Desktops

[Asher]

Drake and Agathon, here's a quiz so you can prove you actually can order them like that.

Identify which is from which.

[Q Classic]

ah, excellent. i got on a friend's computer (he uses gnome rather than KDE) and they also have sub-pixel hinting for font-smoothing.

he didn't have arial, but he did have times new roman, so.

this is in gnome 2.6/openoffice 1.1

[Asher]

Don't Gnome and KDE both use XFT?

And don't individual applications in Linux still need to be compiled to support font AA? (ie, why there's still mozilla-firefox-gtk2-xft builds...)

[Spiffor]

Originally posted by Asher
Identify which is from which.

1. Windows
2. KDE
3. Mac

[Q Classic]

you mean xfs (xfontserver)?
yes and no. you can disable it. i've disabled it on mine, i don't know what he's done.

some apps have to have it compiled in. i'm on gentoo, so i've added it to the default use flags, so. he's on gentoo too, so i dunno... i don't really pay much attention to it anymore, since i've got a stable system and aren't doing as much compiling ('cept for new kernels).

[Asher]

To be honest, I can't tell them apart.

[Asher]

Originally posted by Q Cubed
you mean xfs (xfontserver)?
yes and no. you can disable it. i've disabled it on mine, i don't know what he's done.

some apps have to have it compiled in. i'm on gentoo, so i've added it to the default use flags, so. he's on gentoo too, so i dunno... i don't really pay much attention to it anymore, since i've got a stable system and aren't doing as much compiling ('cept for new kernels).

No, I'm referring to XFT: http://www.scottbolander.com/mozilla-xft.html

[Spiffor]

Originally posted by Asher
To be honest, I can't tell them apart.

So, did you ask the question without knowing the answer?

[Grrr]

On topic peoples .

[Asher]

Originally posted by Spiffor
So, did you ask the question without knowing the answer?

I know the answer, I made the image from Q^3's.

[Q Classic]

xft? it's not compiled into my kde, isn't in the ebuilds, and i didn't use it as a use flag. it also doesn't show up in my world file, so i'd guess not, unless it's part of the basic kde package.

nice, grr. is that one desktop over two monitors? if it is, i'm envious.

[Q Classic]

aha. found xft, it's part of xfree. reading up on it, kde and gnome don't have to use it, but can.

[Grrr]

Originally posted by Q Cubed
xft? it's not compiled into my kde, isn't in the ebuilds, and i didn't use it as a use flag. it also doesn't show up in my world file, so i'd guess not, unless it's part of the basic kde package.

nice, grr. is that one desktop over two monitors? if it is, i'm envious.

Yep, that is one dual monitor desktop, 2560x1024 all the way.

Anyway, here is my lappie's desktop:

[Q Classic]

purdy.

as shown earlier, mine is indistinguishable from windows classic, save for the start menu, where you can first see the difference.

[Asher]

On-topic. Some changes.

[joncha]

Again with the painful desktop. Are you trying to make me go blind?

[Asher]

I love brilliant colors. Adds some life to...life.

[joncha]

So... I'm too lazy to go back and check. What did you change? Did Drake T. spellcheck it for you?

[Asher]

Higher resolution icons, different icons for some things.

Plus the battery life is different.

[joncha]

Actually, the battery monitor is the only thing I did notice.

[Agathon]

Originally posted by Asher

This is Fitts' Law:

MT = a + b log₂(2A/W)

where:
MT = movement time
a,b = regression coefficients
A = distance of movement from start to target center
W = width of target

Then it's false if it's meant to describe what normal people do.

[Asher]

Originally posted by Agathon
Then it's false if it's meant to describe what normal people do.

Fitts' law is "false"?

It's a generalized metric to help determine how quickly somebody can access an element on the screen with the mouse.

It's not meant to describe what normal people do, which is precisely why I was trying to correct you repeatedly...

[Agathon]

This is a silly debate anyway.

There are four different settings in Mac OS X. One is standard anti-aliasing, which is optimal for CRT displays (like mine). The other three methods work using the same method as Cleartype and are best for flat panel displays. Frankly all of them suck compared to the standard anti-aliasing in my view. This is really evident when you need to read Classical Greek, which has all the odd accent marks. Sometimes it's hard to tell which accent is which.

So if you want my view - antialiasing just looks better.

IIRC XP does not anti alias, it uses only one method. That's why it doesn't look so hot on a CRT monitor. To be fair, I've only played with Cleartype on one laptop quite some time ago - I showed my friend how to use it, she hated it though and went back to unsmoothed fonts.

Plus IIRC Cleartype does not work for non-roman fonts. X will smooth anything including all the Oriental characters.

[Agathon]

Originally posted by Asher

Fitts' law is "false"?

It's a generalized metric to help determine how quickly somebody can access an element on the screen with the mouse.

I thought it was supposed to be much more general than that. Didn't the guy do his experiments in the 50s anyway?

It seems to me that given that some people find moving left to right easier than the converse, it can't just be a matter of size and distance. One could imagine various other complications.

[Asher]

Originally posted by Agathon
This is a silly debate anyway.

It only becomes silly and trivial after you've lost.

IIRC XP does not anti alias, it uses only one method.

This makes no sense whatsoever to me. To me this says that anti-aliasing, by definition, has to "use more than one method"? Plus, this is not true anyway. There is ClearType and there's the standard anti-aliasing that's been around since Win98 as well.

Plus IIRC Cleartype does not work for non-roman fonts. X will smooth anything including all the Oriental characters.

You do not recall correctly.

[Inverse Icarus]

how do i turn on this cleartype?

[Asher]

Originally posted by Agathon
I thought it was supposed to be much more general than that. Didn't the guy do his experiments in the 50s anyway?

Yes, mid-50s.

It seems to me that given that some people find moving left to right easier than the converse, it can't just be a matter of size and distance. One could imagine various other complications.

There are many other complications. HCI is a very complex field, most people don't understand just how complex it is.

It's harder than it seems to make a genuinely functional and easy-to-use interface.

[Agathon]

Originally posted by Asher

It only becomes silly and trivial after you've lost.

Hardly. Rather it is exactly what every thread becomes when you can't admit that you are at odds with the facts.

This makes no sense whatsoever to me. To me this says that anti-aliasing, by definition, has to "use more than one method"? Plus, this is not true anyway. There is ClearType and there's the standard anti-aliasing that's been around since Win98 as well.

No it doesn't. It's just that there are two kinds of font smoothing available on Mac OS X and that the standard one works better than anything else for CRT displays. To my eyes Cleartype is clearly inferior on a CRT display. Otherwise the differences are not that great.

You do not recall correctly.

Apparently not. Still can't remember where I'd heard that.

Anyway, that looks ugly.

[Asher]

Originally posted by Uber KruX
how do i turn on this cleartype?

This page tells you what it is and how it works, very generally: http://www.microsoft.com/typography/...ClearType.mspx

Then enable it and calibrate it here: http://www.microsoft.com/typography/...pe/tuner/1.htm

[Agathon]

Originally posted by Asher

Yes, mid-50s.


There are many other complications. HCI is a very complex field, most people don't understand just how complex it is.

It's harder than it seems to make a genuinely functional and easy-to-use interface.

Well then, for once we agree.