How to detect table width or height?

[Dennis]

Is there some way --using, say, DOM or javascript-- to detect the
current pixel width and/or height of a relatively sized table or of
one of its columns or rows. I'm going to be writing javascript to
adjust my page to the viewer's browser window dimensions and this
would sure be great information to have.

Thanks ....
Dennis

[Brian]

[a little editing goes a long way when posting replies on usenet]

Barry Pearson wrote:

With images, you can't satisfactorily cater for any resolution. You eventually
have to make some hard decisions.
**Still** refusing to distinguish between content and presentation, I see.
You can design to ensure that
your pages CAN be displayed on a particular viewport size without necessarily
making it worse for the rest
In the context of css, the easiest, most reliable way to design is for
any resolution, any screen size. Textual content will adapt. Images
will not. That is the nature of the beast. I don't design my photos
to fit in a window size. I make the photo first, then make it
available on the www.
It makes a lot of sense to design yourpages so
that they will display well at 800 x 600.

My pages *do* display well on a screen that is 800px wide. If they
don't, then please explain what you see as a problem.

If you want to pick one window size and design for it, then do so
already. Why are you badgering me to do the same? I won't do it.
Would you like me to tell you that I do cater to one screen size, even
though I don't? Would that make you happy?

If you had massive detail, would it be (say) 1024 x 768,

That's a resolution, right? Well, I suppose, by coincidence, I might
come up with a photo that is exactly 1024px x 768px.

Not good! I hope you would take advice before do that.

Are you for real? *You* proposed that hypothetical, not me. Please
stop arguing for its own sake.

The size she chose allows a reasonable representation of her photos
while remaining accessible on most computer screens.

In other words, she took into account the screen sizes of "most computers"!

Right, Barry. Now try to follow this: "most computer screens" means
that the photos are too large for a handheld device, but not too large
for a computer screen, **whatever** the window size, **whatever** the
resolution. The stylesheet for media="screen" does not cater to an
800 x 600 size by using e.g., body {width: 790px}. To do so would be
folly.

As such, her site is viewable on a wide array of computer screen
configurations.

--
Brian
follow the directions in my address to email me

[Barry Pearson]

Brian wrote:

[a little editing goes a long way when posting replies on usenet]
I have edited every response I've made to what you have said!
Barry Pearson wrote:

With images, you can't satisfactorily cater for any resolution. You
eventually have to make some hard decisions.
**Still** refusing to distinguish between content and presentation, I
see.

I am simply stating a fact. People developing inherently visual material,
especially pixel-oriented material, design it with the target screen sizes in
mind, if they have any sense.

If you or anyone here can pose an alternative, please do. There are vast
numbers of photographers and others who need an answer. But we know what the
current answer is - we design for target screen-occupancy, because we have to.

The decision about the size of the material in pixels CANNOT be ducked. Should
this JPEG be 400 x 400 pixels? 600 x 400? 700 x 500? 1024 x 768?

Those decisions make a massive difference to the value of the web site,
including whether or not scrolling is needed to see the image, the download
time, whether you have to look hard to see an image occupying 10% of your
screen, etc.

[snip] In the context of css, the easiest, most reliable way to design is for
any resolution, any screen size. Textual content will adapt. Images
will not. That is the nature of the beast. I don't design my photos
to fit in a window size. I make the photo first, then make it
available on the www.

[snip]

When you prepare your photographs for the web, you SHOULD do it for a
particular window size. Oh ... I forgot. You have said you are not a
photographer. Enough said!

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/

[Mark Parnell]

Sometime around Tue, 14 Oct 2003 17:45:26 +0100, Barry Pearson is reported
to have stated:

But the maximum viewport size is typically just a bit less than the screen size.

Assuming, of course, that the user has their browser window maximised.
Which many don't.

--
Mark Parnell
http://www.clarkecomputers.com.au

[Brian]

Barry Pearson wrote:

I am simply stating a fact. People developing inherently visual material,
especially pixel-oriented material, design it with the target screen sizes in
mind, if they have any sense.
Insult number 2. But I have plenty of sense, thank you. I do not
design for a screen size. How could I? There are too many possibilities.
If you or anyone here can pose an alternative, please do.
Asked and answered.
The decision about the size of the material in pixels CANNOT be ducked. Should
this JPEG be 400 x 400 pixels? 600 x 400? 700 x 500? 1024 x 768?
Why does that last hypothetical size match a resolution? You're not
confusing the two, are you?
When you prepare your photographs for the web, you SHOULD do it for a
particular window size.
Why? Is there something wrong with my sites?
Oh ... I forgot. You have said you are not a photographer. Enough said!

I guess that's another insult. Whatever. HAND.

--
Brian
follow the directions in my address to email me

[Barry Pearson]

Mark Parnell wrote:

Sometime around Tue, 14 Oct 2003 17:45:26 +0100, Barry Pearson is
reported to have stated:

But the maximum viewport size is typically just a bit less than the
screen size.

Assuming, of course, that the user has their browser window maximised.
Which many don't.

I said "the maximum viewport". That is independent of whether the window is
maximised. On an 800 x 600 screen, the maximum viewport size is just less than
that - even if the browser isn't running!

The point about this discussion is so simple that I'm surprised that there is
debate. Someone developing pixel-oriented material, JPEG, GIF, PNG, etc, has
to decide how many pixels to have in the image. Perhaps sometimes there is
actually no choice. (For example, someone showing a screen-capture may well
accept exactly the size dictated by the material as put onto the screen before
capture). But typically there IS a choice. A photographer has a choice about
whether to make a photograph 400 x 300, or 600 x 450, or 800 x 600, or 4000 x
3000, etc. Someone creating a chart from Excel, or a graph, or a diagram from
a modelling tool, also has to make a decison. (I have done all of these at one
time or another).

So, how should the decision be made? Without considering the screen sizes
available to view such images, and for photographs the network speeds, there
is no reference point. Any of the above sizes could be valid. This isn't like
leaving text sizes to the browser-default, where the author can simply avoid
making the decision. The image-developer HAS to make a decision, which then
gets frozen into the content.

Below, there is some pixel-oriented material where I have had to make such a
decision. It is arguable whether I made the right decision in each case. That
can probably really only be judged by the viewer, who might say "on my screen
that image requires horizontal scrolling", or "on my screen that image only
occupies 10% of the screen and is insignificant", or "that photograph takes
half a minute to download on my connection and I would prefer a smaller
version".

No amount of denial in these NGs will alter the basic fact. The developer HAS
to make the decision, and that decision should have a rational basis in order
to be satisfactory to the viewers.

http://www.parliament.the-stationery...98/9091506.htm

http://www.childsupportanalysis.co.u...se_study_1.htm

http://www.childsupportanalysis.co.u...e_children.htm

http://www.barry.pearson.name/photog...95_09_10_3.htm

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/

[Barry Pearson]

Brian wrote:
[snip]

Until you see the need for a rational basis for making decisions about what
size to make pixel-oriented material, I see no point in continuing this
discussion. You appear to be denying the existence of a problem that
photographers and others developing such material, all over the world, know is
a very real problem.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/

[Steve Pugh]

"Barry Pearson" <ne**@childsupp ortanalysis.co. uk> wrote:

The point about this discussion is so simple that I'm surprised that there is
debate.
Maybe the point isn't as simple as you think.
Someone developing pixel-oriented material, JPEG, GIF, PNG, etc, has
to decide how many pixels to have in the image
That is true.
So, how should the decision be made? Without considering the screen sizes
available to view such images, and for photographs the network speeds, there
is no reference point.
Wrong. The reference point is the image itself - the image must be
large enough for all the detail the creator wants to convey to be
apprarant. This must be balanced against the download time for the
user. the creator must decide upon a compromise between download time
and detail. Placing window size above either of those is going to lead
to problems.
Below, there is some pixel-oriented material where I have had to make such a
decision. It is arguable whether I made the right decision in each case. That
can probably really only be judged by the viewer, who might say "on my screen
that image requires horizontal scrolling", or "on my screen that image only
occupies 10% of the screen and is insignificant", or "that photograph takes
half a minute to download on my connection and I would prefer a smaller
version".

http://www.parliament.the-stationery...98/9091506.htm
Yuck. Was this made in Word? I thought you used Photoshop?
The image should be redone, as it is it needs to be larger in order to
be clearer (detail) and at 7kb can afford to be so (download time).

But what window size was a 434 x 299 image designed for?
http://www.childsupportanalysis.co.u...se_study_1.htm
If redrawn to be clearer these images could be smaller in both
physical and filesize terms and at the same time clearer to read.

Again, what window size were these images designed for?
The squeeze the text alongside them into a too narrow column at
800x600
http://www.childsupportanalysis.co.u...e_children.htm
As the graph is only ther to show the trend rather the precise numbers
involved, (if it is there to show the precise numbers then it has
failed,) I would make it smaller and tidier (anti-aliased fonts,
clearer x-axis labels, etc) and float it alongside the text.
http://www.barry.pearson.name/photog...95_09_10_3.htm
At 74kb with quite poor resolution this fails on both the detail and
download time criteria.

From a photographic point of view I don't like the composition. Having
three main points in a line, but facing right, left, right just
doesn't work. In fact an enlargement of just the girl in red would
probably be what I'd take from this shot. I'd go for a larger image of
that portion of the picture, but try to keep the file size smaller
than it is at the moment.

Of the four examples given the last two were designed for 800x600
windows but the others don't seem to have been designed for any window
size. Are you sure you meant to post these particular URIs in support
of yoru argument?
No amount of denial in these NGs will alter the basic fact. The developer HAS
to make the decision, and that decision should have a rational basis in order
to be satisfactory to the viewers.

No one is denying that. They are denying that window size is a
predominant factor in that decision making process.

Steve

--
"My theories appal you, my heresies outrage you,
I never answer letters and you don't like my tie." - The Doctor

Steve Pugh <st***@pugh.net > <http://steve.pugh.net/>

[Barry Pearson]

Steve Pugh wrote:

"Barry Pearson" <ne**@childsupp ortanalysis.co. uk> wrote:

The point about this discussion is so simple that I'm surprised that
there is debate.
Maybe the point isn't as simple as you think.

Someone developing pixel-oriented material, JPEG, GIF, PNG, etc, has
to decide how many pixels to have in the image

That is true.

So, how should the decision be made? Without considering the screen
sizes available to view such images, and for photographs the network
speeds, there is no reference point.

Wrong. The reference point is the image itself - the image must be
large enough for all the detail the creator wants to convey to be
apprarant. This must be balanced against the download time for the
user. the creator must decide upon a compromise between download time
and detail. Placing window size above either of those is going to lead
to problems.

The amount of detail that is apparent depends on the viewing characteristics
of the user. For example, how many pixels per inch on their screen. Is their
monitor calibrated? Are they viewing in subdued lighting? With my target
photograph audience, they will often have pretty good monitors. For the child
support site, they often won't.

Any decision leads to problems. It is a matter of deciding what should take
priority. You say "download time". And that is ...? Once again, it s a matter
of trying to work out what my target audience uses. As it happens, some of the
most important people in my target audience have fast networks and top-end
screens.

Below, there is some pixel-oriented material where I have had to make
such a decision. It is arguable whether I made the right decision in
each case. That can probably really only be judged by the viewer, who
might say "on my screen that image requires horizontal scrolling", or
"on my screen that image only occupies 10% of the screen and is
insignificant ", or "that photograph takes half a minute to download
on my connection and I would prefer a smaller version".

http://www.parliament.the-stationery...cmselect/cmsoc sec/798/9091506.htm
Yuck. Was this made in Word? I thought you used Photoshop?
The image should be redone, as it is it needs to be larger in order to
be clearer (detail) and at 7kb can afford to be so (download time).
(I use Powerpoint for diagrams). Indeed. That is what I thought when I did the
original (523 x 359 pixels, 8.6 KB), at:

http://www.childsupportanalysis.co.u...se_study_1.htm

But Hansard decided I had got it wrong for their readers, so they reduced the
size to 434 x 299. Evidently I didn't study their requirements well enough. (I
don't know of any other coloured diagrams in Hansard, so I had nothing to go
on).

[snip]http://www.childsupportanalysis.co.u...c/wftc_case_st udy_1.htm
If redrawn to be clearer these images could be smaller in both
physical and filesize terms and at the same time clearer to read.
I have a 1400 x 1050 117 ppi screen on my old laptop, and these images are
really too small for it. I know of people with much greater numbers of pixels
per inch who see them even smaller. I think bigger, not smaller, is the way to
go in future. But, of course, that is because you and I are viewing on
different physical screens with different characteristics .

This simply illustrates the dilemma faced by people who develop pixel-oriented
material. Material that I used to develop for VGA screens now looks wrong on
the majority of screens. But what choice did I have then? Hardly to ignore the
problem. No - I had to take into account the screen size at the time. In 5
years time I may examine typical network speeds and screens sizes and rework
some of the images, or at least develop new material to the new size.
Again, what window size were these images designed for?
The squeeze the text alongside them into a too narrow column at
800x600
I used 800 x 600 when I was deciding on the size. I developed to a theme,
which meant accepting that some images were wider than others. (Rather than
resizing images and text to fit a narrow width). In the end, I feel they work
on 800 x 600, and don't squeeze the text too much. The sort of compromise that
people developing pixel-oriented images keep having to face. (They still work,
although far less well, on a 640 x 480 screen. But I chose to accept the
problems at 640 x 480 in order to get a bigger image on 800 x 600, often on a
poor quality screen).

[snip]http://www.barry.pearson.name/photog...95_09_10_3.htm

At 74kb with quite poor resolution this fails on both the detail and
download time criteria.

I allow myself up to about 100KB for a 700 pixel wide photograph. It is, once
again, a compromise. (On that web site I put photographs there with 2 sizes,
and the smaller size is about 38KB). On broadband it downloads plenty fast
enough.

What does with "poor resolution" mean? Should it have more pixels? Isn't it
sharp enough at the current number of pixels? (Can you see the diagonal wire
grid on the gate behind the girl in black? Can you see a glimmering of hair
surrounding the head of the girl in red?

The pre-web version is 5302 x 2470 pixels, about 35 megabytes. Whatever I do,
detail will get lost. Since I can't satisfy everyone, I try to satisfy the
characteristics of my target audience, as far as I can find out. That, of
course, involves talking to them, and asking them on NGs. "Marketing" - it
hasn't gone away.

[snip] Of the four examples given the last two were designed for 800x600
windows but the others don't seem to have been designed for any window
size. Are you sure you meant to post these particular URIs in support
of yoru argument?

Yes. They illustrate the dilemma faced by developers of pixel-oriented
material. The first one shows how important it is to understand the audience -
Hansard reduced my image because I didn't study their requirements. I believe
all of them work (in their original form) on 800 x 600 screens.

No amount of denial in these NGs will alter the basic fact. The
developer HAS to make the decision, and that decision should have a
rational basis in order to be satisfactory to the viewers.

No one is denying that. They are denying that window size is a
predominant factor in that decision making process.

For pixel-oriented material, the screen characteristics are among the most
important factors. Available number of pixels, and the display in pixels per
inch, all have an impact.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/

[Brian]

Barry Pearson wrote:

Until you see the need for a rational basis for making decisions about what
size to make pixel-oriented material,
I already told you how I decided. You chose to argue that I couldn't
be right.
I see no point in continuing this discussion.

Bye.

--
Brian
follow the directions in my address to email me

[Dennis]

On Tue, 14 Oct 2003 23:10:57 +0100, "Barry Pearson"
<ne**@childsupp ortanalysis.co. uk> wrote:

Brian wrote:

[a little editing goes a long way when posting replies on usenet]

I have edited every response I've made to what you have said!

Barry Pearson wrote:

With images, you can't satisfactorily cater for any resolution. You
eventually have to make some hard decisions.

**Still** refusing to distinguish between content and presentation, I
see.

I am simply stating a fact. People developing inherently visual material,
especially pixel-oriented material, design it with the target screen sizes in
mind, if they have any sense.

If you or anyone here can pose an alternative, please do. There are vast
numbers of photographers and others who need an answer. But we know what the
current answer is - we design for target screen-occupancy, because we have to.

The decision about the size of the material in pixels CANNOT be ducked. Should
this JPEG be 400 x 400 pixels? 600 x 400? 700 x 500? 1024 x 768?

Those decisions make a massive difference to the value of the web site,
including whether or not scrolling is needed to see the image, the download
time, whether you have to look hard to see an image occupying 10% of your
screen, etc.

[snip]

In the context of css, the easiest, most reliable way to design is for
any resolution, any screen size. Textual content will adapt. Images
will not. That is the nature of the beast. I don't design my photos
to fit in a window size. I make the photo first, then make it
available on the www.

[snip]

When you prepare your photographs for the web, you SHOULD do it for a
particular window size. Oh ... I forgot. You have said you are not a
photographer . Enough said!

I hope you'll allow me a little extra leeway here since I'm going off
the newsgroup subject area (CSS and HTML), but my remarks are germane
to the issue. The problem of getting multiple columns of text in the
viewport (that prompted me to start this thread) has led me to decide
on using Flash as my main medium. I think Flash will do what CSS is
too immature to yet do. Now, when CSS "catches up", it will surpass
Flash because of its huge potential in separating content from
presentation. It will make a snap out of all the tedious coding I'm
now doing in Flash ActionScript. But to speak to your photo
resolution dilemma, right now today, if you were to use Flash, you'd
know at every moment the height and width of the viewer's browser
window. So you could either just download the image that best fits
their screen or let the Flash player scale whatever size image you
use. Either way, you fill up the screen --if that's what you want--
from a Palm to the widest Apple. I'm not knocking CSS (completely)
here, but I am telling it to get a move on!

Dennis