I wouldn't say that the large number of windowmanagers bother me. Most of them manage to do what I need (move windows) and since most of what I do is inside the windows, there are few wm features that could improve my productivity and hence, I tend to go with whatever windowmanager that runs when I log onto a newly installed dist.
If there is anything that would increase my productivity, it would be (in no particular order):
1. Fewer toolkits with more standardized keybindings. This is happening so I'm not all that bothered. GTK+ and Qt are both fine though I would be happy if they for some reason would decided to use the same keybindings for standard events (which seems unlikely at the moment).
2. No-fuss international character set support. I'm Swedish, so naturally I want to use Swedish characters. The problem has actually gotten worse since modern dist-makers started to include support for UTF-8, as many common apps doesn't support UTF-8 yet.
No I think it's ment to be used the other way around. This way, I can mount my UN*X school account that allows shell access on my Linux computer at home (where you usually have root access)./S
I may have a few quibbles with the Microsoft GUI but two things that I seriously lack in Linux is consistent behaviour for copy/paste commands and working (as in "no configuration needed") keymappings/character encoding for my national characters in _all_ applications. This has never been a problem in any windows version (counting all the way from 3.0).
Why doesn't anybody mimic that?
Music industry is dying, gaming is next I suspect
on
More on 64-bit Gaming
·
· Score: 1, Troll
I think the gaming industry is in for a nasty surprise, just like the one the music industry is having now. I see companies pumping more and more money into their games. Many games have enormous budgets already and here (at least in the GameSpy article) they are talking about the "advantages" of having loads of memory to make bigger and more complex games. If a game's gona use up all the memory available to a 64-bit CPU, it's gona have to be huge, it's gona take forever to develop and it's gona cost huge ammounts of money, i.e. it better be a success and sell to everybody. They'll probably be marketed down our throats, sporting few or no new ideas and promoted along with the latest lame engineered pop idol music album by enormous mega-media conglomerates seeking to make profit from their projects with budgets rivaling that of many countries.
I study control theory at a university in Sweden and there we use MATLAB for a lot. When the speed of MATLAB routines doesn't suffice, we use MATLAB C-extensions (mex-files) and some commercial Fortran-based packages (e.g. NPSOL).
Prototyping is very often done in MATLAB, where the built in libraries and matrix manipulation capabilities take you a long way before you have to do anything complicated on your own. Simulink (a MATLAB tool) useful for creating models of dynamic systems but I hear that Modelica is also used sometimes.
... even if the entire music _industry_ goes bankrupt and dies, people will still be creative and make new music. I might not even be such a bad idea if the CD-labels did go down, as it might make way for something new (or something old, like the already mentioned music download services).
Yes, writing software for a static plattform is much easier than for a ever changing one like the PC but the example brought up was Morrowind for the Xbox. The Xbox is a static plattform but the game is apparently bug-riddled anyway.
Somehow, even though I never run any strange ad-ons to my Windows installation and that I run on known brands hardware only, I still get a lot of crashes from some PC games (Daggerfall, the prequel to Morrowind to name one). On the other hand, some games run very solidly, even without patches (pretty much all games from Blizzard to name a few). My conclusion therefore must be that there are many low-quality (bug-wise at least) games for the PC that could have been much better, given some effort. Interestingly, the problem seems to have grown worse since people began using the internet and was able to download patches, something unheard of in the early days of computer gaming.
So, in conclusion, I still say that PC software is plagued by bad quality culture. Some companies do good and should have credit for that, but on the whole, the situation is unacceptable. I agree that console developers have a somewhat easier task given more standardized hardware. But even so, many games are ported between diffrent consoles seemingly without any quality reduction, while many ports from the PC to the consoles seem to retain their low quality.
I think what we see here is a bit of PC software culture beeing pushed into the console arena. I've been a console & PC gamer for the last 20 years and own most of the major consoles. I have yet to see a game on a non-Microsoft platform crash.
I stopped buying PC games because I got tired of saving every 2 minutes so I would not loose anything if the game crashed. I think console games are as bug free as they are because people can't patch and wouldn't accept a crashing game.
Bottom line is: if you want software to get better, don't buy crap with lots of bugs. If Microsoft manages to migrate their "Buy-crap-and-wait-for-the-upgrade" attitude to the console world, I guess the only thing to do is to go back to playing tabletop chess...
Nothing wrong with abstraction and my point is not the speed of the application, but to constantly rewrite an application just 'coz you thought of a new way to abstract something will get you nowhere. In my experience, it's better to abstract a little each release. You never really know for sure where you have to go next time and if you're unlucky, all that extra abstraction will just end up in the way. On account of the imon-issue, last time when somebody asked Raster about that and I was around he pretty clearly stated that he didn't care about other kernels and that if anyone had an interest in using efm on another OS than Linux, they could port Enlightenment themselves. This might have changed since then, which would be about a year ago.
As for the politics thing, that might also have changed since I checked the e-devel list last so I'll drop the attitude point. That still doesn't change the fact that the application they're creating is somewhat radical in many ways (not saying it's bad) and that many design decisions are accordingly radical and as such easy to be disliked by someone from a diffrent school of development. For instance, I wouldn't be caught dead with creating a unix application that used non-text config files (which was the direction Raster was heading a couple of re-writes ago atleast), not because binary files don't have any advantages but because text config files are somewhat of a tradition (a working one at that) in the unix environment and breaking that tradition doesn't buy me all that much in the end.
I can see several reasons as to why E17 is taking so long. I briefly considered taking part of the effort since E16 was my wm of choice for a long time, but I gave it all up. A slightly condensed list of why:s include:
Suffering from Not-invented-here syndrome. The E team spends a _lot_ of time re-implementing things that already exist. They don't even make use of eachother's code as much as they could.
This of course takes lots and lots of time.
Abstraction mania. Imlib2, Evas, Ebits, Etox... the list of abstraction layers grows longer for each time I check out Raster's website. I'm not sure why all these layers need to exist since E17 (actually it's just Efsd I think... at the moment) need to be run on top of a (patched) Linux kernel to work. It's not like they're aiming for portability. I wonder when they're gona be content with what tools they have and just finish the damned thing, as I presume there will be an E18 eventually.
Politics. Many don't agree very much with a lot of the design decisions made by the core E team and they're not always very gracious when telling you they don't like your suggestions either. It's of course Raster's god given right to do as he pleases with his code, but it's a strategy that doesn't really attract a lot of contributors.
Additionally, I'd like to agree with several other posts to this discussion: most people don't care if one wm/desktopshell is 5%, 10% or even 50% faster than the other (we're still talking about fractions of seconds here) or if their buttons are semi-transparent and anti-aliased. They care about applications, and last time I looked, "most people" are going to be the ones that decide what the Next Big Thing will be.
I got the notion that maybe women are not so interested in the technology designed by men. Possibly women would find technology more interesting if more of it was designed by women.
Slashdot Mirror
I wouldn't say that the large number of windowmanagers bother me. Most of them manage to do what I need (move windows) and since most of what I do is inside the windows, there are few wm features that could improve my productivity and hence, I tend to go with whatever windowmanager that runs when I log onto a newly installed dist.
If there is anything that would increase my productivity, it would be (in no particular order):
1. Fewer toolkits with more standardized keybindings. This is happening so I'm not all that bothered. GTK+ and Qt are both fine though I would be happy if they for some reason would decided to use the same keybindings for standard events (which seems unlikely at the moment).
2. No-fuss international character set support. I'm Swedish, so naturally I want to use Swedish characters. The problem has actually gotten worse since modern dist-makers started to include support for UTF-8, as many common apps doesn't support UTF-8 yet.
3. backspace/del.
Don't get me started.
Oookay... rant mode off.
No I think it's ment to be used the other way around. This way, I can mount my UN*X school account that allows shell access on my Linux computer at home (where you usually have root access). /S
I may have a few quibbles with the Microsoft GUI but two things that I seriously lack in Linux is consistent behaviour for copy/paste commands and working (as in "no configuration needed") keymappings/character encoding for my national characters in _all_ applications. This has never been a problem in any windows version (counting all the way from 3.0).
Why doesn't anybody mimic that?
I think the gaming industry is in for a nasty surprise, just like the one the music industry is having now. I see companies pumping more and more money into their games. Many games have enormous budgets already and here (at least in the GameSpy article) they are talking about the "advantages" of having loads of memory to make bigger and more complex games. If a game's gona use up all the memory available to a 64-bit CPU, it's gona have to be huge, it's gona take forever to develop and it's gona cost huge ammounts of money, i.e. it better be a success and sell to everybody. They'll probably be marketed down our throats, sporting few or no new ideas and promoted along with the latest lame engineered pop idol music album by enormous mega-media conglomerates seeking to make profit from their projects with budgets rivaling that of many countries.
64-bit gaming? Bah!
I study control theory at a university in Sweden and there we use MATLAB for a lot. When the speed of MATLAB routines doesn't suffice, we use MATLAB C-extensions (mex-files) and some commercial Fortran-based packages (e.g. NPSOL). Prototyping is very often done in MATLAB, where the built in libraries and matrix manipulation capabilities take you a long way before you have to do anything complicated on your own. Simulink (a MATLAB tool) useful for creating models of dynamic systems but I hear that Modelica is also used sometimes.
... even if the entire music _industry_ goes bankrupt and dies, people will still be creative and make new music. I might not even be such a bad idea if the CD-labels did go down, as it might make way for something new (or something old, like the already mentioned music download services).
Yes, writing software for a static plattform is much easier than for a ever changing one like the PC but the example brought up was Morrowind for the Xbox. The Xbox is a static plattform but the game is apparently bug-riddled anyway.
Somehow, even though I never run any strange ad-ons to my Windows installation and that I run on known brands hardware only, I still get a lot of crashes from some PC games (Daggerfall, the prequel to Morrowind to name one). On the other hand, some games run very solidly, even without patches (pretty much all games from Blizzard to name a few). My conclusion therefore must be that there are many low-quality (bug-wise at least) games for the PC that could have been much better, given some effort. Interestingly, the problem seems to have grown worse since people began using the internet and was able to download patches, something unheard of in the early days of computer gaming.
So, in conclusion, I still say that PC software is plagued by bad quality culture. Some companies do good and should have credit for that, but on the whole, the situation is unacceptable. I agree that console developers have a somewhat easier task given more standardized hardware. But even so, many games are ported between diffrent consoles seemingly without any quality reduction, while many ports from the PC to the consoles seem to retain their low quality.
I think what we see here is a bit of PC software culture beeing pushed into the console arena. I've been a console & PC gamer for the last 20 years and own most of the major consoles. I have yet to see a game on a non-Microsoft platform crash.
I stopped buying PC games because I got tired of saving every 2 minutes so I would not loose anything if the game crashed. I think console games are as bug free as they are because people can't patch and wouldn't accept a crashing game.
Bottom line is: if you want software to get better, don't buy crap with lots of bugs. If Microsoft manages to migrate their "Buy-crap-and-wait-for-the-upgrade" attitude to the console world, I guess the only thing to do is to go back to playing tabletop chess...
Nothing wrong with abstraction and my point is not the speed of the application, but to constantly rewrite an application just 'coz you thought of a new way to abstract something will get you nowhere. In my experience, it's better to abstract a little each release. You never really know for sure where you have to go next time and if you're unlucky, all that extra abstraction will just end up in the way. On account of the imon-issue, last time when somebody asked Raster about that and I was around he pretty clearly stated that he didn't care about other kernels and that if anyone had an interest in using efm on another OS than Linux, they could port Enlightenment themselves. This might have changed since then, which would be about a year ago.
As for the politics thing, that might also have changed since I checked the e-devel list last so I'll drop the attitude point. That still doesn't change the fact that the application they're creating is somewhat radical in many ways (not saying it's bad) and that many design decisions are accordingly radical and as such easy to be disliked by someone from a diffrent school of development. For instance, I wouldn't be caught dead with creating a unix application that used non-text config files (which was the direction Raster was heading a couple of re-writes ago atleast), not because binary files don't have any advantages but because text config files are somewhat of a tradition (a working one at that) in the unix environment and breaking that tradition doesn't buy me all that much in the end.
I can see several reasons as to why E17 is taking so long. I briefly considered taking part of the effort since E16 was my wm of choice for a long time, but I gave it all up. A slightly condensed list of why:s include:
This of course takes lots and lots of time.
Additionally, I'd like to agree with several other posts to this discussion: most people don't care if one wm/desktopshell is 5%, 10% or even 50% faster than the other (we're still talking about fractions of seconds here) or if their buttons are semi-transparent and anti-aliased. They care about applications, and last time I looked, "most people" are going to be the ones that decide what the Next Big Thing will be.
I got the notion that maybe women are not so interested in the technology designed by men. Possibly women would find technology more interesting if more of it was designed by women.