Fixed site width vs. dynamic: what is good web design?

[Jonathan N. Little]

Well we disagree I guess. Won't be the first time.

Well you haven't been very good at supporting your position with any
supportive argument. The ol' "'cuz" does not support the debate. In my
argument I used to support my position that how with web-delivered
content the "canvas", device, and actual user access of the content
cannot be controlled by the designer; and therefore impacts the
effectiveness of various layouts. And you counter "I just don't believe
that it is the only way to design a website." Essentially, "'cuz".
That's the problem I think here in the US; emphasis is on "faith". It
seems to have destroyed critical thinking. You can hold a position
without the need for all that analysis and synthesis of evidence and
factors to come to a reasoned conclusion. You can just wave your hand
and say, this is it because it is how I believe. The end.

Now I didn't say liquid-layout is the *only* way to design a website.
What I have said is that liquid-layout is best suited for how the web
works. The fixed-layout will always be at the disadvantage because of
the flexible delivery of web-content. Therefore if one decides on a
fixed liquid-layout they have to realize the limiting and negative
impact of that design choice and whether on not the "control" provided
by a fixed layout out weighs the detriments with respect to the
application. And my point is that too many designers arbitrarily choose
fixed-layouts without taking a critical look at how the web works.

If you feel my argument is incorrect, challenge it, and convince me with
evidence. I have not always held this position. Coming from a art and
graphics background my first web designs were very "fixed". As the web
has evolved so has my design process, and it still continues to evolve.

 

[Neredbojias]

Well bust mah britches and call me cheeky, on Mon, 11 Feb 2008 20:27:27 GMT
dorayme scribed:

Now, what was I saying? O yes, modesty...

No bugger that, I am reminded of my lost cause:

<http://members.optushome.com.au/droovies/opinion/drugLaws.html>

Drugs are often considered the opiate of the masses. However, I think
their real benefit is in providing an excuse for those who just don't get
life, the universe, and everything.

 

[Toby A Inkster]

That's the problem I think here in the US; emphasis is on "faith". It
seems to have destroyed critical thinking.

You should check out Ben Elton's latest book "Blind Faith". It's his
updated take upon "Nineteen Eighty-Four", set in overcrowded post-flood
(i.e. global sea level rises) London, some time around the year 2200. The
surveillance state is enforced not by Big Brother, but by an inverse-
voyeuristic society who willingly broadcast every aspect of their lives to
the world; encouraged by religious leaders, preaching a bizarre mixture of
Christianity, Princess Diana worship and self-love. Ones beliefs are
protected by law, while the truth about many aspects of the world are
pushed underground.

Elton can be a variable writer, but this one's a good read.

--
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]
[OS: Linux 2.6.17.14-mm-desktop-9mdvsmp, up 15 days, 22:29.]

Mince & Dumplings
http://tobyinkster.co.uk/blog/2008/02/10/mince-dumplings/

 

[Travis Newbury]

Well you haven't been very good at supporting your position with any
supportive argument.

Fixed width is important when the presentation of content is as
important as the content itself. Entertainment (movies, sports,
celebrity, etc..) sites, interactive sites catering to children, and
sites devoted to gaming are some that fall into this category.

These types of sites are generally not visited by phones or PDA, and
they tend to be very graphical in nature. People that visit these
types of sites more than likely allow Flash and Javascript, and are
looking to be entertained by the site.

Now, if you disagree that there are sites where the presentation is as
important as the content, then you will not find my argument valid.
So we still disagree.

 

[Jonathan N. Little]

Fixed width is important when the presentation of content is as
important as the content itself. Entertainment (movies, sports,
celebrity, etc..) sites, interactive sites catering to children, and
sites devoted to gaming are some that fall into this category.

These types of sites are generally not visited by phones or PDA, and
they tend to be very graphical in nature. People that visit these
types of sites more than likely allow Flash and Javascript, and are
looking to be entertained by the site.

Now, if you disagree that there are sites where the presentation is as
important as the content, then you will not find my argument valid.
So we still disagree.

Agree. But a as I say there are too many sites that sell computer parts,
office supplies, and other stuff that "felt the compelling need"...

I also remember years ago, when my children were young and broadband was
a real rarity of a game site, www.origin.com. Remember old Wing
Commander? They had a "wonderful" flash site that looked impressive once
it downloaded. Pain when all you wanted was the damn game patch... There
where countless of that ilk.

 

[Travis Newbury]

Agree. But a as I say there are too many sites that sell computer parts,
office supplies, and other stuff that "felt the compelling need"...

So we are almost in complete agreement then. If the site's purpose is
direct e-commerce (the computer parts store, etc...), or information
distribution (say google, wikpedia, ect..) then yes, Flexible width,
no javascript or flash for functionality is probably the way to go.

That is why I say, you can not discount one method or another before
you understand the site. I even go as far as to say discounting any
technology before one understands the reason for the sites existence,
and the likely visitors to the site is kind of foolish and is a
disservice to the client.

I also remember years ago, when my children were young and broadband was
a real rarity of a game site,www.origin.com. Remember old Wing
Commander?

I remember Ginger Lynn the porn star had a part in the cut scenes.

They had a "wonderful" flash site that looked impressive once
it downloaded. Pain when all you wanted was the damn game patch... There
where countless of that ilk.

Flash then and Flash now are two totally different animals. BUT (and
this is a HUGE BUTT) the vast majority of the Flash developers out
there haven't a clue how to make good Flash. Most come from a
graphics background. Those of us that come from a software
development background and use Actionscript for just about everything
including animations create fast small Flash files that dynamically
load things as they are needed. So people not on broadband can use it
as easily as those with broadband.

I showed an example of that here several months ago with a Flash and
ASP/SQL Server based LMS (learning management system) I built for
Siemens that was under 50k, but managed classes, students, locations,
and course content. An artist could not have done the same.

 

[Toby A Inkster]

Fixed width is important when the presentation of content is as
important as the content itself.

That is patent nonsense. Just looking at a web page it's impossible to
even *tell* whether the design is fixed width or not.

e.g. http://examples.tobyinkster.co.uk/fixed-test.jpeg

The only way you can tell if the design has a fixed or flexible width is
by dragging the corner of your browser window to actually resize the page.
Apart from the insanely curious, most people will not sit there resizing
their browser to check the flexibility of your design.

So if people can't tell whether or not the site is fixed width, how can
fixing a particular width be aesthetically beneficial?

Apart from the aforementioned insanely curious, the only other time when
people notice whether or not your page is fixed width is when they notice
that your design requires horizontal scrolling to be able to see all of
the content, or leaves great gobs of useless white space at the side of
their screen.

--
Toby A Inkster BSc (Hons) ARCS
[Geek of HTML/SQL/Perl/PHP/Python/Apache/Linux]
[OS: Linux 2.6.17.14-mm-desktop-9mdvsmp, up 16 days, 9:08.]

Mince & Dumplings
http://tobyinkster.co.uk/blog/2008/02/10/mince-dumplings/

 

[Travis Newbury]

The only way you can tell if the design has a fixed or flexible width is
by dragging the corner of your browser window to actually resize the page.
Apart from the insanely curious, most people will not sit there resizing
their browser to check the flexibility of your design.

Not true at all. You can not have a presentation based site with
flexible as it would be different on everyone's machine. Different
browsers size, different layout. Presentation falls apart. Flexible
width is useless when the layout of the page is important

 

[rf]

On Feb 14, 10:14 pm, Toby A Inkster <[email protected]>
wrote:

Flexible
width is useless when the layout of the page is important

Utter nonsense.

Flexible layout is perfectly useful when the layout of the page is important
and that layout has been designed to be flexible.

All pages start out [being blank] with a flexible layout. It is only when so
called designers impost some arbitrary width on some enclosing element on
the page that they become non flexible. (Then said designers invariably come
here asking how to "centre" that fixed width box in the browser canvas. How
do you centre some 1200 pixel wide box in a 800 pixel wide canvas?)



Impose the width restriction and then design the page and you end up with a
fixed width page.

Fail to impose the width restriction and then design the page and you end up
with a flexible page.

You previously give the example IIRC of pages that must contain things like,
say, a map. Well, look at google maps these days. It used to be fixed width
but is now totally flexible. The map adjusts to exactly fit whatever canvas
it is supplied with.

If the pelicans over at google can do this Travis, why can't you? Nothing
IMHO *needs* to be a certain fixed width if the designer *really* knows what
s/he is doing and can design for the medium rather than some head-in-sand
concept of "screen width".

Would you attempt learn to play the piano but restrict yourself to the
middle two feet of keys? No. You would adapt to the medium and use all the
keys. Why do you insist that it work this way with web pages? Why can you
not adapt and use the entire canvas?

You sir, have IMO proved that you basically fail in this adaption.

 

[Travis Newbury]

Utter nonsense.

What use is flexible width when I haver a static (like print) design?

You sir, have IMO proved that you basically fail in this adaption.

No, I have failed at nothing. I simply disagree with your premise
that flexible width is the only way to go.

 

[rf]

What use is flexible width when I haver a static (like print) design?

If you have a static (like print) design then it belongs in a piece of
bloody paper, not on the web. That is what print means, paper.

 

[Ben C]

On Feb 14, 10:14 pm, Toby A Inkster <[email protected]>
wrote:

Flexible
width is useless when the layout of the page is important

Utter nonsense.

Flexible layout is perfectly useful when the layout of the page is important
and that layout has been designed to be flexible.

All pages start out [being blank] with a flexible layout. It is only when so
called designers impost some arbitrary width on some enclosing element on
the page that they become non flexible. (Then said designers invariably come
here asking how to "centre" that fixed width box in the browser canvas. How
do you centre some 1200 pixel wide box in a 800 pixel wide canvas?)

Easy: give it used left and right margins of -200px each. This is
precisely defined by CSS specifications.

 

[Travis Newbury]

If you have a static (like print) design then it belongs in a piece of
bloody paper, not on the web. That is what print means, paper.

And that's where we disagree isn't it

 

[Jonathan N. Little]

That is patent nonsense. Just looking at a web page it's impossible to
even *tell* whether the design is fixed width or not.

I agree. For site were presentation is important does not me it must be
fixed! That's like saying for a artwork to express depth and dimension
it must be sculpture and not a painting. It just takes a different
approach. I have seen very arty sites that very creatively flow within
the viewport that look like they were designed for that "exact"
dimension. I have tried to do that with my own site. It is just easier
to plunk it down on 800px-wide layout.

e.g. http://examples.tobyinkster.co.uk/fixed-test.jpeg

The only way you can tell if the design has a fixed or flexible width is
by dragging the corner of your browser window to actually resize the page.
Apart from the insanely curious, most people will not sit there resizing
their browser to check the flexibility of your design.

Well with a fixed-site you will know:

|
The only way you can tell if the desi|
by dragging the corner of your browse|
Apart from the insanely curious, most|
their browser to check the flexibilit|
|

I borrowed your text Toby for my demonstration.

 

[Jonathan N. Little]

And that's where we disagree isn't it

See that is where I feel you are in error. If a content is destine for
web delivery then the goal should be suit the media, design with no
preset canvas dimensions from the start. You may not always be
successful, but it should be a goal like a sinless life...this is in
contrast to the attitude where when presentation is important to a
project the foregone conclusion is you must treat the web as paper and
the layout must be fixed! Doomed to sin... ;-)

 

[Travis Newbury]

See that is where I feel you are in error. If a content is destine for
web delivery then the goal should be suit the media, design with no
preset canvas dimensions from the start. You may not always be
successful, but it should be a goal like a sinless life...this is in
contrast to the attitude where when presentation is important to a
project the foregone conclusion is you must treat the web as paper and
the layout must be fixed! Doomed to sin... ;-)

We could (and have) argue all day long, but the bottom line is the
clients I work with have specific goals in mind for their sites. I
deliver that for them. This makes their visitors happy, which puts
money in the client's bank account, which makes the client happy,
which puts money in my bank account, which makes the wife happy. And
we all know, if the wife isn't happy NO ONE is happy....

So no matter if you believe this is right or wrong, it is what I asked
for time and time again. And when I deliver the product everyone is
happy. So until they ask for something else, I will continue to make
sites that make the clients happy.

 

[dorayme]

Easy: give it used left and right margins of -200px each. This is
precisely defined by CSS specifications.

Excellent. Even if it was not so specified, the remark would
still be damn good. <g>

But, call me humourless, at browser on 800px wide,

<http://netweaver.com.au/alt/negativeMarginCenteringJustLeft.html>

works as well as

<http://netweaver.com.au/alt/negativeMarginCentering.html>

In fact at other widths too it is hard to...

 

[Ben C]

Excellent. Even if it was not so specified, the remark would
still be damn good. <g>

Actually it isn't quite specified like that. If you set a fairly small
width and auto left and right margins then the browser gives them equal
values and you get centering.

But if you set a width larger than the available width, then all you get
is a negative right margin (or left if direction is rtl).

But, call me humourless, at browser on 800px wide,

<http://netweaver.com.au/alt/negativeMarginCenteringJustLeft.html>

works as well as

<http://netweaver.com.au/alt/negativeMarginCentering.html>

In fact at other widths too it is hard to...

If you just set the left margin, the right one gets computed for you. In
this case it will end up with the same value as what you set in the
first case.

The browser just solves this equation for margin-right:

margin-left + width + margin-right = containing width

(Padding and borders are in the equation too, but you didn't set any).

The main difficulty with your illustration however is that most browsers
don't seem to allow left scrolling.

I can't imagine why not-- in an Arabic document for example, content is
just as likely to run off to the left as it is to run off to the right
in an English one. As far as I can tell the specification of overflow:
scroll is that you should be able to scroll to see all the contents
somehow, i.e. including ones that are off to the left.

Many web sites (www.bbc.co.uk is one) deliberately offset things miles
off to the left with CSS so you can't see them, relying on the fact that
you can't scroll over there.

Here is a test case: http://www.tidraso.co.uk/misc/negativeMargins.html

 

[dorayme]

Actually it isn't quite specified like that. If you set a fairly small
width and auto left and right margins then the browser gives them equal
values and you get centering.

But if you set a width larger than the available width, then all you get
is a negative right margin (or left if direction is rtl).


If you just set the left margin, the right one gets computed for you. In
this case it will end up with the same value as what you set in the
first case.

The browser just solves this equation for margin-right:

margin-left + width + margin-right = containing width

(Padding and borders are in the equation too, but you didn't set any).

The main difficulty with your illustration however is that most browsers
don't seem to allow left scrolling.

You have put you finger on it, I think this was the source of
most of my bemusement.

I can't imagine why not-- in an Arabic document for example, content is
just as likely to run off to the left as it is to run off to the right
in an English one. As far as I can tell the specification of overflow:
scroll is that you should be able to scroll to see all the contents
somehow, i.e. including ones that are off to the left.

Many web sites (www.bbc.co.uk is one) deliberately offset things miles
off to the left with CSS so you can't see them, relying on the fact that
you can't scroll over there.

Here is a test case: http://www.tidraso.co.uk/misc/negativeMargins.html

A nice illustration.