Re: Xinerama Status?

2007-06-22 Thread Thomas Themel
Hi,
Clemens Fruhwirth ([EMAIL PROTECTED]) wrote on 2007-06-22:
> Can someone give me a status update on the latest Xinerama support?
> Just a few lines, whether it works or if it's broken. Depending on the
> answer I might want to bump my xft patches to the latest RC and create
> a separate distribution tar ball.

It works for me with the latest rc, after being broken in the one
before.

ciao,
-- 
[*Thomas  Themel*] Das passiert den meisten Systemen die gut funktionieren,
[extended contact] man hört nichts davon, sie geraten in Vergessenheit und
[info provided in] verschwinden dann.
[*message header*]  - Erwin Burgstaller in at.linux


Re: Xinerama

2007-04-11 Thread Jarrod Lowe

On 09/04/07, Thomas Themel <[EMAIL PROTECTED]> wrote:

Hi,
Thomas Themel ([EMAIL PROTECTED]) wrote on 2007-04-09:
> I'll publish a complete version including the build/install stuff and
> docs after a little testing (maybe after I figure out the bindmap stuff
> that tuomov was referring to :)).

Okay, since it works for me, here goes:

darcs get http://wannabehacker.com/repos/mod_xinerama-3/

or via HTTP: http://wannabehacker.com/src/mod_xinerama-20070409.tar.gz

It works on my personal unecological penis enlargement using a pretty
much unmodified ion3 config. If you encounter any problems, please let
me know.  Thanks to tuomov for the outline (and of course the window
manager running it).


Very nice! I was a bit confused for a bit when dopath("mod_xinerama")
did nothing, but adding it to cfg_ion.lua and "killall -USR1 ion3"
applied it. I haven't been using it very long, but it seems to "just
work."

Thanks muchly!

--
Jarrod Lowe


Re: Xinerama

2007-04-10 Thread Seb

On 4/9/07, Thomas Themel <[EMAIL PROTECTED]> wrote:

Hi,
Thomas Themel ([EMAIL PROTECTED]) wrote on 2007-04-09:
> I'll publish a complete version including the build/install stuff and
> docs after a little testing (maybe after I figure out the bindmap stuff
> that tuomov was referring to :)).

Okay, since it works for me, here goes:

darcs get http://wannabehacker.com/repos/mod_xinerama-3/

or via HTTP: http://wannabehacker.com/src/mod_xinerama-20070409.tar.gz

It works on my personal unecological penis enlargement using a pretty
much unmodified ion3 config. If you encounter any problems, please let
me know.  Thanks to tuomov for the outline (and of course the window
manager running it).

ciao,


Thank you very much! Great job.
I have been thinking about doing it myself but I never found the time...

Seb

--
I've never played by the rules, but I've played with the rules.


Re: Xinerama

2007-04-09 Thread Thomas Themel
Hi,
Thomas Themel ([EMAIL PROTECTED]) wrote on 2007-04-09:
> I'll publish a complete version including the build/install stuff and
> docs after a little testing (maybe after I figure out the bindmap stuff
> that tuomov was referring to :)).

Okay, since it works for me, here goes:

darcs get http://wannabehacker.com/repos/mod_xinerama-3/

or via HTTP: http://wannabehacker.com/src/mod_xinerama-20070409.tar.gz

It works on my personal unecological penis enlargement using a pretty
much unmodified ion3 config. If you encounter any problems, please let
me know.  Thanks to tuomov for the outline (and of course the window
manager running it).

ciao,
-- 
[*Thomas  Themel*] Freedom includes the freedom to be foolish,
[extended contact] to be sick. Free choice includes the freedom to
[info provided in] choose badly.
[*message header*]  - P.W. Huber, Orwell's Revenge


Re: Xinerama

2007-04-09 Thread Thomas Themel
Hi,
Nick Murdoch ([EMAIL PROTECTED]) wrote on 2007-04-07:
> How would I go about installing that module?

I'll publish a complete version including the build/install stuff and
docs after a little testing (maybe after I figure out the bindmap stuff
that tuomov was referring to :)).

ciao,
-- 
[*Thomas  Themel*]
[extended contact] /* Conforms to the XENIX standard. */
[info provided in] - Microsoft Platform SDK errno.h
[*message header*]


Re: Xinerama

2007-04-07 Thread Nick Murdoch
Thomas Themel wrote:
> Hi,
> 
> I just had a first stab at implementing a Xinerama module as outlined in
> Tuomo's 2007-01-17 post[0], see attached mod_xinerama.c.

Well done! I've been missing ion3's new features since I switched back
to ion2 :)

How would I go about installing that module?



Re: Xinerama

2007-04-07 Thread Thomas Themel
Thomas Themel ([EMAIL PROTECTED]) wrote on 2007-04-07:
> Hints?

After a good night's sleep, I figured out that MPLEX_ATTACH_UNNUMBERED
is what I was looking for.

ciao,
-- 
[*Thomas  Themel*] "Die meisten Raucher sterben nicht an
[extended contact]  Lungenkrebs, sondern erfrieren auf
[info provided in]  Balkonen."
[*message header*]   -- Timo Maier in <[EMAIL PROTECTED]>


Doing more with less & Re: Xinerama support

2007-02-01 Thread Tuomo Valkonen
(Perhaps it would be better to take the part of this thread not directly
related to Xinerama off-list. Some people may be getting annoyed. But,
whatever: I don't particularly like on-topic police myself, due to crappy
technologies. Mailing lists suck, as there's no simple light-weight way to
redirect discussions elsewhere, as in Usenet. But Usenet/NNTP is also too
clumsy too with its rigid or spammed hierarchies.)


On 2007-02-01, G.H. <[EMAIL PROTECTED]> wrote:
> Athcool: I have tried any software possible (?) but I prefer 'powernowd'
> which was successful in mad tasks such as my old amd64 3500's

"old"? Doesn't sound old at all. My AthlonXP 2500+ (from mid-2003, IIRC) 
doesn't even have that frequency control shit yet, and I don't consider
it old. It only supports whatever athcool turns on (it doesn't actively
control anything). (My older Duron 1GHz would, indeed, be quite sufficient
for my needs, but its mobo leaked the juices out of its capacitors only
after a year, without showing any signs until later, and so I had to get
new mobo/cpu/memory, as finding out the broken one of them would've 
been too laboursome.)

> And I think the main goal on using GNU/Linux is about pushing out the
> capabilities of cheap hardware... I know...

Doesn't seem like people are doing that. No, they're getting bigger
penis enlargements instead of...

> Multihead:
> ---
> I don't agree with some arguments proposed in early posts. There are very
> many ways of serious criticism on the techno-empiricism exposed here: In
> my modest opinion (which is not mine indeed) the brain is just tied to the
> senses but no directly "co-responsible" in the cognitive functions, so is
> not possible to measure the cognitive capabilities of one individual or
> his psychical-motor development in the perspective of his angle of vision
> [...] That is indeed a reductionist explanation, and the possibilities
> into the multihead settings are not just focused on vision but rater in
> the interaction and expressiveness.
> 

... doing more with less. That's what my multihead criticism is all about:
doing more with less. Multihead is wasteful: you don't technically need it
to do what you want to do, unless penis enlargement is precisely what you
want. As for the psychological side, i.e. your need for penis enlargements,
I don't care. I just care what you purportedly need that wasteful extra
screen (or other gizmo) for, and whether you technically really need it for
that. I can be the judge of technical need if you provide the necessary
information, and most of the time, there is not one, because people don't
dare to say that it's penis enlargement precisely, what they want.

You know, you probably could survive without that computer too. You 
don't need it to survive. But you need it for many other things, that
I can agree with. And a screen too. But you don't need multiple screens,
or a really big screen to do these things, just a modest-sized screen.
You also shouldn't need a very powerful computer, and certainly not
an x86 frying pan, but unfortunately these things are a bit out of
any single person's control... 

(There are processors that demand far less energy for the same computing
power as the x86 mount de manure, although their peak power may not be as
high. But by using multiple low-power processors (or multicore) in parallel,
you could get a lot of computing... That's where I think things will
eventually head, if people get into their senses: many low-power 
computing units instead of a single centralised frying pan.)

As you can see, it's all about doing more with less, and multihead is
antithetic to that for most uses. As for a penis enlargement, well,
rather a screen than a car, if people would just admit that's what they
want it for. But better without either. (An urban automobile is like a
penis with AIDS and pus, and an SUV the over-bloated same. Now, a
train/metro/etc., that's a decent-sized healthy penis... and far more
energy-efficient than that gas-guzzling murderous tin can.)

> but dividing the screen by two in 17" monitors perhaps is not good
> enough.

It is. That's all I've ever had, and works great. For programming, for
latex+pdf/dvi, for everything I've ever had to do. At least with Ion. And
if you're using something worse, well, you could do more with less by
switching to Ion. Or creating something even better. Of course you might
be doing something that I don't do, that really does demand more screen
space even in my opinion, but those handful of people most certainly
people do not include techno-toy fetists bragging about their zillion
monitor configurations, and crying after broken Xinerama support in Ion.

> Finally, I am just thinking about the technical possibility of making
> work Ion with some Xinerama alternative, perhaps MergedFB on ati
> devices. Maybe repeating somebody in the list, would be possible to
> point in that direction considering there are not m

Re: Xinerama support

2007-02-01 Thread G.H.
I will point some things about Xinerama which I don't found in the list, but I 
am
tempted to expose some non-Xinerama issues which many people have pointed too.
So skip until Multihead section if you are interested but don't have time 
enough.

Athcool:
I have tried any software possible (?) but I prefer 'powernowd' which was 
successful
in mad tasks such as my old amd64 3500's overclocking: I used to have my 
processor with
his default fan at speed limits not googled ;> and even impossible rates for 
Windows XP
(which downsized by a half the bogomips in my old machine) craziness [...] The 
thing
remained cool all the time.

Suspend:
In the Linux side, other capabilities like software suspend really sucked in 
every computer
tested, including my new Laptop HP dv8305. And I even don't need to repeat 
anything about
'network interfaces', /dev/cryptopsy, neither including 5 in 1 cards support or 
the mystery
behind the Linux keyboard support: my keyboard lost 4 keys.

However, HP dv8000 series are good enough and I can't realize how to adjust the 
budget
only for suspending my chick: Thinkpads are actually expensive. And I think the 
main goal on
using GNU/Linux is about pushing out the capabilities of cheap hardware... I 
know...

Video Drivers:
I have tortured my self trying to get ati drivers working. Though I realized 
how to
get the 3D extension working properly after _three months_ of struggling with 
linux 2.6.17
+ Xorg 7.1 + fglrx drivers. I find interesting ati drivers doesn't depend at all
on Xinerama to get 'multihead' support, there is 'MergedFB'. I think nvidia has 
already an
extension like that, but I cannot remember.

Multihead:
---
I don't agree with some arguments proposed in early posts. There are very many 
ways of serious
criticism on the techno-empiricism exposed here: In my modest opinion (which is 
not mine indeed)
the brain is just tied to the senses but no directly "co-responsible" in the 
cognitive functions,
so is not possible to measure the cognitive capabilities of one individual or 
his psychical-motor
development in the perspective of his angle of vision [...] That is indeed a 
reductionist
explanation, and the possibilities into the multihead settings are not just 
focused on vision
but rater in the interaction and expressiveness. 


That said, multihead would be a good "penis extension" if you think about 
graphic design,
the torturing GUI programming (there are no VIS nowadays ;)), writing in 
pdf-latex and so on.
Ion 3ds-20061223 is great, behaves better with those named floating windows, 
but dividing
the screen by two in 17" monitors perhaps is not good enough.
  
Finally, I am just thinking about the technical possibility of making work Ion 
with some
Xinerama alternative, perhaps MergedFB on ati devices. Maybe repeating somebody 
in the list,
would be possible to point in that direction considering there are not much 
documentation about?

Ion is the best thing that happened into the GNU/Linux 'desktops' (a true 
revolution which
perhaps Tuomo itself even don't realize) so I don't abandon the hope ... 

-- 
If god exists, would be not necessary to invent him.
Mario G.H.



Re: Xinerama support

2007-02-01 Thread Matthieu Moy
Tuomo Valkonen <[EMAIL PROTECTED]> writes:

> All I've ever had swsusp do, is crash and panic.

I had a funny suspend to RAM too: suspend worked perfectly, but is
just forgot to resume, so, I had to hard-reboot the box to restart it.
Hmm, it slightly decreases the advantage of suspend to RAM ;-).

-- 
Matthieu


Re: Xinerama support

2007-01-31 Thread Tuomo Valkonen
On 2007-01-31, Joseph Goncalves <[EMAIL PROTECTED]> wrote:
> If it wasn't working on my laptop then I wouldn't of started using Linux=20
> on my laptop. Luckily for me suspend2 works well on my laptops.

All I've ever had swsusp do, is crash and panic. And even if it didn't 
crash, I seriously doubt it works seamlessly with X (and the proprietary
NVidia drivers). Not that I have been able to try it recently, because 
upgrade the kernel is practically impossible. And how about encrypted 
swaps? Probably needs some serious hacking not nicely and _safely_ 
integrated into distros to get it working.

[[Running over a year old 2.6.14. Newer versions AFAIK no longer have
decent SATA drivers with the devices as predictable /dev/hde etc., 
instead demanding random SCSI mapping insanity and hence udev. An udev
itself is crypto-slugware, the kid of crap that I will not have on my 
system. Plus compiling the kernel is a week long task: one day trying 
make oldconfig to notice that it doesn't work, a few days for the 
initial configuration, a few more days fixing it, and then a few days
trying to restore the  system to be able to run the old kernel. 
Linux sucks. 

And the stock kernels in the distributions boot years loading every
single module in a random order, so that you dont't get the integrated
buzzchip as the first sound card, the wrong ethernet interface as eth0
and so on. And the tools to customise the initrd demand -- you guessed
it -- the plague of udev. Linux sucks.

Today the network went down and apparently the kernel can't handle my
static and dynamic IP not being in the same subnets now, not finding
the routes. Newer kernels might not have the bug, but I can't upgrade
for the above-mentioned reasons. Linux sucks.

I wish FreeBSD got ice1712 driver so I could perhaps switch to
Debian/kFreeBSD. Linux sucks.]]

-- 
Tuomo



Re: Xinerama support

2007-01-31 Thread Joseph Goncalves
On Tuesday 23 January 2007 07:50, Tuomo Valkonen wrote:
> On 2007-01-22, Bruce Stephens <[EMAIL PROTECTED]> wrote:
> > (One of our concerns at work was heat from CRTs (perhaps bizarrely,
> > given that we're in south west London, but we really had been in an
> > office where heat was a significant problem, though we aren't,
> > now).)
>
> [And speaking of heat and power consumption, the state of software
> suspend in Linux totally sucks, and yet it's one of the most
> important features I can think of. It sucks having to leave the
> computer on when you don't use it, because it's such a _chore_ to
> bring it and all the programs up again, especially as session
> management support sucks too (and there's no protocol it for terminal
> programs). The hardware suspend support of desktop mobos also tends
> to be poor, PSU fan still keeping humming and so on. (And I haven't
> heard of that good things of the Linux laptop support either.) That's
> one very good reason to use Windows: hibernate. Of course, even
> better would be if the computer could run on _really_ low power,
> shutting down all the fans, not using the disks except very
> infrequently, and so on, but still react to ssh connection attempts
> etc. ...). Well, at least with 'athcool' I managed to make the system
> run 10C cooler most of the time.]
If it wasn't working on my laptop then I wouldn't of started using Linux 
on my laptop. Luckily for me suspend2 works well on my laptops.


pgpFT1kExc46j.pgp
Description: PGP signature


Re: Xinerama support

2007-01-31 Thread Tuomo Valkonen
On 2007-01-31, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> just use right hardware ;) for example, hibernate and suspend to ram
> works fine on thinkpads.

Does it also spin down the couple of external hds? (Because the laptop 
surely isn't going to have enough disks pace.) How about the box 
providing a decent sound card? (I've yet to see a tolerable integrated
chip that didn't completely distort the sound.)

Laptops make quite lousy and expensive tabletop computers, what with
an unnecessary screen and keyboard. (The foldable 17" computers aren't
that good either. No, they're not laptops. More like dragtops.) Laptop
keyboards are crap, just like tapping the table. And with the extra 
keyboard on the way, the screen becomes unusable as well (in addition 
to being already otherwise rather tiny for tabletop usage -- and too 
big to be mobile, really [1]).

---

  [1] http://modeemi.fi/~tuomov/b/archives/2006/08/04/T21_13_29/

-- 
Tuomo



Re: Xinerama support

2007-01-31 Thread alex

just use right hardware ;) for example, hibernate and suspend to ram
works fine on thinkpads.

thanks, Alex

> Tuomo Valkonen (TV) writes:

 TV> [And speaking of heat and power consumption, the state of software
 TV> suspend in Linux totally sucks, and yet it's one of the most important
 TV> features I can think of. It sucks having to leave the computer
 TV> on when you don't use it, because it's such a _chore_ to bring it 
 TV> and all the programs up again, especially as session management support
 TV> sucks too (and there's no protocol it for terminal programs). The 
 TV> hardware suspend support of desktop mobos also tends to be poor, PSU fan
 TV> still keeping humming and so on. (And I haven't heard of that good
 TV> things of the Linux laptop support either.) That's one very good reason
 TV> to use Windows: hibernate. Of course, even better would be if the computer
 TV> could run on _really_ low power, shutting down all the fans, not using 
 TV> the disks except very infrequently, and so on, but still react to ssh 
 TV> connection attempts etc. ...). Well, at least with 'athcool' I managed
 TV> to make the system run 10C cooler most of the time.]

 TV> -- 
 TV> Tuomo



Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Tuomo Valkonen <[EMAIL PROTECTED]> wrote:
> Here, AFAIK, you're supposed to be able to inspect the product just 
> like you could in a shop, and to do that, you need to plug it in.

The difference of course being, that in a shop you can't inspect
that _particular_ item you get, but the display article. And being
able to inspect the one you get is the crux of getting a TFT without
dead pixels.

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Bruce Stephens <[EMAIL PROTECTED]> wrote:
>>> (At least in .fi, it's best to order TFTs online, BTW: the law dictates
>>
>> Wow. Sounds like a civilised place to live. :)

It isn't.

> I believe the law here's usually interpreted (by vendors, anyway) such
> that if you actually open the box (and, even worse, plug the device
> in), then you can no longer use the DSR to return it.

Here, AFAIK, you're supposed to be able to inspect the product just 
like you could in a shop, and to do that, you need to plug it in.
Basically the returned product has to be in a shape that it can still
be sold as new.

> In the meantime using two relatively cheap TFTs seems reasonable.

A single TFT is more reasonable, unless you _really_ need the space,
(i.e. can justify your need to me and other critics). Most don't, 
least of all for their home setups. (I suppose Seth here may have
a proper need, if he _has to_ _monitor_ the _content_ -- and not 
just  something in general happening -- of an IRC discussion 
constantly (modulo the product being supported being of any positive
value), while doing something else. A rather small screen should 
suffice for that, though, e.g. that of a laptop if the computer 
itself was the laptop.)

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Bruce Stephens
Seth Arnold <[EMAIL PROTECTED]> writes:

[...]

> (I rarely need references while programming beyond 'man' in the
> scratchpad, so the single display isn't a serious concern when
> programming. :)

Depends what you're doing, of course.  Often I'm working with ITU
standards (X.511, etc.), and I have those in PDF.  For that matter,
quite a few API docs seem to be in annoying frames-requiring HTML,
which don't really squash that well.

[...]

>> (At least in .fi, it's best to order TFTs online, BTW: the law dictates
>> an unconditional 14 day right of return, so if there's a single dead 
>> pixels, just return it, no questions asked. Can't do that with IRL 
>> shops.) Anyway, unless something really remarkable happens (like 
>
> Wow. Sounds like a civilised place to live. :)

Depends on how the member state has implemented the EU directive, of
course.  And how that law's interpreted by courts and companies.  In
the UK I believe the intention is that the Distance Selling
Regulations provide a similar experience to consumers for things
they've bought at a distance to what they'd get in a shop.

Specifically, if we buy over the web we can't (necessarily) see the
physical box, read what it's got on it, etc.

I believe the law here's usually interpreted (by vendors, anyway) such
that if you actually open the box (and, even worse, plug the device
in), then you can no longer use the DSR to return it.

And the legislation only applies to consumers, not business purchases.

>> 4800x3600 on 17"), I'm likely to use this display as well, until it 
>> dies.
>
> We can dream..

Yeah, I'm waiting for the flexible, printed, high-resolution display.

In the meantime using two relatively cheap TFTs seems reasonable.
There's lots of experience producing them, so presumably there's less
wastage than with less common varieties (unusually high resolution,
maybe even widescreen, though I guess widescreen ones are common
enough now).


Re: Xinerama support

2007-01-22 Thread Seth Arnold
On Mon, Jan 22, 2007 at 09:01:28PM +, Tuomo Valkonen wrote:
> Umm... why do you need to check on IRC an Email all the time? Perhaps

I support internal and external users as part of my job; people expect
(and get :) good response to their questions and problems.

> not being able to concentrate on the work at hand is your problem?

If my job required more dedicated programming, absolutely. In fact,
when I do the bulk of my programming, I bring my laptop to a location
without network access and program free of distractions. (I rarely need
references while programming beyond 'man' in the scratchpad, so the
single display isn't a serious concern when programming. :)

> for year (kicking and hitting helps with dying hardware!), and then got

"Technical tap" :) swearing rarely helps though :(

> this TFT, using an ancient laptop as a serial terminal while the shop 
> took two weeks to deliver it (even though they had it in stock). 

Ugh. Yes, you -can- work in that environment, but that doesn't make it
ideal. :)

> (At least in .fi, it's best to order TFTs online, BTW: the law dictates
> an unconditional 14 day right of return, so if there's a single dead 
> pixels, just return it, no questions asked. Can't do that with IRL 
> shops.) Anyway, unless something really remarkable happens (like 

Wow. Sounds like a civilised place to live. :)

> 4800x3600 on 17"), I'm likely to use this display as well, until it 
> dies.

We can dream..


pgpymsfuRFSnc.pgp
Description: PGP signature


*****SPAM***** Re: Xinerama support

2007-01-22 Thread Tilman Baumann
Spam detection software, running on the system "cbs.krieg.ag", has
identified this incoming email as possible spam.  The original message
has been attached to this so you can view it (if it isn't spam) or label
similar future email.  If you have any questions, see
admin for details.

Content preview:  Besides all economic issues with two power sucking 
  devices versus one and gapless widescreen with one psu. My Company 
  gives me only one crappy 16" TFT with bad resolution (1280x1024). But 
  lucky me, i saved a 19" CRT from the crap yard a while ago. So thats 
  what i have to work with. Period. There is no way i can get a 23" TFT, 
  only because it is the better solution. [...] 

Content analysis details:   (2.1 points, 2.0 required)

 pts rule name  description
 -- --
 0.1 MISSING_HEADERSMissing To: header
 2.0 BAYES_80   BODY: Bayesian spam probability is 80 to 95%
[score: 0.9028]


--- Begin Message ---
Besides all economic issues with two power sucking devices versus one  
and gapless widescreen with one psu.


My Company gives me only one crappy 16" TFT with bad resolution  
(1280x1024). But lucky me, i saved a 19" CRT from the crap yard a  
while ago.

So thats what i have to work with. Period.
There is no way i can get a 23" TFT, only because it is the better  
solution.


I can live with the argument that no one wants to fix/implement  
xinerama, because it's a pain, ugly or simply noone is motivated to  
implement something he does not use.

Or whatsoever. That's a fair argument.

That i can't use it ore someone thinks it's a bad idea does not mean  
i don't want it anymore.

So please don't tell me i don't want it.  ;)

Some extra inches penis enlargement are a great thing for me. Moving  
my eys is much more intuitive and efficient than switching desktops.
One screen for useless distractions like chat and email, and one to  
work on. That is how it worked for me for years. :)



Am 22.01.2007 um 08:34 schrieb Tuomo Valkonen:


On 2007-01-22, Seth Arnold <[EMAIL PROTECTED]> wrote:
Sometimes it is more economical to purchase two lower-resolution  
devices
than one high-resolution device -- at least, that was the case  
five years
ago when I changed from a 1600x1200 fuzzy CRT to two 1280x1024  
LCDs. I
didn't want to give up the screen space, but a 1600x1200 LCD was  
simply

not in my budget at the time.


You don't need that much space. Your field of vision isn't that big at
a "working distance" (as opposed to "monitoring distance") from the
screens, so you could just as well switch a workspace as turn your  
head

to the other display. Like I say in the FAQ, the real reason for
multihead is penis enlargment most of the time.


(I'd even bet money that my two LCDs draw
less power than that crummy old CRT. :)


Yes. My 17" TFT is supposed to draw 33W "typically" (whatefer that  
means),
while CRTs can easily draw 100W and more.  But TFTs may very well  
pollute
and waste energy more during their manufacturing. (There's the  
issue with
dead pixels, and hence scrap panels for one thing. Any real  
statistics on

the proportion of rejected panels?)

--
Tuomo



--- End Message ---


Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Seth Arnold <[EMAIL PROTECTED]> wrote:
> This makes it far easier for me to monitor
> email and IRC (on the left display) while reading reference content (on
> the right display). Swapping workspaces requires erasing the reference
> content when bringing up the malleable content.

Umm... why do you need to check on IRC an Email all the time? Perhaps
not being able to concentrate on the work at hand is your problem?

> Perhaps I've waited long enough to treat myself to a new display with all
> the space I need. :) 

Nah, there's no too long time. I used my previous dislay (17" CRT)
for about 7 years until it died, having been in its dying spasms
for year (kicking and hitting helps with dying hardware!), and then got
this TFT, using an ancient laptop as a serial terminal while the shop 
took two weeks to deliver it (even though they had it in stock). 
(At least in .fi, it's best to order TFTs online, BTW: the law dictates
an unconditional 14 day right of return, so if there's a single dead 
pixels, just return it, no questions asked. Can't do that with IRL 
shops.) Anyway, unless something really remarkable happens (like 
4800x3600 on 17"), I'm likely to use this display as well, until it 
dies.

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Bruce Stephens <[EMAIL PROTECTED]> wrote:
> (One of our concerns at work was heat from CRTs (perhaps bizarrely,
> given that we're in south west London, but we really had been in an
> office where heat was a significant problem, though we aren't, now).)

[And speaking of heat and power consumption, the state of software
suspend in Linux totally sucks, and yet it's one of the most important
features I can think of. It sucks having to leave the computer
on when you don't use it, because it's such a _chore_ to bring it 
and all the programs up again, especially as session management support
sucks too (and there's no protocol it for terminal programs). The 
hardware suspend support of desktop mobos also tends to be poor, PSU fan
still keeping humming and so on. (And I haven't heard of that good
things of the Linux laptop support either.) That's one very good reason
to use Windows: hibernate. Of course, even better would be if the computer
could run on _really_ low power, shutting down all the fans, not using 
the disks except very infrequently, and so on, but still react to ssh 
connection attempts etc. ...). Well, at least with 'athcool' I managed
to make the system run 10C cooler most of the time.]

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Seth Arnold
On Mon, Jan 22, 2007 at 07:34:03AM +, Tuomo Valkonen wrote:
> You don't need that much space. Your field of vision isn't that big at
> a "working distance" (as opposed to "monitoring distance") from the 
> screens, so you could just as well switch a workspace as turn your head
> to the other display.

I quite like having my malleable content on one screen and my reference
content on the other screen. This makes it far easier for me to monitor
email and IRC (on the left display) while reading reference content (on
the right display). Swapping workspaces requires erasing the reference
content when bringing up the malleable content.

This constant erasing and replacing content works against my innate
ability to remember where content is on a page. When I had a single
1600x1200 display, I could fit reference content and malleable content
on screen quite nicely. But it was fuzzy. So I tried 1280x1024 LCD and
found myself swapping content enough that it detracted from my ability to
get work done. So I bought a second LCD and have enjoyed the extra space.

Perhaps I've waited long enough to treat myself to a new display with all
the space I need. :) (Pity my video card can't do more than 1280x1024
DVI output. I _really_ don't want to be in the market for a new video
card too. Blech.)

> Like I say in the FAQ, the real reason for 
> multihead is penis enlargment most of the time.

Perhaps. But some of us find it more useful for reading reference matieral
while working on other stuff than advertising our qualities as mates. :)

Honestly, expect better from your users. :)

The argument for not supporting Xinerama that I can easily appreciate is
what it does to your code. Saying "just don't use it" isn't nearly as
compelling as "it fuglifies the code".


pgp7noXrWpjyn.pgp
Description: PGP signature


Re: Xinerama support

2007-01-22 Thread Bruce Stephens
Tuomo Valkonen <[EMAIL PROTECTED]> writes:

[...]

> Well, there's the issue of bigger displays probably having more
> rejects, but the widescreen display is about least half smaller than
> a second display,

Physically it'll be significantly smaller, but I'd guess the physical
stuff (glass, plastic, PSU, etc.) aren't significant compared to the
cost of the TFT.

But I'm guessing.  I wouldn't want to argue that nobody should choose
a widescreen monitor rather than two squarer ones.  Or a good quality
CRT.  All seem defensible choices, depending on what you have already,
what space you have, whether you're worried about heat.

(One of our concerns at work was heat from CRTs (perhaps bizarrely,
given that we're in south west London, but we really had been in an
office where heat was a significant problem, though we aren't, now).)

[...]



Re: Xinerama support

2007-01-22 Thread Steven N. Severinghaus

Tuomo Valkonen said the following on 2007-01-22 11:13:
> ...^Z works [g]reat.)

It's truly amazing what you can fit into one monitor, isn't?

One monitor
  Multiple virtual consoles (ick!)
Multiple workspaces per console
  Multiple frames per workspace
Multiple tabs with xterms per frame
  Multiple screens per xterm
Multiple vims per screen
  Multiple tabs per vim
Multiple windows per vim tab

Who could possibly need one more "multiple" at the top of that tower? ;)

-Steve (happy with one!)

--
Steven N. Severinghaus
[EMAIL PROTECTED]
http://severinghaus.org/


Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Bruce Stephens <[EMAIL PROTECTED]> wrote:
> but I imagine the costs (environmental and other) even out (physically
> bigger screens probably don't cost much more than smaller ones; it
> probably depends more on the number of pixels).

Yeah, I'd suspect so. It probably isn't easy to produce displays with lots
of tiny pixels, and there are bound to be more defects. Infact, I wonder 
if we'll go back to analogue tech, one day. If it would be possible to 
produce thin pixel-less displays somehow, working like plotters. I don't
know how you'd do that, and it might not work for non-vector graphics, 
such as moving video, but... it's an interesting prospect anyway. Maybe 
nano-machines arranging themselves in instructed shapes... :) (Infact,
the E-paper techs I am aware of are quite "mechanical", and I wonder if 
you could control the ink balls floating freely instead of suspended 
in pixels...)


> Quite true, although I'm not so confident about the relative costs.

Well, there's the issue of bigger displays probably having more rejects,
but the widescreen display is about least half smaller than a second
display, there isn't a second PSU with its lossess and costs, less
casing, and so on.

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Bruce Stephens
Tuomo Valkonen <[EMAIL PROTECTED]> writes:

[...]

> Actually, it's not so much about the size of the screen. I don't find
> 19" 4:3 offers anything over 17" 4:3 at the same resolution, or even
> a slightly bigger resolution (e.g. 1600x1200 vs. 1280x1024... which would
> be more of an enhancement on a 17" screen than between 19" and 17", but not
> much of enhancement in either case, as both resolutions are still so poor). 

Sure, by "size" I really mean in pixels.  I'm constrained by company
policy to use TFTs (we decided that for reasons which probably aren't
so valid any more, so possibly I could argue against that), so that
limits my choices.  I'm disappointed I can't get the same kind of
resolution on a desktop screen that is available on laptops, but I
imagine the costs (environmental and other) even out (physically
bigger screens probably don't cost much more than smaller ones; it
probably depends more on the number of pixels).

> It's more about the shape. There's horizontally a bit of space left in 
> the field of vision after a 4:3 screen, that a widescreen display should
> fill. You can comfortably fit two A4 papers on a widescreen display, 
> wasting less resources than multihead, that normally will have a lot 
> of screen space outside your field of vision.

Quite true, although I'm not so confident about the relative costs.

Certainly I don't feel that much advantage in having two screens---a
single one with a similar number of pixels would do just as well
(especially with ion).  I guess if ion regards my existing two screens
as one big one that won't be too bad; it would be slightly annoying on
the occasions where I want to make some application full-screen,
though.


Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Renee Klawitter <[EMAIL PROTECTED]> wrote:
> Tuomo Valkonen <[EMAIL PROTECTED]> writes:
> That's true, but you do not have to search for the workspace. You can
> put the apps you often have to look at (debugger/browser/manuals) on
> one screen and your "working area" at the other one and do not have to
> type in or cycle through app names ...

Umm.. I just have the documentation and such stuff on one side of the
screen, and the code on the other. Works great. One project per
workspace. 

Debugger? What's that? fprintf is my debugger! (For the occasional
'gdb backtrace' I usually just use the same xterm running the editor, 
^Z:ing back to the shell. The same I use for compiles. Although there
would be space for a small compilation/debugging terminal/frame on the
documentation side of the screen, I've never got used to switching
there. ^Z works reat.)

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Thomas Themel <[EMAIL PROTECTED]> wrote:
> Hi,
> Renee Klawitter ([EMAIL PROTECTED]) wrote on 2007-01-22:
>> Tuomo Valkonen <[EMAIL PROTECTED]> writes:
>> One question: How do you debug ion ? One tft two xservers ?
>
> Not sure whether that's how Tuomo does it, but that sounds like a job
> for Xnest.

Xnest is crap. Firstly, it's buggy and crashes all the time, or causes
programs not to work. Secondly, bindings will overlap, and Xnest can't
go into a grabbed mode where it would receive all the key presses, as
far as I know. VmWare would work better than Xnest! But two servers 
on one TFT is what I use. How'd you even do anything like that with
two servers on separate screens? Two keyboards and mice? What an 
unusable hack.

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Thomas Themel
Hi,
Renee Klawitter ([EMAIL PROTECTED]) wrote on 2007-01-22:
> Tuomo Valkonen <[EMAIL PROTECTED]> writes:
> One question: How do you debug ion ? One tft two xservers ?

Not sure whether that's how Tuomo does it, but that sounds like a job
for Xnest.

ciao,
-- 
[*Thomas  Themel*] I'd say that making microchips, or collecting garbage
[extended contact] for that matter, has much more "meaning" than "raising
[info provided in] consciousness of ethnic heritage" and other such bullshit.
[*message header*]- Anonymous poster on cypherpunks


Re: Xinerama support

2007-01-22 Thread Renee Klawitter
Tuomo Valkonen <[EMAIL PROTECTED]> writes:

>
> You don't need that much space. Your field of vision isn't that big at
> a "working distance" (as opposed to "monitoring distance") from the 
> screens, so you could just as well switch a workspace as turn your head

That's true, but you do not have to search for the workspace. You can
put the apps you often have to look at (debugger/browser/manuals) on
one screen and your "working area" at the other one and do not have to
type in or cycle through app names ...

> to the other display. Like I say in the FAQ, the real reason for 
> multihead is penis enlargment most of the time.

Maybe so, but we could use gnome too, couldn't we. Is just not
convenient enough ...

One question: How do you debug ion ? One tft two xservers ?

/Renee.





-- 
For contacting me: klawitter(at)email(dot)de
--
"The major difference between a thing  that might go wrong and a thing
that cannot  possibly go  wrong is that  when a thing  that cannot
possibly go wrong goes wrong it usually turns out to be impossible
to get at or repair."  Douglas Adams, Mostly Harmless



Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Tuomo Valkonen <[EMAIL PROTECTED]> wrote:
> You can comfortably fit two A4 papers on a widescreen display, 

And should I remind of the floating splits of Ion? I also mostly just
split the screen in half horizontally (Mod1+K S), and then float the
halves (Mod1+M, float...) for e.g. the browser that unfortunately needs
so much space on crappy Web pages, and for workspaces with LaTeX work.

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Tuomo Valkonen
On 2007-01-22, Bruce Stephens <[EMAIL PROTECTED]> wrote:
> So I use two monitors.  Alternatives would be a bigger screen, and
> that would work fine, I expect.  Obviously printing things out, though
> it's handy to have searchable things, and I routinely refer to quite a
> lot of material.  Or switching between workspaces, of course.  It's
> just that much more convenient just to have another monitor.

Actually, it's not so much about the size of the screen. I don't find
19" 4:3 offers anything over 17" 4:3 at the same resolution, or even
a slightly bigger resolution (e.g. 1600x1200 vs. 1280x1024... which would
be more of an enhancement on a 17" screen than between 19" and 17", but not
much of enhancement in either case, as both resolutions are still so poor). 
It's more about the shape. There's horizontally a bit of space left in 
the field of vision after a 4:3 screen, that a widescreen display should
fill. You can comfortably fit two A4 papers on a widescreen display, 
wasting less resources than multihead, that normally will have a lot 
of screen space outside your field of vision.

-- 
Tuomo



Re: Xinerama support

2007-01-22 Thread Arnaud Legrand

Today Bruce Stephens wrote:


What we're doing (well, what I'm doing, most of the time) is no
different to having something printed out for reference while I'm
working.


A friend of mine uses two monitor to have at the same time her emacs
and a high definition preview of her documents when LaTeXing her
articles (she rotates one of her monitors to have a Portrait preview
of the page). As far as I'm concerned, I prefer splitting my screen in
half. I don't know whether we can talk about penis enlargement for her
;) but, although there is definitely some show off in this setting, I
can understand it might be convenient.

Cheers,

Arnaud


Re: Xinerama support

2007-01-22 Thread Bruce Stephens
Tuomo Valkonen <[EMAIL PROTECTED]> writes:

[...]

> You don't need that much space. Your field of vision isn't that big
> at a "working distance" (as opposed to "monitoring distance") from
> the screens, so you could just as well switch a workspace as turn
> your head to the other display. Like I say in the FAQ, the real
> reason for multihead is penis enlargment most of the time.

What we're doing (well, what I'm doing, most of the time) is no
different to having something printed out for reference while I'm
working.

In some cases the reference is some sane reflowable format, so there'd
be no difficulty in fitting everything on one (normal-sized) screen.
But often, it's PDF, which (I find, anyway) doesn't fit readably on a
1280x1024 screen concurrently with an editor.  Same seems true of
quite a bit of web-based documentation.

So I use two monitors.  Alternatives would be a bigger screen, and
that would work fine, I expect.  Obviously printing things out, though
it's handy to have searchable things, and I routinely refer to quite a
lot of material.  Or switching between workspaces, of course.  It's
just that much more convenient just to have another monitor.

[...]



Re: Xinerama support

2007-01-21 Thread Tuomo Valkonen
On 2007-01-22, Seth Arnold <[EMAIL PROTECTED]> wrote:
> Sometimes it is more economical to purchase two lower-resolution devices
> than one high-resolution device -- at least, that was the case five years
> ago when I changed from a 1600x1200 fuzzy CRT to two 1280x1024 LCDs. I
> didn't want to give up the screen space, but a 1600x1200 LCD was simply
> not in my budget at the time. 

You don't need that much space. Your field of vision isn't that big at
a "working distance" (as opposed to "monitoring distance") from the 
screens, so you could just as well switch a workspace as turn your head
to the other display. Like I say in the FAQ, the real reason for 
multihead is penis enlargment most of the time.

> (I'd even bet money that my two LCDs draw
> less power than that crummy old CRT. :)

Yes. My 17" TFT is supposed to draw 33W "typically" (whatefer that means), 
while CRTs can easily draw 100W and more.  But TFTs may very well pollute 
and waste energy more during their manufacturing. (There's the issue with 
dead pixels, and hence scrap panels for one thing. Any real statistics on
the proportion of rejected panels?)

-- 
Tuomo



Re: Xinerama support

2007-01-21 Thread Seth Arnold
On Sun, Jan 21, 2007 at 06:36:52PM +, Tuomo Valkonen wrote:
> On 2007-01-21, Ulrich Pfeifer <[EMAIL PROTECTED]> wrote:
> > You are saying it *should* be easy, but it currently is not - right?
> Yep.

I don't really need my windows to span both displays. I'm perfectly fine
with windows being confined entirely on one screen or the other. But
I do like moving windows between the two screens occasionally. (Only
when starting a new application, and it comes up on the 'wrong' display.
Because I can easily move windows, I've never figured out the winprops
support. Would winprops work for the 'standard X multihead'?)

> Or forget multihead. It's stupid anyway.

Sometimes it is more economical to purchase two lower-resolution devices
than one high-resolution device -- at least, that was the case five years
ago when I changed from a 1600x1200 fuzzy CRT to two 1280x1024 LCDs. I
didn't want to give up the screen space, but a 1600x1200 LCD was simply
not in my budget at the time. (I'd even bet money that my two LCDs draw
less power than that crummy old CRT. :)

In any event, thanks for the fine window manager. I hope to be able
to adapt my work habits to fit the new version at some point in the
future. :)

Thanks


pgpmoHifcksTf.pgp
Description: PGP signature


Re: Xinerama support

2007-01-21 Thread Jarrod Lowe

On 21/01/07, Ulrich Pfeifer <[EMAIL PROTECTED]> wrote:

Hello,

t> Or forget multihead. It's stupid anyway.

it serves me very well and I will stick with the latest development
snapshot for the next years I guess.  I don't think I have the
knowledge to implement the module you sketched.

Ulrich


I agree - I'll also be sticking to the last snapshot with support for
multiple monitors and migrating windows between them (whatever that is
called in X terms) until someone re-adds support. I simply need the
space. It is a pity, but I have no feel for coding graphical
environments (always get my X's and Y's wrong :) so I'll probably be
forced to abandon ion before to long.


--
Jarrod


Re: Xinerama support

2007-01-21 Thread Ulrich Pfeifer
Hello,

t> Or forget multihead. It's stupid anyway.

it serves me very well and I will stick with the latest development
snapshot for the next years I guess.  I don't think I have the
knowledge to implement the module you sketched.

Ulrich


Re: Xinerama support

2007-01-21 Thread Tuomo Valkonen
On 2007-01-21, Renee Klawitter <[EMAIL PROTECTED]> wrote:
> What do you mean by artificially ? Sounds to me as if this could 
> easily be changed ... ?

X doesn't allow "reparenting" windows from one root window to
another. There's no fundamental technical reason why it shouldn't
be possible when both root windows have the same bit depths 
(and hence the same X "Visuals"). There doesn't seem to be
anything in the Xlib API that would prevent that either. Some
X server implementations and Xlib internals might need some
updating, and some crappy programs might also get confused
by allowing this, but issues are not insurmountable, and the
only real reason why it "can't" be done is, is some hallowed
specification saying so.

-- 
Tuomo



Re: Xinerama support

2007-01-21 Thread Tuomo Valkonen
On 2007-01-21, Ulrich Pfeifer <[EMAIL PROTECTED]> wrote:
> You are saying it *should* be easy, but it currently is not - right?

Yep.

> So I either stay with the current ion version or implement the module
> you proposed - right?

Or forget multihead. It's stupid anyway.

-- 
Tuomo



Re: Xinerama support

2007-01-21 Thread Ulrich Pfeifer
Hello Tuomo,

still not sure ...

t> I'm saying that you don't need root windows spanning multiple physical
t> screens. Plain old X multihead with completely separate screens is
t> quite sufficient

But I cannot move applications from the one head I don't see any more
when I leave the office easily to the other head - right?

You are saying it *should* be easy, but it currently is not - right?

So I either stay with the current ion version or implement the module
you proposed - right?

Ulrich Pfeifer


Re: Xinerama support

2007-01-21 Thread Renee Klawitter
Tuomo Valkonen <[EMAIL PROTECTED]> writes:

> of Xinerama. Especially if the migration of sub-windows from one
> root window to another wasn't artificially restricted, as it 

What do you mean by artificially ? Sounds to me as if this could 
easily be changed ... ?

/Renee.


-- 
For contacting me: klawitter(at)email(dot)de
--
"The major difference between a thing  that might go wrong and a thing
that cannot  possibly go  wrong is that  when a thing  that cannot
possibly go wrong goes wrong it usually turns out to be impossible
to get at or repair."  Douglas Adams, Mostly Harmless



Re: Xinerama support

2007-01-21 Thread Tuomo Valkonen
On 2007-01-21, Ulrich Pfeifer <[EMAIL PROTECTED]> wrote:
> Are you saying I do not need Xinerama to achieve this or are you
> saying that I need to implement that module if I ever want to upgrade
> ion?

I'm saying that you don't need root windows spanning multiple physical
screens. Plain old X multihead with completely separate screens is
quite sufficient, and better than the ugly big-ass root window hack
of Xinerama. Especially if the migration of sub-windows from one
root window to another wasn't artificially restricted, as it 
presently is. But a single screen is also very sufficient for 99.99% 
of users.

-- 
Tuomo



Re: Xinerama support

2007-01-21 Thread Arnaud Legrand

Hi,

Today Ulrich Pfeifer wrote:


In my setup I do have a laptop and an external LCD with a different
resolution when I am in the office.  I currently fold my applications
from the monitor to the laptop screen when I leave the office.


As far I'm concerned, I also have a laptop and an external LCD with a
different resolution. I'd thought about a setting like yours but I
finally found out it wasn't really convenient (mainly because I did
not really make use of the laptop monitor). What I do now is simply
cloning the externel LCD to my laptop (thus some parts are not
displayed on the laptop) and when I go home, I just unplug the
externel monitor, use xrandr and restart ion (with ioncore.restart).
It works great.

OK, I have an nvidia card and I use the twinview extension to clone
the external display but I don't think this is actually needed. And
anyway, ion (3ds-20061223) works perfectly fine with it.

Cheers,

Arnaud


Re: Xinerama support

2007-01-21 Thread Ulrich Pfeifer
Hello Tuomo,

Tuomo> You either don't really need the information, or you'd
Tuomo> simply want plain old X multihead with window migration
Tuomo> between screens, something that is only _artificially_
Tuomo> restricted when the bit depths are the same.

I don't know enough about X to fully understand your mail.

In my setup I do have a laptop and an external LCD with a different
resolution when I am in the office.  I currently fold my applications
from the monitor to the laptop screen when I leave the office.

Are you saying I do not need Xinerama to achieve this or are you
saying that I need to implement that module if I ever want to upgrade
ion?

Thanks,

Ulrich

-- Xinerama configuration:
-- Screen   "Screen0"
-- Screen   "Screen1" LeftOf "Screen0"

function go_home ()
   local right_screen = ioncore.find_screen_id(0)
   local left_screen  = ioncore.find_screen_id(1)
   for _, r in pairs(left_screen:managed_list()) do
   right_screen:attach(r)
   end
end

function to_other_screen (r)
   local left_screen  = ioncore.find_screen_id(1)
   local right_screen = ioncore.find_screen_id(0)
   local ws = ioncore.find_manager(r, "WGroupWS")
   local sc = ioncore.find_manager(ws, "WScreen")

   if sc == left_screen then
right_screen:attach(ws)
   else
left_screen:attach(ws)
   end
end


Re: Xinerama support

2007-01-18 Thread Tuomo Valkonen
I should perhaps add that even with a mod_xinerama and _any_
reasonable effort, Xinerama will still have serious problems, at
least with programs that try to position their own floating windows,
and possibly also ones that use crappy methods to look up windows.
The former simply is not doable without updates to the ICCCM to be
aware of nested WM window configurations. Currently the ICCCM does
not specify how applications should handle multiple levels of WM
windows, but also does not disallow it. So one app handles that one
way, and another the other way. 

It is not reasonable to not have the confusing "virtual root windows"
for each physical screen if you want the workspaces on each screen to
be separate for floating windows, the windows that this problem
mostly affects. And if you don't want them to be separate, you don't
really need Xinerama support. Configurations with screens of totally
differing sizes appearing as one are absolutely stupid (but so is
multihead in general, most of the time). Hence, Xinerama is a mostly
useless piece of crap of an extension. You either don't really need 
the information, or you'd simply want plain old X multihead with window
migration between screens, something that is only _artificially_ 
restricted when the bit depths are the same. 

But the general trend among the X folks is to write crappy mutually
incompatible extensions for everything, instead of extending the X
protocols. (There's _no_ need for Xrender, Xft and that shit. It
would've been sufficient to add extra transparency and blurring bits
for programs to set when they can cope with that stuff, and use the
old stuff.)

-- 
Tuomo