into it I cannot know a head of time.

That's what requirements gathering and analysis are for. You don't create a
website until you have requirements.
I have to program for maximum flexibility.

Websites are not programmed. They are authored.
A link, followed by a political quote, followed by a link, followed by
an image of themselves, followed by a PHP script that gives the date,
followed by another link, seems to me like a heterogenous collection.

Seems like a random clutter of ill-thoughout components. What user benefits
is this haphazard scattering supposed to achieve? Some respectable
references would be nice at this point.
 
I

Isofarro

lawrence said:
I took nearly all of your posts and sent them to the two graphic
designers I work with. There reaction was basically "What is all
this?"

So they don't really know HTML. Sounds quite familiar.
They couldn't make sense of it. They had no idea what a
structured document was, nor why it might be important.

Quite a common failing with web designers. Nothing to be completely
embarrassed about.

They are
coming from a graphic design background, with a strong emphasis on the
word "graphic". They know how to make pages that look pretty.

Glamour websites died along with boo.com.

It was
interesting to watch their reaction. They felt like they were reading
a foreign language. They were wary.

That tends to happen when they are faced with the distinct possibility that
what they believe is wrong. Now they need to get over that bump in the
road, and work with the strengths of the medium (like every true artist).
Using tags to offer logical
structure to a document sounded to them like a bit of programming,

No, its called markup. Newspapers, books and magazines have been doing it
for centuries.
We've planned a meeting for Monday during which we're going to go over
this thread and I'll explain to them what a structured document is.

You'll probably have to explain the following concepts first: Hypertext,
then Markup, only then attempt structure. Pointing out that the web isn't
solely a visual experience will also help at this point. (and yes, blind
people do use the web).

Still, when they ask me why a structured document is important, I
admit I stutter.

Why is structure important in any content medium? Why are books divided into
logical structures like chapters and parts. Why are page numbers found in a
consistent location within a particular book? Why are magazines broken into
articles and tabulated into a tables of content? Why do newspapers have a
strongly hierarchical series of headlines. Why are the top stories on the
outer pages of a newspaper. These are all logical structures - so such a
concept should not be alien to anyone with basic literacy skills.
To my mind, the only reason that "structure" is
important is so that the document can be processed by software.

One tiny piece. Structure brings consistency and familiarity. When you look
at the front page of a newspaper - how do you tell which story is regarded
by that edition as the most important story? You tell by reading the main
headline (a logical structure). Now you can use that logical structure and
pick up almost any decent newspaper and figure out which is their most
important story (including a foreign language newspaper).
I can
not think of any interesting code I might write to process something
like <ul>.

Consider reading through the thread properly before regurgitating. It is
common courtesy.
 
L

lawrence

Toby A Inkster said:
Which only goes to show that you have a very limited imagination.

How about a spreadsheet that uses HTML tables?

Or a document display system that can generate automatic summaries of
documents, based on their <hX> headings?

A web crawler that searches documents for <dt>/<dd> pairs to populate a
dictionary?

A C++ compiler with a built in browser, so you can surf the web and then
press a button to compile the C++ code between <code> and </code>?

Presentation software (like Powerpoint)? (This has already been done by
Opera).

A word processor that can compile a bibliography at the end of the
document automatically, using links and cite attributes?

This is quite an interesting list of ideas for interacting with HTML.
I thank you for the list.
 
L

lawrence

This topic is apparently off-topic on alt.html, so I'll take my
question elsewhere. Can anyone point to a design site where people
discuss non-technical issues of design and visual presentation? I
already read Edward Tufte's site and Christina Wodtke's site, but
something specific to the web would be helpful.
 
W

William Tasso

lawrence said:
This topic is apparently off-topic on alt.html, so I'll take my
question elsewhere. Can anyone point to a design site where people
discuss non-technical issues of design and visual presentation? I
already read Edward Tufte's site and Christina Wodtke's site, but
something specific to the web would be helpful.
...

The o/p is not available from my news service but from the subject header
and your notes,
news:comp.infosystems.www.authoring.site-design
seems like a sensible choice.
 
L

lawrence

I've a follow up question about the structure of HTML. It is it better
to have a lot of empty tags, like this:

<h1></h1>
<h2></h2>
<h2></h2>

Or to have non-empty div tags like this:

<div class="headline1">Welcome to this site!</div>
<div class="headline2">Top stories:</div>


It occurs to me that the PHP code might just print the content, like:

$dbArray = $sql->getWeblogContent();
extract($dbArray);

<h1> $headline </h1>

<h2> $subheadline1 </h2>

<div> $maincontent </div>


In this way, the graphic designers who design the sites could have
total freedom about what HTML tags they wrap that code in. They'd
write all the tags themselves. However, sometimes there would be no
content and there would be a lot of empty tags. How bad is that? Is
Google smart enough to simply ignore empty tags?


The other possibe strategy is to have the code print nothing if there
is nothing to print, like this:


$dbArray = $sql->getWeblogContent();
extract($dbArray);

if ($headline) echo "<h1> $headline </h1>";
if ($subheadline1) echo "<h2> $subheadline1 </h2>";
if ($maincontent) echo "<div> $maincontent </div>";

The advantage of this is there are no empty tags, the disadvantage is
that the HTML is hard-coded, and the graphic designer can't change any
of it. And, as this thread makes clear, some designers will take issue
with the choice of tags. More so, true freedom of design should mean
changing those tags. On my own personal site, the weblog title is 12px
and the each weblog entry title is big. I'd use a <p> tag for the
weblog title and an <h2> for the weblog entry title. But on most
sites, the weblog title is the biggest thing on the page, and the
entry titles are more like 12px.

The question, really, is how damaging is it to the sense of
"structure", when speaking of a structured document, to have a lot of
empty tags? Toby's convinced me that the empty tag approach is best
because of the need to maximize the freedom of the designers. More so,
if the code writes no tags, then the separation between code and
design is cleaner. But, again, I wonder if it is considered poor form
to have a lot of empty tags on the page?
 
W

William Tasso

lawrence said:
...
I wonder if it is considered poor form
to have a lot of empty tags on the page?

it is certainly a waste of bandwidth and it doesn't say much about the
structure of the document
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
474,085
Messages
2,570,597
Members
47,220
Latest member
AugustinaJ

Latest Threads

Top