Applet printing

[JWS]

I finally managed to produce an applet, and put it on my page
(using the <applet> tag). I tried to make it look nice, with the
same background colour as the page itself. It looks just like the
other pictures on the page, the only difference being that it
moves when you press "run" (http://www.jw-stumpel.nl/bounce.html,
bottom of the page).

However, when I print (or print-preview) the page, Firefox (on
Linux and Windows) just shows a blank area where the applet should
be. IE7 prints the applet area correctly.

Does anyone know if this is normal behaviour? Should this be
considered a bug? Is there some trick (perhaps by changing
something in the applet itself, om some trick in print css) to
make FF print applets? Google did not get me very far with this.

Regards, Jan

 

[dorayme]

I finally managed to produce an applet, and put it on my page
(using the <applet> tag). I tried to make it look nice, with the
same background colour as the page itself. It looks just like the
other pictures on the page, the only difference being that it
moves when you press "run" (http://www.jw-stumpel.nl/bounce.html,
bottom of the page).

However, when I print (or print-preview) the page, Firefox (on
Linux and Windows) just shows a blank area where the applet should
be. IE7 prints the applet area correctly.

Does anyone know if this is normal behaviour? Should this be
considered a bug? Is there some trick (perhaps by changing
something in the applet itself, om some trick in print css) to
make FF print applets? Google did not get me very far with this.

Regards, Jan

Just a little thing, you should reconsider your

line-height: 120%;

Best to leave it out completely, or consider using a ratio:

line-height: 1.2; if you must have it in.

The immediate effect of *your value* is to make the main heading
at the top collide when user font-size is upped for comfort and
it runs to 2 lines.

Perhaps you will need more expert help with Java about why the
radio buttons are too crowded together making the text disappear
as in the 6 in the 16 and the 0 in the 50 for choice of Nmasses.

 

[JWS]

Just a little thing, you should reconsider your

line-height: 120%;

Best to leave it out completely, or consider using a ratio:

line-height: 1.2; if you must have it in.

I thought I had that fixed already by specifying some other
line-height for h1, but apparently I hadn't. Thanks for pointing
it out.

Anyway I now followed your suggestion of changing 120% to 1.2
(without doing anything special with h1). To my amazement this
made a difference; the line spacing inside the h1 (if it is longer
than one line) is now correct. It doesn't change anything in the
rest of the page. I thought (from the CSS specs) that 1.2 and 120%
were just the same.

Perhaps you will need more expert help with Java about why the
radio buttons are too crowded together making the text disappear
as in the 6 in the 16 and the 0 in the 50 for choice of Nmasses.

This does not happen in any of the computers (Windows and Linux
only) or browsers that I tested it with. The distances between the
buttons are (must be in this case) fixed in pixels, so it seems
that your system has a bigger default font size than the systems I
used. No particular font was specified; I assumed (apparently
wrongly) that the java default font is always the same. BTW Are
the texts "Nmasses" and "Spring display" also too big? I mean, do
they run into the buttons on the right? They are "text" fonts
which may be different from "UI" fonts.

Regards, Jan

 

[mbstevens]

I finally managed to produce an applet, and put it on my page
(using the <applet> tag). I tried to make it look nice, with the
same background colour as the page itself. It looks just like the
other pictures on the page, the only difference being that it
moves when you press "run" (http://www.jw-stumpel.nl/bounce.html,
bottom of the page).

However, when I print (or print-preview) the page, Firefox (on
Linux and Windows) just shows a blank area where the applet should
be. IE7 prints the applet area correctly.

Most Linuxes don't come with browsers' Java installed.
Are you sure it is installed? -- easy enough to do if it isn't.

This is a reason you might think about whether an applet will
be usable to all your visitors. I stopped using them years ago.

 

[JWS]

Most Linuxes don't come with browsers' Java installed.
Are you sure it is installed? -- easy enough to do if it isn't.

I am pretty sure:

-- I installed Java 6 (a.k.a. 1.6) on my Linux system.
-- I actually wrote the applet myself (on my system).
-- It runs perfectly well on my system. But it does not print
from Firefox on Linux (it does not print from Firefox when I
boot the computer into Windows, either; so far it only prints
from IE7 on Windows XP).

This is a reason you might think about whether an applet will
be usable to all your visitors. I stopped using them years ago.

Yes, from Googling I got the impression that there is a general
dislike of Java applets nowadays, among Web page creators. I do
not quite understand why. Could you elaborate? What alternatives
do you use for putting interactive programs on the Web? I am just
a Java newbie (what you could call a "very very late adopter") but
I was pretty impressed that my applet (which animates fairly
complicated mathematical things) asks nothing more from the server
than to serve up one file, of only 6003 bytes -- less than even a
small .png, .gif, or .jpg picture, for instance.

Regards, Jan

 

[mbstevens]

I am pretty sure:

-- I installed Java 6 (a.k.a. 1.6) on my Linux system.

You also have to make sure that the browser knows how to reach it.

-- I actually wrote the applet myself (on my system).
-- It runs perfectly well on my system. But it does not print
from Firefox on Linux (it does not print from Firefox when I
boot the computer into Windows, either; so far it only prints
from IE7 on Windows XP).


Yes, from Googling I got the impression that there is a general
dislike of Java applets nowadays, among Web page creators. I do
not quite understand why. Could you elaborate?

Actually, I ran across a microscope site a while back that had
an applet animating light paths that I thought was a good use.

But too many people have browsers that do not 'do' applets.

Most applets you see are misused -- for instance, for navigation,
cute animations of waves and such.

What alternatives
do you use for putting interactive programs on the Web?

Server side processing, or DHTML/Ajax _if_ you can make them
degrade gracefully. Flash/Java are still OK for some limited
uses.

I am just
a Java newbie (what you could call a "very very late adopter") but
I was pretty impressed that my applet (which animates fairly
complicated mathematical things)

That might be a good use...

 

[dorayme]

I thought I had that fixed already by specifying some other
line-height for h1, but apparently I hadn't. Thanks for pointing
it out.

Anyway I now followed your suggestion of changing 120% to 1.2
(without doing anything special with h1). To my amazement this
made a difference; the line spacing inside the h1 (if it is longer
than one line) is now correct. It doesn't change anything in the
rest of the page. I thought (from the CSS specs) that 1.2 and 120%
were just the same.

Also try the other thing I suggested and leave it out altogether
(which is more elegant in that less is more, but I do not know if
you have some special need for this figure, eg. in your applet?).
As I understand it, the defaults for most browsers is about 1.2
so no real need in general to put this in. For your purposes
perhaps no one will care about typographical subtleties when you
are discussing the very interesting subject of material bodies
slamming into other bodies? <g>

This does not happen in any of the computers (Windows and Linux
only) or browsers that I tested it with. The distances between the
buttons are (must be in this case) fixed in pixels, so it seems
that your system has a bigger default font size than the systems I
used. No particular font was specified; I assumed (apparently
wrongly) that the java default font is always the same. BTW Are
the texts "Nmasses" and "Spring display" also too big? I mean, do
they run into the buttons on the right? They are "text" fonts
which may be different from "UI" fonts.

I am on a Mac and the view is rather the same in both Safari and
Firefox (to take a couple of excellent browsers), perhaps this
shot of the relevant bit will help you:


http://netweaver.com.au/test/pics/jan_applet.png

Just a thought for you: since your applet is not user sizable,
err on the side of caution and have bigger fonts in it, better to
do this than make them hard to read for some people.

An applet like this looks quite suited to your modelling aims and
well done for giving it a go!

 

[JWS]

Also try the other thing I suggested and leave it out
altogether

No, I wouldn't do that; it is a taste thing. The 20% 'leading'
makes the text (IMHO) more readable than when the lines are just
packed together without strips of lead between them. I tried it
again just now, and decided I really want the 20% extra space.

Do you have an explanation of why 1.2 does not equal 120%?

[..] I am on a Mac and the view is rather the same in both
Safari and Firefox (to take a couple of excellent browsers),
perhaps this shot of the relevant bit will help you:

http://netweaver.com.au/test/pics/jan_applet.png

Thanks much for this. It is really nice with this Mac look. FYI on
my own (Linux/Firefox) system it has this much more impoverished look:

http://www.jw-stumpel.nl/app-snap.png

Not so nice, but at least you can see all the text. I use a
'minimalistic' type of Java which produces small files, and relies
on the user's system to provide GUI elements like buttons, etc.
Basically my studies have not progressed to the point where I can
program more sophisticated Java.

Just a thought for you: since your applet is not user sizable,
err on the side of caution and have bigger fonts in it, better
to do this than make them hard to read for some people.

AFAIK Java applets are never user-sizable. The size (in pixels)
has to be set in the HTML which carries the applet (between the
<applet> tags, <object> tags, or whatever). In this case
the width is 800 pixels. There is simply no room for a bigger
font; in fact, in the case of the Macintosh, it seems I would have
to make the font smaller. It looks though that by juggling it a
little bit, I could make it fit even on the Mac at your present
font size.

But this depends on the font size not changing. If you change the
font size (or type) on your Mac, does the font in the applet
change too, or is it fixed? In Windows and Linux, it is fixed. And
BTW, how about print-previewing applets on the Mac?

An applet like this looks quite suited to your modelling aims

Well, yes, that's what I thought too. I would think that Java
applets are excellent for instructional / educational texts, and
am in fact amazed that they have never really taken off.

and well done for giving it a go!

Thanks.

Regards, Jan

 

[Sherman Pendley]

I finally managed to produce an applet, and put it on my page
(using the <applet> tag). I tried to make it look nice, with the
same background colour as the page itself. It looks just like the
other pictures on the page, the only difference being that it
moves when you press "run" (http://www.jw-stumpel.nl/bounce.html,
bottom of the page).

One of the few well-justified uses of an applet I've seen lately. It's
not required to view the page, but it does add value to the page for those
who can use it. Very well done!

However, when I print (or print-preview) the page, Firefox (on
Linux and Windows) just shows a blank area where the applet should
be. IE7 prints the applet area correctly.

As an additional data point, it doesn't print on FireFox/Mac either, but
it *does* print on Safari/Mac.

sherm--

 

[dorayme]

No, I wouldn't do that; it is a taste thing. The 20% 'leading'
makes the text (IMHO) more readable than when the lines are just
packed together without strips of lead between them. I tried it
again just now, and decided I really want the 20% extra space.

Do you have an explanation of why 1.2 does not equal 120%?

As I said at the time of raising the idea of leaving it out, the
default for most browsers is about 1.2. In other words, the
browsers already use a css style sheet and on this style sheet
you will see a value for line height. It is not 1. That would be
too squashed for many fonts and sizes as you are correctly
implying above.

As you might appreciate, a pure number acts as a multiplier of
the elements font-size. But a percentage is different. I rewrote
the following to include a spiel on percentages which might
explain the situation to you a bit. Please forgive me if there
are any typos or mistakes. Will correct any that are pointed out
of course:

http://netweaver.com.au/test/lineHeight.html

If you change the
font size (or type) on your Mac, does the font in the applet
change too, or is it fixed? In Windows and Linux, it is fixed. And
BTW, how about print-previewing applets on the Mac?

It is fixed. And yes, it appears fine in my print preview from
Safari. But not in Firefox.

(Work on trying to get the applet to actually play on the printed
page <g>)

 

[Ben C]

No, I wouldn't do that; it is a taste thing. The 20% 'leading'
makes the text (IMHO) more readable than when the lines are just
packed together without strips of lead between them. I tried it
again just now, and decided I really want the 20% extra space.

Do you have an explanation of why 1.2 does not equal 120%?

[...]
As you might appreciate, a pure number acts as a multiplier of
the elements font-size. But a percentage is different. I rewrote
the following to include a spiel on percentages which might
explain the situation to you a bit. Please forgive me if there
are any typos or mistakes. Will correct any that are pointed out
of course:

http://netweaver.com.au/test/lineHeight.html

I think the point is that if you inherit a percentage line-height, you
get that percentage of the font-size of the ancestor that you inherited
line-height from.

This is because you don't inherit it as a percentage, but you inherit
the result of that percentage applied to the ancestor it was set on.

But if you inherit a pure number line height, you inherit it _as a pure
number_, and you get that multiplied by your own font-size.

The spec is rather unclear on this:

<number>

The used value of the property is this number multiplied by the
element's font size. Negative values are illegal. The computed value [p.
88] is the same as the specified value.

<percentage>

The computed value [p. 88] of the property is this percentage
multiplied by the element's computed font size. Negative values are
illegal.

They sound the same, until you consider inheritance: you compute the
computed value before bequeathing a property to your descendents.

So, in your example, #percentage has line-height: 72% and font-size:
50px. Its computed line-height is therefore 36px. So <p class="small">
inherits 36px directly, not 72%.

But if you give #percentage line-height of .72, then <p class="small">
inherits .72 (since for <number>, above, "the computed value is the same
as the specified value"). That's why in that case it gets .72 of its
_own_ font-size.

 

[Andrew Thompson]

...What alternatives
do you use for putting interactive programs on the Web? ..

Your applet is well suited to being launched using Java
Web Start* (JWS, coincidental with your posting name!).
It would end up as 'free floating', but then, this type of
animation applet hardly needs a browser window wrapped
around it.

If you wanted to take it further, it seems also well suited to
being deployed as a screensaver, I could throw it up on-screen
and watch those harmonic patterns for hours. If interested,
check the SaverBeans** project.

* <http://www.physci.org/jws/#jtest>
** <https://screensavers.dev.java.net/>

Andrew T.

 

[JWS]

As I said at the time of raising the idea of leaving it out, the
default for most browsers is about 1.2.

But not for Mozilla products on Linux! I definitely need the line
in the css. That is why I didn't understand, at first, your
suggestion to leave it out. I just rebooted into Windows and
messed with my css, and you are right, 20 % of leading is applied
by IE and Firefox by default (I do not know where exactly this is
set). So I would suggest that for the benefit of the Linux users,
line-height: 1.2 (not 120%) should be put into the css. It does
not harm the Windows and (presumably) Mac users.

Linux users can also set it by default on the user side by editing
userContent.css, but I would guess most users do not even know
that this file exists.

Regards, Jan

 

[dorayme]

On 2007-11-03, dorayme

I think the point is that if you inherit a percentage line-height, you
get that percentage of the font-size of the ancestor that you inherited
line-height from.

This is because you don't inherit it as a percentage, but you inherit
the result of that percentage applied to the ancestor it was set on.

Certainly right. I chose the absurdly big 'ancestral' 50px to
help see a case of this.

The spec is rather unclear on this:

I agree. As so often, One stares at the specs, forms about 4
alternative pics in the brain as to what the meaning or
implications really are, shrugs, makes some test cases and the
pennies start to drop - with any luck. Then one goes back to the
specs and they look a bit clearer. <g>

 

[Ben C]

Certainly right. I chose the absurdly big 'ancestral' 50px to
help see a case of this.


I agree. As so often, One stares at the specs, forms about 4
alternative pics in the brain as to what the meaning or
implications really are, shrugs, makes some test cases and the
pennies start to drop - with any luck. Then one goes back to the
specs and they look a bit clearer. <g>

Indeed, I only figured out what the spec was talking about after looking
at your example.

But if you read the bit about "cascading" it does explain (more or less)
that it's computed values that are inherited. That suddenly makes the
line-height section make more sense.

 

[dorayme]

Indeed, I only figured out what the spec was talking about after looking
at your example.

But if you read the bit about "cascading" it does explain (more or less)
that it's computed values that are inherited. That suddenly makes the
line-height section make more sense.

Yes, it does. All a matter of it registering, be nice if it was
at the time one consults the specs. I have always been impressed
by how some people here (including you of course) seem able to
read the specs with as much clarity as you do. I tend to have to
do it with more hindsight. I recall Jukka Korpela pointing out
about the trap of units for line-height in an exchange with me
ages ago and it was a salutary lesson.

 

[Ben C]

Yes, it does. All a matter of it registering, be nice if it was at the
time one consults the specs. I have always been impressed by how some
people here (including you of course) seem able to read the specs with
as much clarity as you do. I tend to have to do it with more
hindsight.

It's all hindsight on my part too, but you do reach a point (after a lot
of trying things out) where the penny drops and you can tell what most
of it is talking about.

But then it still isn't so easy to explain it because you find yourself
starting to use the same incomprehensible jargon yourself.

I recall Jukka Korpela pointing out about the trap of units for
line-height in an exchange with me ages ago and it was a salutary
lesson.

As far as I can see one almost always would want to use 1.2 rather than
120%. That's the simple lesson to take away here.