XML is a framework for inventing a language, rather than a specific language itself - so, just because all the java apps use XML doesn't mean there's much portability between knowing the config of one app and knowing it of another.
I do a lot of work with XML config files and they are a horribly verbose waste of time. They are usually peppered with namespace declarations etc. which are never variant anyway, and only serve to confuse the user.
The easiest and most useful program I manage configs for is Apache - with a homebrew, but simple and to the point config language.
Just listing the dependencies isn't good enough. More dependencies might be a good thing from a disk space POV if the libraries are more atomic - i.e. do one thing and one thing well.
The KDE libraries tend to bundle a lot of functionality together - smaller number of libraries, but greater individual library size.
Well, I'd like a PPC for the hardware, but I'd get very tired of OS X very quickly. I've borrowed an ibook for one day and that was enough.
Basically, with Debian I can make my machine do whatever I want. In order to use OSX I had to install a third-party virtual desktop manager, which was a bit flaky.
So he found linux HW hard and you found it (relatively) easy. Both arguments may not reflect the situation a normal user is going to experience in the slightest.
To throw my own irrelevant experiences into the mix, all the HW shipped with my PC works with a fresh XP and a fresh Debian/sarge.
.net got GUI development completely right? Please enlighten me, for I am not well versed in the.net stuff, but getting GUI development completely right is something I'd like to understand.
How does.net handle seperating the UI logic from the code? Can you take an application and switch the UI to something different, like a text-based or web-based one, with minimal reworking of the application logic? How are accessibility issues dealt with? How do you unit test the UI?
So some other helpful chap modifies a mozilla patch to make it work for firefox, puts it in the build, and it doesn't solve the problem - hm, did he not test the patch? Why is the patch still in there if it doesn't work? Maybe the article words this funny and I'm mistaken, and I'm not going to read the code to find out, but that sounds a bit strange.
I'm not remotely excited about `true' transparency and drop shadows - features which are no use beyond looking pretty for me. However I can see that in order to get support, which is what an OSS project thrives on, the screenshot-happy users need to be pleased. So in that regard I think such developments are a trade-off.
What significant X developments would impact me? Well, has anyone tried adding a 3rd-party driver to X? I have to download the entire x source (apt-get source xfree86) and stick a diff from aiptektablet into the debian patch directory and build the whole freakin` thing (which, with.o's, clocks in over 300MB) in order to get my graphics tablet to work. Thats an enormous download and compile job when a mere fraction of that code and the result is needed. Streamlining and simplifying this process would allow more people to experiment with x hacks and make our lives easier.
I think hardware support for vector rendering will be a great benefit to how quickly window-manager and toolkit operations are performed - anyone profiled a GTK2 app recently, and seen the slice pango takes up?
Finally there is a lot of innovation going on outside of the x.org project which I think is equally as important as the framework - examples of next-generation window management such as ion and devil's pie show where I think things are moving for power-users.
Thats what workspaces are for. Lets face it, you cannot read both at the same time, what is important is the ease with which you can switch focus. A keyboard shortcut or a flick of the mouse wheel to move quickly between two workspaces satisfies your needs without making your display noisy and headache-inducing.
In addition, there are some routines in there which make it run really slow on old hardware. It demands the full resources of my (450mhz / 512mb) which I find a little extreme for such a simple game.
I used Debian for a long time before I got tired of it and switched back to a source based distro (Gentoo) where you don't have to wait around for people to package and make binaries.
No, now you just wait around for the compile job to finish - solving the same set of equations that all the other gentoo users are independently solving at the same time.
How many times have you really needed a feature that is present in the most cutting-edge versions anyway? I asked myself this question and the answer was `nearly never'.
I think the obsession with running bleeding edge software is some kind of rebellion against the way things were back with windows. However I like to think I'm past rebelling and actually using my computer for something, rather than watching a compile tick by for its own sake.
Of course this is interesting, and shows the talents and ideas that can occur in the world of free/open software.
It also highlights the biggest achilles heel for the open source community: too many of the talented authors waste their time on things like this, when real quality software problems could be tackled instead.
I think the preloading thing is a mix-up. It is rumoured (I haven't personally confirmed this but I have no reason to doubt it) that office is prelinked rather than loaded - i.e. the libraries it depends on have dedicated locations in memory which favour office over other applications (boundaries etc.) and lowers the start-up time due by removing the need to assign VM locations to the dynamic libraries then.
Often at startup we would just load as little of the app we could to render the main frame and then load the actual functional code in the background.
I can't see whats objectionable about this. I'd much rather have bits of mozilla (or whatever) load in the relatively large window of idleness that happens whilst I move the mouse pointer over to the address bar with mammalian slowness, or in the wait for the network to return a response, than up front when I am (unwillingly) idle.
How about a closed source kernel driver and an open source X driver? Or, an open source kernel driver, an open source X driver, and a common, closed-source library in the middle?
Since X is now in quite a volatile state, X.org being considered canonical and the opportunity for radical changes to be made; having to stick to the existing driver interface due to 3rd party vendors having closed source drivers is a major handicap.
The plus points couldn't be implemented into linux any more readily with the source available than now if the licence wasn't GPL.
It would not be a trivial job to do either - they are utterly different implementations of a kernel, I imagine that whatever code could be salvaged would need major massaging.
Can you (or anyone else) highlight some significant advantages of the solaris kernel over linux which should be ported across if it became possible to do so?
Challenge/Response is fundamentally broken. For more information, take a look at some discussions on the topic from debian-user: here's one. There's a few google-harvested discussions on the topic too.
Question: is the current MINIX licence GPL-compatible? I've given it a scan and it seems pretty liberal. Is it possible or feasible for code to be taken from linux (or vice versa) within the remit of these licences?
Hotmail tagging the invites as junk may just be an accident. What about ISPs purposefully blocking all mail from gmail accounts? Nobody here seems to be focusing on that. This seems a much more serious problem.
From the forum mentioned:
"Thank you for contacting us. Unfortunately, due to potential privacy and confidentiallity considerations, we are not accepting inbound email from Google's Gmail service. In order for us to respond to your email, you will need to resend it from a different email account."
But what business of the ISPs is it if I choose to submit to the terms and conditions?
Slashdot Mirror
XML is a framework for inventing a language, rather than a specific language itself - so, just because all the java apps use XML doesn't mean there's much portability between knowing the config of one app and knowing it of another.
I do a lot of work with XML config files and they are a horribly verbose waste of time. They are usually peppered with namespace declarations etc. which are never variant anyway, and only serve to confuse the user.
The easiest and most useful program I manage configs for is Apache - with a homebrew, but simple and to the point config language.
Just listing the dependencies isn't good enough. More dependencies might be a good thing from a disk space POV if the libraries are more atomic - i.e. do one thing and one thing well.
The KDE libraries tend to bundle a lot of functionality together - smaller number of libraries, but greater individual library size.
Well, I'd like a PPC for the hardware, but I'd get very tired of OS X very quickly. I've borrowed an ibook for one day and that was enough.
:)
Basically, with Debian I can make my machine do whatever I want. In order to use OSX I had to install a third-party virtual desktop manager, which was a bit flaky.
With Debian I'd stick ion2 on and be happy
We've got up2date behind a HTTP proxy with no problems. I'd keep at it IIWY
So he found linux HW hard and you found it (relatively) easy. Both arguments may not reflect the situation a normal user is going to experience in the slightest.
To throw my own irrelevant experiences into the mix, all the HW shipped with my PC works with a fresh XP and a fresh Debian/sarge.
but what does it do?
.net got GUI development completely right? Please enlighten me, for I am not well versed in the .net stuff, but getting GUI development completely right is something I'd like to understand.
.net handle seperating the UI logic from the code? Can you take an application and switch the UI to something different, like a text-based or web-based one, with minimal reworking of the application logic? How are accessibility issues dealt with? How do you unit test the UI?
How does
So some other helpful chap modifies a mozilla patch to make it work for firefox, puts it in the build, and it doesn't solve the problem - hm, did he not test the patch? Why is the patch still in there if it doesn't work? Maybe the article words this funny and I'm mistaken, and I'm not going to read the code to find out, but that sounds a bit strange.
I'm not remotely excited about `true' transparency and drop shadows - features which are no use beyond looking pretty for me. However I can see that in order to get support, which is what an OSS project thrives on, the screenshot-happy users need to be pleased. So in that regard I think such developments are a trade-off.
.o's, clocks in over 300MB) in order to get my graphics tablet to work. Thats an enormous download and compile job when a mere fraction of that code and the result is needed. Streamlining and simplifying this process would allow more people to experiment with x hacks and make our lives easier.
What significant X developments would impact me? Well, has anyone tried adding a 3rd-party driver to X? I have to download the entire x source (apt-get source xfree86) and stick a diff from aiptektablet into the debian patch directory and build the whole freakin` thing (which, with
I think hardware support for vector rendering will be a great benefit to how quickly window-manager and toolkit operations are performed - anyone profiled a GTK2 app recently, and seen the slice pango takes up?
Finally there is a lot of innovation going on outside of the x.org project which I think is equally as important as the framework - examples of next-generation window management such as ion and devil's pie show where I think things are moving for power-users.
Thats what workspaces are for. Lets face it, you cannot read both at the same time, what is important is the ease with which you can switch focus. A keyboard shortcut or a flick of the mouse wheel to move quickly between two workspaces satisfies your needs without making your display noisy and headache-inducing.
In addition, there are some routines in there which make it run really slow on old hardware. It demands the full resources of my (450mhz / 512mb) which I find a little extreme for such a simple game.
I used Debian for a long time before I got tired of it and switched back to a source based distro (Gentoo) where you don't have to wait around for people to package and make binaries.
No, now you just wait around for the compile job to finish - solving the same set of equations that all the other gentoo users are independently solving at the same time.
How many times have you really needed a feature that is present in the most cutting-edge versions anyway? I asked myself this question and the answer was `nearly never'.
I think the obsession with running bleeding edge software is some kind of rebellion against the way things were back with windows. However I like to think I'm past rebelling and actually using my computer for something, rather than watching a compile tick by for its own sake.
Of course this is interesting, and shows the talents and ideas that can occur in the world of free/open software.
It also highlights the biggest achilles heel for the open source community: too many of the talented authors waste their time on things like this, when real quality software problems could be tackled instead.
I think the preloading thing is a mix-up. It is rumoured (I haven't personally confirmed this but I have no reason to doubt it) that office is prelinked rather than loaded - i.e. the libraries it depends on have dedicated locations in memory which favour office over other applications (boundaries etc.) and lowers the start-up time due by removing the need to assign VM locations to the dynamic libraries then.
More info at http://www.crast.us/james/articles/prelink.php
Often at startup we would just load as little of the app we could to render the main frame and then load the actual functional code in the background.
I can't see whats objectionable about this. I'd much rather have bits of mozilla (or whatever) load in the relatively large window of idleness that happens whilst I move the mouse pointer over to the address bar with mammalian slowness, or in the wait for the network to return a response, than up front when I am (unwillingly) idle.
how comes Word still starts in under 3 seconds when running on Windows emulation on Linux?
I've never seen this myself - do you know this from personal experience, or are you also propagating an urban myth?
Compression doesn't necessarily mean a drop in quality. There's lossy and non-lossy compression.
http://www.geocities.com/cryect/pistolflashlight.h tml
This, and more via the excellent http://www.doomworld.com/
How about a closed source kernel driver and an open source X driver? Or, an open source kernel driver, an open source X driver, and a common, closed-source library in the middle?
Since X is now in quite a volatile state, X.org being considered canonical and the opportunity for radical changes to be made; having to stick to the existing driver interface due to 3rd party vendors having closed source drivers is a major handicap.
Taskbar grouping is basically the same thing as grouping via tabs, which is very popular in some browsers and window managers.
The plus points couldn't be implemented into linux any more readily with the source available than now if the licence wasn't GPL.
It would not be a trivial job to do either - they are utterly different implementations of a kernel, I imagine that whatever code could be salvaged would need major massaging.
Can you (or anyone else) highlight some significant advantages of the solaris kernel over linux which should be ported across if it became possible to do so?
Challenge/Response is fundamentally broken. For more information, take a look at some discussions on the topic from debian-user: here's one. There's a few google-harvested discussions on the topic too.
Screen is more than a terminal multiplexer. Take a look at the documentation.
Question: is the current MINIX licence GPL-compatible? I've given it a scan and it seems pretty liberal. Is it possible or feasible for code to be taken from linux (or vice versa) within the remit of these licences?
Hotmail tagging the invites as junk may just be an accident. What about ISPs purposefully blocking all mail from gmail accounts? Nobody here seems to be focusing on that. This seems a much more serious problem.
From the forum mentioned:
"Thank you for contacting us. Unfortunately, due to potential privacy and confidentiallity considerations, we are not accepting inbound email from Google's Gmail service. In order for us to respond to your email, you will need to resend it from a different email account."
But what business of the ISPs is it if I choose to submit to the terms and conditions?