Will there be full hardware-acceleration of the graphical effects, like Compiz provides,
Yes. There already is, in fact. Aaron Seigo has prepared a screencast which will be linked with this week's Commit Digest. Among other things, it shows the true transparency of Plasma applets and the panel. Also, you can check out these old (non-Plasma related) kwin_composite videos: http://www.youtube.com/watch?v=4WBLlc6xCQ4
or will KDE 4 continue to have ugly faux transparency?
Eavesdropping on IRC, I get the feeling that aseigo hates this hack more than you do;)
The author is a KDE developer so he naturally favors KDE, but as far as I can tell, there is no obvious unfairness in his analysis, and it is very easy to reproduce the tests yourself.
In fact, the GNOME guys did just this (with a more recent, and thus better-optimized, version of GNOME), and although GNOME fared a little better than it did in Lubos's test, the results were largely the same:
The "KDE is bloated!" meme is just a silly myth that needs to die, in my opinion, and I have no idea why the reality surprises so many people - after all, aren't the Germans noted for their efficiency?:)
Here are a few currently running (UK) eBay auctions that have known-working-under-Linux cards (all are ralink, and I own one of each of these devices, unless some kind of version-bump has changed the chipset!) :
All should Just Work on, say, Ubuntu Dapper onwards (they did for me, at least), but I should note that I have not tried WPA with them. Obviously, eBay auctions are not exactly appropriate for corporate acquisitions, but hopefully this will help you in your search!
I actually love KDE, but the Kopete interface is quite too much cluttered in my opinion. However I sometimes try it and the latest versions are not as bad as the old ones.
For some reason I can't get the zoom stuff or anything involving the "Windows" key working though. Also, I wasn't able to get any video going in real time-- obviously I can't watch a DVD with the CD already in the drive... and streaming wasn't working. Also, I have two monitors but they weren't both detected.
Same here - you can edit this in gconf-editor, though - I think (purely from memory) that it's in Apps->Compiz->Zoom, then you can edit the key to something like Alt plus scrollwheel, like I did.
My computer has a fx5200 - not exactly shitty, but hardly top of the line, either - and it worked like a charm:) I endedup going to the "Moon Song" on www.rathergood.com while watching videos on Apple Trailers and thumbnailing them as a test, and it worked, I must say, absolutely stunningly well - I was amazed at how smooth and fluid it all was.
Didn't work on my old Savage Twister laptop, though, although based on my experiences with Dapper Flight 4, Xorg 7 has DRI support for my Savage card out-of-the-box.
My only complaint is that the LiveCD contained no sample movies or 3D games (like, e.g., Scorched3D or Tuxracer).
You have to unzip it first:). "Zip" files are well-understood and have a wealth of (portable) open-source implementations available. The extracted directories contain files that are either human-readable (the document text, plus styles and formatting) or files that were embedded in the document, kept in their original format (e.g. embed a PNG in your documents, and you can unzip your document and grab your PNG image back with nothing more than an unzipping utility!).
Indeed. Last weekend, I downloaded the FreeCiv source from Ubuntu's repositories, and cross-compiled for ARM. It worked, literally, first-time on my YOPY. Of course, it's going to take a bit of GUI tweaking to fit it onto the screen properly (FreeCiv was really designed to work on 640x480 and above, whereas the YOPY 3700 has a 240x320 screen), but the actual game itself runs fine. I was very pleasantly surprised:)
I always thought that this one deserved the Worst Slashdot Article Ever award. It is, however, responsible for one of my all-time favourite quotes:
I hate Linux. I think it's the most over rated piece of software ever built and survives simply out of spite and not because it is terribly good at doing something because it is not!
It reads like it was written by a petulant 12 year old. Hilarious stuff!
Actually, having now read some of the comments, I retract this statement - one mentioned example about two renderable webpages with the same hash being created blew my mind. Amazing.
There is at least some truth to this, I think: creating a file with not only the same md5 as a given file but also the same file-type (e.g. plain-text if the original is plain-text; a valid.zip file if the original is a.zip file; a valid.iso if the original is a.iso, although I'm not sure on this last example as I don't know how strict the "structure" of a.iso is; etc) is far, far harder. In fact, it could potentially even be the case that there does not even exist a file with both the same hash and the same file-type as the original, or at least, not one that is not so vast as to be either computationally unfeasible, or just plain too darn big to download:)
Great point; I'm tired of people conflating OSS with "hobbiest" programming. While in practice there is often an overlap between the two concepts, they are fundamentally distinct (OSS can be created by a scruffy misfit in his mother's basement, or a high-powered team of crack programmers in IBM's Really Clever Research department) and lumping the two together serves only to muddy the waters of discussion.
You are correct; the previous one was a IDN spoofing vulnerability, which I thought was largely a flaw in the IDN specification itself, rather than in any particular implementation thereof (is this correct...?). This time around, however, the flaw lies in the Firefox code itself.
PDF is a completely open format with a large number of implementations (many of which are open source); for example, xpdf runs on my YOPY. Also, OO.o can export to PDF natively, although the 1.x.y series does not include hyperlinks except via a 3rd party plug-in. LaTeX can create nice PDFs too, if you're feeling hardcore;)
why it is up to openoffice to try to read microsoft's documents.
Because Microsoft Office's formats are the defacto standards, and Microsoft have made no effort to support OO.o's format (despite open standards and even a reference implementation, although the latter would be of limited use to them). It's not really fair, but that's just the way it is - in order for an Office Suite to be of use, it must unfortunately be compatible with Microsoft's (closed) formats.
you know its a free download go get it and install it so you can read the people's documents that are created in openoffice.
Very few people currently use Open Office, and even then a sizeable proportion will probably use it for importing and exporting MS Office-formatted documents. As a side-note, read this: http://en.wikipedia.org/wiki/Network_effect
openoffice uses open standards for file saving and microsoft doesn't - this isn't rocket science people. just run them side by side until you totally switch to open office.
This is fine in a completely sealed environment, but becomes impractical when people external to your company send/ require MS-formatted documents, as is very often the case.
Having goverments not only switch to, but stipulate that any one who deals with them must supply/ be able to receive open formats is probably about the best thing we can hope for.
I purchased a copy of the ANSI C++ Standard last year. I can read it on damn-near anything (it works perfectly under xpdf, so I could probably even use it on my YOPY:)). The only caveat is that my name and address is emblazoned on every page! Personally, I like this method, although of course it means I have to be extra-vigilant about making sure it doesn't fall into the hands of anyone who'd put it up on P2P:)
Many distros even map ctl-alt-esc to launch xkill.
Some even use GNOME-System-Monitor/ Ksysguard which look pretty much identical to Window's Task Manager! This point of the OP's was very, very weak, as was his assertion that "95% of linux hardware drivers require editing a configuration file through a text editor". Of the God-Knows-How-Many internal and external pieces of hardware I own, only one (the graphics card, and then only if I want 3D acceleration) has required me to edit config files. 95% is ridiculous hyperbole and does nothing but detract from his points.
This is actually almost exactly the tack I take nowadays when discussing Linux, due largely to two points:
a) If you over-sell Linux and imply (as some people had done to me) that it can do damn-near everything Windows can do, and is better in virtually every way, you are setting them up for a disappointment that will lead to such a bitter backlash that they will probably actively despise Linux for a long period of time. This is what happened to me a couple of years ago - people sold Linux to me as if it were the Second Coming, and when I tried it the gulf between expectation and reality made me turn against big time. Thankfully I tried it again (pretty much by force of circumstance, rather than choice; as far as I had been concerned, based on my initial disappointment, I was done with it forever) and saw the light a year or so later:)
b) A measured, educated discussion of its good and bad points will sway people far more, especially if you concentrate on the bad points. I've seen this happen time and time again, and the best theory I can come up with is that, if the person respects your judgement and doesn't just think you are an idiot zealot (the "measured, educated discussion" can help dispell this notion:)), then listing a bunch of its flaws but still continuing to use it will make people curious about the good qualities that keep you using it, and make them want to try it themselves. I suppose it's like a weird kind of reverse-psychology, but that doesn't quite hit the nail on the head, I don't think.
Of course Linux too has to support the universe of hardware configurations Windows has too, but this is only possible because anyone can compile the drivers to fix them on demand, something only super users are willing to do.
This is correct up to a point, but it's worth noting that mature (and Free!) Linux drivers usually end up in the official kernel. Probably even more drivers are added by distro packagers. As a result, each successive distro release supports more and more drivers out of the box - and distro releases are pretty frequent.
For less well-supported hardware, though, you are right, and there's no way I'd leave my mum to do something like compile a driver for any of her hardware:)
Having to "recompile a kernel" does not constitute an "option", in my book:) Also, modern Desktop Linux distributions definitely do aspire to be user-friendly; it's just that adding the nebulous quality that is "user-friendlinesss" is an immensely tricky task.
There are several pairs of goals that preclude or inhibit the other ("security" vs "convenience" is probably one), but I don't think "user-friendliness" and "options and flexibilty" need necessarily be in opposition in any way, especially with a system as open from kernel to DE's as Linux is. Give me a system that can install easily on a wide-range of hardware (and Linux is making great strides in this area - the near-random bundling of parts that comprises my desktop PC requires far less effort to configure and set-up under Linux than it does under Windows, though other people's mileage will vary tremendously) but still give me the ability to tweak the source (or just give me lots of configuration options - I'm a KDE fan, if you haven't guessed) and I'll be happy as a clam. Give me the latter without the former, and all you've done is make my life unnecessarily difficult.
Has there been much progress in the last few years in supporting winmodems under Linux?
Not really, which has always confused me a little bit: one would expect the hardware interface to a WinModem to be very simple, and I'm sure there are libraries that perform the kind of "software modem" processing which could be coupled to a very small driver. Can someone in the know let me know where I'm going wrong? I know that there are drivers for the Lucent WinModem on my mum's IBM T22, but they are closed source with a GPL outer-coating, and I'm wondering if they will continue to work when the jump to GCC 4.0 is made...
Anyway, eBay is a good place for modems - I managed to snag one of those 56k/Ethernet PCMCIA cards for about £8 (postage included) a while back. They are very thin and elegant, and "just work". I'd be surprised if external/ PCI cards are any more expensive than this.
"Proper" NTFS writing (i.e. working out how the NTFS file system is structured, and writing a reliable open-source driver to manipulate it based on this information) is probably a long way off - the current "writing" is basically limited to over-writing non-disjointed files with files of equal or lesser length, which is really not that useful. The method you linked to basically just employs Window's own driver to do the dirty work, so is a black-box solution which may or may not even be legal (very probably not, if you don't have a legal Windows install, but IANAL). I haven't tried it for almost couple of years, now, but at the time it was slow - like "200k/s writing speed", slow. It's doubtless improved since then, though.
I think both the major desktops are beginning to start the optimisation drive round about now; the thing is that software development usually proceeds in fits and starts, with different parts of the software development process going through the phases of Making It Work, Making It Work Well and Securely, and, finally, Making It Work Fast. I'm going to stick my neck out and say that even though both Desktop Environments have been around before the year 2000, both have undergone more intensive development, or at least had more features added, over the last year or so than ever before in their history (actually, this goes for Desktop Linux in general, from the kernel to X to the toolkits to the DE's to the distros themselves), so there are a lot of rough and unoptimised new additions in there.
Fortunately, unlike a certain other purveyor of Desktop OS's, the devs are actually fairly committed to making everything faster and less resource hungry (witness the GNOME optimisation bounties, and the efforts of the Ubuntu team). Robert Love gave a very interesting talk on optimisation of the desktop environments (I can't find a link right now, but the talk was called "Optimising GNOME", although some of the library-level changes could be conscripted by KDE and anyone else, really). KDE posted some resource-consumption figures for the (very rough and unoptimised) KDE4 port of Kate, and it already looks significantly better. Add in the upcoming xgl et al, and things should hopefully get to the absolutely perfect state of getting faster and faster while still adding features that every developer yearns for:)
Of course, it's pretty much impossible to continuously increase functionality without paying some price in terms of resource-consumption, so you might be better off going to less featureful DE's like, say, XFCE, if you prefer speed over functionality.
Wish I had mod-points (and hadn't posted in this topic) - I think this is a great suggestion, and something that would be extremely interesting to see.
Slashdot Mirror
Yes. There already is, in fact. Aaron Seigo has prepared a screencast which will be linked with this week's Commit Digest. Among other things, it shows the true transparency of Plasma applets and the panel. Also, you can check out these old (non-Plasma related) kwin_composite videos: http://www.youtube.com/watch?v=4WBLlc6xCQ4
Eavesdropping on IRC, I get the feeling that aseigo hates this hack more than you do ;)
http://spooky-possum.org/cgi-bin/pyblosxom.cgi/kd
The "KDE is bloated!" meme is just a silly myth that needs to die, in my opinion, and I have no idea why the reality surprises so many people - after all, aren't the Germans noted for their efficiency?
Here are a few currently running (UK) eBay auctions that have known-working-under-Linux cards (all are ralink, and I own one of each of these devices, unless some kind of version-bump has changed the chipset!) :
USB
PCI
PCMCIA
All should Just Work on, say, Ubuntu Dapper onwards (they did for me, at least), but I should note that I have not tried WPA with them. Obviously, eBay auctions are not exactly appropriate for corporate acquisitions, but hopefully this will help you in your search!
http://etotheipiplusone.com/kopete-bland.png
Please excuse crappy WinDeco
--Si.
My computer has a fx5200 - not exactly shitty, but hardly top of the line, either - and it worked like a charm :) I endedup going to the "Moon Song" on www.rathergood.com while watching videos on Apple Trailers and thumbnailing them as a test, and it worked, I must say, absolutely stunningly well - I was amazed at how smooth and fluid it all was.
Didn't work on my old Savage Twister laptop, though, although based on my experiences with Dapper Flight 4, Xorg 7 has DRI support for my Savage card out-of-the-box.
My only complaint is that the LiveCD contained no sample movies or 3D games (like, e.g., Scorched3D or Tuxracer).
You have to unzip it first :). "Zip" files are well-understood and have a wealth of (portable) open-source implementations available. The extracted directories contain files that are either human-readable (the document text, plus styles and formatting) or files that were embedded in the document, kept in their original format (e.g. embed a PNG in your documents, and you can unzip your document and grab your PNG image back with nothing more than an unzipping utility!).
Indeed. Last weekend, I downloaded the FreeCiv source from Ubuntu's repositories, and cross-compiled for ARM. It worked, literally, first-time on my YOPY. Of course, it's going to take a bit of GUI tweaking to fit it onto the screen properly (FreeCiv was really designed to work on 640x480 and above, whereas the YOPY 3700 has a 240x320 screen), but the actual game itself runs fine. I was very pleasantly surprised :)
Actually, having now read some of the comments, I retract this statement - one mentioned example about two renderable webpages with the same hash being created blew my mind. Amazing.
There is at least some truth to this, I think: creating a file with not only the same md5 as a given file but also the same file-type (e.g. plain-text if the original is plain-text; a valid .zip file if the original is a .zip file; a valid .iso if the original is a .iso, although I'm not sure on this last example as I don't know how strict the "structure" of a .iso is; etc) is far, far harder. In fact, it could potentially even be the case that there does not even exist a file with both the same hash and the same file-type as the original, or at least, not one that is not so vast as to be either computationally unfeasible, or just plain too darn big to download :)
Great point; I'm tired of people conflating OSS with "hobbiest" programming. While in practice there is often an overlap between the two concepts, they are fundamentally distinct (OSS can be created by a scruffy misfit in his mother's basement, or a high-powered team of crack programmers in IBM's Really Clever Research department) and lumping the two together serves only to muddy the waters of discussion.
You are correct; the previous one was a IDN spoofing vulnerability, which I thought was largely a flaw in the IDN specification itself, rather than in any particular implementation thereof (is this correct...?). This time around, however, the flaw lies in the Firefox code itself.
PDF is a completely open format with a large number of implementations (many of which are open source); for example, xpdf runs on my YOPY. Also, OO.o can export to PDF natively, although the 1.x.y series does not include hyperlinks except via a 3rd party plug-in. LaTeX can create nice PDFs too, if you're feeling hardcore ;)
Having goverments not only switch to, but stipulate that any one who deals with them must supply/ be able to receive open formats is probably about the best thing we can hope for.
I purchased a copy of the ANSI C++ Standard last year. I can read it on damn-near anything (it works perfectly under xpdf, so I could probably even use it on my YOPY :)). The only caveat is that my name and address is emblazoned on every page! Personally, I like this method, although of course it means I have to be extra-vigilant about making sure it doesn't fall into the hands of anyone who'd put it up on P2P :)
a) If you over-sell Linux and imply (as some people had done to me) that it can do damn-near everything Windows can do, and is better in virtually every way, you are setting them up for a disappointment that will lead to such a bitter backlash that they will probably actively despise Linux for a long period of time. This is what happened to me a couple of years ago - people sold Linux to me as if it were the Second Coming, and when I tried it the gulf between expectation and reality made me turn against big time. Thankfully I tried it again (pretty much by force of circumstance, rather than choice; as far as I had been concerned, based on my initial disappointment, I was done with it forever) and saw the light a year or so later :)
b) A measured, educated discussion of its good and bad points will sway people far more, especially if you concentrate on the bad points. I've seen this happen time and time again, and the best theory I can come up with is that, if the person respects your judgement and doesn't just think you are an idiot zealot (the "measured, educated discussion" can help dispell this notion :)), then listing a bunch of its flaws but still continuing to use it will make people curious about the good qualities that keep you using it, and make them want to try it themselves. I suppose it's like a weird kind of reverse-psychology, but that doesn't quite hit the nail on the head, I don't think.
For less well-supported hardware, though, you are right, and there's no way I'd leave my mum to do something like compile a driver for any of her hardware :)
There are several pairs of goals that preclude or inhibit the other ("security" vs "convenience" is probably one), but I don't think "user-friendliness" and "options and flexibilty" need necessarily be in opposition in any way, especially with a system as open from kernel to DE's as Linux is. Give me a system that can install easily on a wide-range of hardware (and Linux is making great strides in this area - the near-random bundling of parts that comprises my desktop PC requires far less effort to configure and set-up under Linux than it does under Windows, though other people's mileage will vary tremendously) but still give me the ability to tweak the source (or just give me lots of configuration options - I'm a KDE fan, if you haven't guessed) and I'll be happy as a clam. Give me the latter without the former, and all you've done is make my life unnecessarily difficult.
Anyway, eBay is a good place for modems - I managed to snag one of those 56k/Ethernet PCMCIA cards for about £8 (postage included) a while back. They are very thin and elegant, and "just work". I'd be surprised if external/ PCI cards are any more expensive than this.
"Proper" NTFS writing (i.e. working out how the NTFS file system is structured, and writing a reliable open-source driver to manipulate it based on this information) is probably a long way off - the current "writing" is basically limited to over-writing non-disjointed files with files of equal or lesser length, which is really not that useful. The method you linked to basically just employs Window's own driver to do the dirty work, so is a black-box solution which may or may not even be legal (very probably not, if you don't have a legal Windows install, but IANAL). I haven't tried it for almost couple of years, now, but at the time it was slow - like "200k/s writing speed", slow. It's doubtless improved since then, though.
Fortunately, unlike a certain other purveyor of Desktop OS's, the devs are actually fairly committed to making everything faster and less resource hungry (witness the GNOME optimisation bounties, and the efforts of the Ubuntu team). Robert Love gave a very interesting talk on optimisation of the desktop environments (I can't find a link right now, but the talk was called "Optimising GNOME", although some of the library-level changes could be conscripted by KDE and anyone else, really). KDE posted some resource-consumption figures for the (very rough and unoptimised) KDE4 port of Kate, and it already looks significantly better. Add in the upcoming xgl et al, and things should hopefully get to the absolutely perfect state of getting faster and faster while still adding features that every developer yearns for :)
Of course, it's pretty much impossible to continuously increase functionality without paying some price in terms of resource-consumption, so you might be better off going to less featureful DE's like, say, XFCE, if you prefer speed over functionality.
Wish I had mod-points (and hadn't posted in this topic) - I think this is a great suggestion, and something that would be extremely interesting to see.
If you went to the site and manually downloaded, yes; if you used the auto-update, no.
s/work/drop pants/g