Why would he give a 1000 word description of everything he did when two sentences works fine? And why do you (or any one else) need to know that stuff?
If your friend treats you to dinner because he 'got lucky in the stock market,' do you ask for his bank-book and a record of his portfolio?
Really? That's strange because I am having a hell of a time playing back video on my 286. Where is the lighening fast program that will play it at the same speed I do now?
So, in other words, you're not running the same programs faster; you're running bigger, fancier programs that do things you couldn't do before.
Is that what he said? It sounded to me like he was asking if Sybase is significantly better. If the performance/features are the same, and you can get the source to one (not to mention use it on multiple processors or with databases large than 5 GB), then then why not use the latter? You get more. Was he demanding that Sybase open source their stuff? No.
But, by all means, let's all jump down his throat for asking if and what significant advantages it has over OSS solutions.
That's not really the same thing. It doesn't do the smooth image scaling and such. In its previous form, it just took screenshots, slowly, and eventually showed a picture of them all.
However, if you want skippy to work better, you need to get skippyXD, which uses the XDamage extension, new in 6.8. That allows for live updates of windows while they're being displayed by skippy.
In other words, you can't do it well without things like composite and damage.
According to the text of the article, they switched the default theme to winter, so it looks more like this by default. Those screenshots on the E website are (if you look close) from E 16.0, and this is 16.7.
Also, I don't know what you heard, but E16 is just a window manager, like Fluxbox or KWin or Metacity. It isn't an never was trying to be a whole desktop environment. In fact, it used to be the default window manager for Gnome before Sawfish replaced it.
In other words, if you want panels and desktop icons and stuff, then you need to run Gnome or KDE with Enlightenment as the window manager, or you need to use iDesk or something like that to provide that extra functionality. E by itself is closer to the minimalist window managers.
E17 will be more like Gnome, KDE or XFCE, but that's been years in the making and may yet be years before it's released. But E16 was never trying to be like that. What you're doing is sort of like complaining that Fluxbox doesn't do everything that KDE does. E isn't designed to do fancy stuff out of the box and be GUI configurable in all aspects. That's what KDE and Gnome are for.
I don't have a Gmail account either, so I'll speculate too. I don't see why you couldn't take your sorting methods, and instead of outputting a web page, just output some XML or some other information on the threading structure of that mail box. Then you can display it as a web page, or as a folder in Kmail or whatever. You might need to store your mail in some Gmail database or whatever, but it could be separated. Who's to say they don't already do it this way?
The problem with your example is that the interface is simple enough to not be much of an issue. It's just a list of e-mails sorted some way, and probably a sidebar or something. All the interesting stuff is in the sorting and searching algorithms, and the interface is little more than dumping the output to the screen.
So in that case, it's easy to outsource the gui, but it's hard to make/design the backend. That doesn't mean the backend is inherently part of the interface.
If you have a bunch of data points, you can display that data as a graph or a chart or a table. But the underlying data is the same. In my mind, your example is similar. You have a bunch of e-mails. You can sort and relate those e-mails however you want without actually displaying anything. But if you want to work with them, you need an interface. That can be a text-based list or a gui with a bunch of different panels, or a bunch of 3d rendered stacks of boxes. The underlying data is the same. Interface design is about making sure the presentation of that data is effective for the user.
Maybe sorting e-mails is a bit of a crossover, but I still think it falls more on the functionality side. Sure, you can have visionaries design the algorithms and have a monkey code them. But you can have visionaries design user interfaces and then have a monkey code those as well. What I and the original poster are saying is that the ui visionaries should work on ui stuff, and the functionality visionaries should work on the functionality stuff.
I don't think you understood what the original poster said.
He didn't say, "Usability is for losers. People should learn overly complex interfaces." What he said is that you should separate 'what it does' from 'how it looks.'
Take k3b, for example. People usually say it's the best burner for Linux. I just click on a button to open a project, and then drag the files I want to burn onto a window. Or I can drag songs directly out of my media player, juk. Then I click a burn button and it makes the cd.
But, does k3b implement its own cd burning functionality? No, it uses cdrecord and cdrdao and whatever other command line utilities are out there. People who know what they're doing write the cdr writing stuff, and then the k3b people use those to make a nice interface for it.
Take gift as another example. It runs a file sharing client as a daemon, and then you use various gui programs to interact with the daemon. Don't like KDE? Use the Gnome interface. Or vice versa. Don't like any of the interfaces? Make up some totally new filesharing user interface. But underneath it all is the same piece of code providing the actual functionality, written by people who know how to implement it well.
When functionality and interface are separate, you can choose your interface. People can develop new, better interfaces while preserving the existing functionality, and without having to rewrite it, and you don't force people who aren't good at interface design to make all of the decisions for it.
That's what the original poster was saying. Not "usability doesn't matter."
While the point about not straight-cloning an interface may be a good one, what do you suggest?
"OSS should stop cloning MS and do something completely different, only it should be more usable." That's great, do you have any visionary ideas? How many ways are there to make a word processor or e-mail client work?
As for your window buttons example, you can put them wherever you want on the window. By default they're where they are on Windows presumably to take advantage of user familiarity. You can have as many panels as you want and put them wherever you want, and have buttons or start menus or whatever. You can even remove all desktop icons and have a right-click menu like fluxbox in KDE. Or you can turn off the desktop all together.
So, is your suggestion that the default setup should be arbitrarily different from MS? Or should people magically come up with totally revolutionary ideas? Then the projects will get flamed because "it's too hard for people to learn to use."
This is exactly what holds Slashdot back. Arrogant people spouting the same old stuff labeled as 'insightful.'
The original poster complained about some stuff. The reply pointed out some places where his complaints are inconsistant with reality (i.e. you don't need to recompile the kernel for nVidia drivers).
This person is not necessarily representative of the Linux community. Yes, he came off elitist. However, we're also reading Slashdot, where every elitist nerd comes to post his tripe. There are plenty of friendly people in the Linux community ready to help out newbies. Slashdot is not the place they hang out to do tech support, though.
Also, the "not worthy of my time" type comment was aimed at pointing out the other arguments the original poster made that aren't consistant with reality, not with providing tech support.
But, posting "blah blah Linux elitism blah blah will never succeed," is guaranteed to be modded insightful, even though it's bullshit. You think there aren't elitist Windows or Mac users? If you do, you're deluding yourself.
There's tons of "fucking terrible" software for windows, too. Go into any computer store and there will be racks of CD burning software, and most of it sucks.
Then go look for their rack of "Windows Software for only $10." That stuff probably sucks, too. Look in the games section. It's like 10% good, well known games, and then 90% crappy knock-offs.
Most commercial software comes with huge disclaimers that say they aren't responsible when your computer blows up or whatever, just like open source software.
In other words, open source software isn't very different from commercial software, from an average buyer standpoint (I agree, there probably isn't support for some of the big stuff, like CAD). There's some really great stuff, and then there's tons of crap.
The only difference is that with open source, there's no $200 difference in price between the good stuff and the shit.
In order to damage chromosomes, you have to have enough energy to do something to them, like break various types of bonds.
For example, 60 Hz radiation from powerlines, which is oft seen in studies resembling this one, is, if I recall, below the amount of energy needed to even start a molecule of water rotating, and needless to say it's far below that needed to break the hydrogen and covalent bonds in DNA.
I think AM radio signals are similarly below this threshold, so if it does cause cancer, you need to come up with something more complex than "radiation damages DNA."
PCI Express is a replacement for PCI and AGP on desktop class motherboards (I guess PCI-X might be better for servers, but I don't know).
Its advantages are that it has switched uplinks, so, if I understand correctly, each device can have its maximum bandwidth between any other component. PCI shares its bandwidth between all devices.
PCI Express 16x replaces AGP, and roughly doubles the bandwidth, I think. Then there's 8x, 4x, 2x and 1x for devices with lower bandwidth requirements. And you could probably expand to 32x if you really need more bandwidth than 16x. It's all about the number of "lanes" you devote to a card.
Someone here has a link to an article on this stuff, in case you want a description from someone who actually knows what they're talking about.
That does not serve your point at all. We both know that the common user will have a choice between WinXP and XP Pro
Most people don't really have a choice at all, other than perhaps a checkbox on Dell.com. And if Linux were that way, it wouldn't matter either. Most people will have that decision made for them by a knowledgable person, and won't be installing it themselves.
But, assuming they are installing it themselves...
Also, even in friendly distros like Mandrake they don't always just set you up with a "clear" default unless you know enough to choose that during the install. Normally there is a list of several window managers you can choose to log into, include KDE, GNOME, IceWM, and a few others.
I'm pretty sure this isn't true. I don't know about Mandrake, but Red Hat does Gnome by default. If you want KDE, you need to find the "Switch to KDE" menu option somewhere. SuSE, I think, standardizes on KDE. It's not true that people have to choose this unless they want to (say, if they choose the expert installation).
Oh, and the fact that you can run GNOME apps on a KDE desktop and vice versa is great, but it is still confusing and totally non-obvious.
No it's not. They're installed. They're in the "start menu." You click the button and they run. The distro themes make Qt and GTK look the same. What's not obvious?
Your saying that GTK and QT apps look out of place together shows that you haven't used a new Linux distribution in a while (or if you have, you haven't payed any attention). In Fedora, for example, the KDE and Gnome desktops look almost exactly the same (when I had my friend switch from one to the other, I thought nothing had actually changed. It had, but it just looked exactly the same).
We all have this type of blindness to varying degrees.
People say that a lot. How do you know it's right? You think you're blind, but you don't really know, because you can't go back to seeing to test it out.
Modern Linux distributions are just as GUI-clickable/usable as Windows is for the great majority of people.
And for the more admin oriented people, there are plenty of distributions out there that implement clicky network card configuration and so on, just like Windows. And it hides just as much of what's really going on as Windows, so I have just as little of an idea of how to fix it when it breaks. I guess if I used the GUI tools all the time I'd know.
That's how people want it, I guess. But your assertations that we are somehow blind to the staggering complexity of Linux is, basically, horsefeathers. Sure, we may be used to command lines, but modern distributions let you do almost all the common stuff with the GUI, which is good enough for most people. Anything more complex they'd need to call their Windows guru friend for anyway (and I don't even know what you'd be dealing with then, either on Linux or Windows).
If you got Linux in the same situation as Windows now is (preinstalled and so on), there's no reason most people couldn't use it fine, including those who need to do a little tweaking once in a while.
You should probably study up and become an investment banker, and then convince every company selling Linux distributions to join in a big merger and become one company. I'm sure all of them will be pleased, and you'll make a lot of money.
First of all, which Linux? Redhat, slack, debian, Mandrake or maybe SuSE? That alone is reason enough for a customer to just pass it by and pick up a copy of windows.
They'll use whichever one comes installed on their PC, or whichever one their local geek installs. Just like Windows.
Oh then you need to pick a desktop, KDE, gnome, Windowmaker, etc etc etc etc.
Most distributions choose one desktop as standard, and then they use themes to make GTK and QT look the same (RedHat and Mandrake do, at least). Most people won't have any idea which they're running. I had a friend who installed Fedora, and switching between Gnome and KDE left the desktop almost identical.
Most people won't know Windowmaker exists.
99.9% of people will just pass and go for the system that just does it for them.
You mean Linux, which comes with 99.9% of the software most people use preinstalled? Or do you mean Windows where I have to run to Tucows to get a free CD ripper, or an IRC client, and so on. Oh, and which of the 4 different Office boxes should I get? Do I need FrontPage? This is so confusing!
Linux is 4 times more bloated with multple EVERTHING.
Yeah, it sure is a pain that Linux requires me to install 5 of everything. I sure wish I didn't have to go through and select every checkbox and install every software package on the install CDs.
It's a good thing that Windows doesn't have people building 58 ftp clients for it, or I'd have to go install every single one of those, too.
Unified interfaces are better, it's why windows and OSX are successful on the desktop and Linux is not.
Oh, you mean how Windows uses a different widget set for Office than it does for IE? And how Windows Media Player doesn't look the same either?
Yeah, that's much better than Redhat's Bluecurve theme making all the GTK and QT applications look identical (and that probably covers 99% of the applications that get installed/used).
Everything attempts to look like windows
I see. When Linux is different from Windows, it's hard to use, but when it's the same, it's just unimaginitive geeks mindlessly parroting Windows. You're totally right.
and the xservers today are no different from the ones from yesteryear.
Yeah, you've obviously been following the XOrg development lists. You're so informed. I wish several of Keith Packard's improvements were slated to be included in the next release in 2 1/2 weeks. And maybe after that they could plan on implementing hardware acceleration and such.
I see no forward movement
Yeah. It'd probably help you see if you opened your eyes, but that's too much trouble.
The 2.4 kernels (and possible 2.2, I'm not sure) still have new releases coming out.
How do you want to support new hardware and new features without adding to the kernel? Also, the kernel is modular, so you can choose to either not compile certain features at all, or to compile certain features as modules so that you don't use memory unless you need it.
And if you don't like how much stuff is in Gnome, you can choose not to compile some of it, or you can use one of the other 4 examples he gave that aren't old. KDE is generally regarded as faster, but there are other applications that slim down more. XFCE is a minimalist DE, and Fluxbox (and tons of others) are minimalist window managers.
The purpose of Gnome is to have a whole bunch of applications that integrate and work together to provide for most/all of a user's needs. If you don't want that, then, yes, use something else.
Yeah, you certainly don't use the functionality.:)
Some of the KDE stuff is kind of nice. For example, kmplayer wraps Xine/mplayer in a KPart so you can view videos in web pages (I know, there are similar plugins for Mozilla).
There are other interesting things you could do with it as well. for example, you could theoretically specify a vim/emacs kpart (only vim exists, as far as I know) as the default text editor part, and then kwrite and kate and so on would all automatically have vi/emacs behavior.
But I won't claim that you'll gain much with all of that. It's interesting and nice to see all applications integrating together so well, but if you have specific apps you prefer, you're probalby not going to get enough out of switching to make it worth it.
I don't use Gnome, so I can't tell you about it. I can give some KDE examples, though.
Lots of KDE apps are designed as kparts with wrappers around them to be actual applications. For example, KGhostview can view PDFs. But since it's written as a kpart, Konqueror can load up that kpart in itself and display PDFs. In fact, Konqueror isn't much but a wrapper for a whole bunch of kparts, which is how it can work as a browser (khtml) and also display images, text files, PDFs and so on.
Another example is kwrite and kate. Both use the text editor KPart, but kwrite just has a toolbar, while kate has a console built in (probably another kpart), and a file browser sidebar (another kpart), so it's more of a programmer's text editor.
Kontact is another example. It's a lot like Evolution/Outlook, but for mail it uses a KMail part, and for your datebook it uses a korganizer part (or something like that).
I believe Gnome has a similar API (bonobo, CORBA based). I don't think it's used quite as much as in KDE (a lot of "Gnome" apps are simply GTK apps, after all, and don't really make use of this kind of stuff. True Gnome apps could, though), but the functionality is there.
I don't have any definitive answer to that question. But there are reasons besides speed to write in C.
For example, it's easier to make language bindings for other languages from a C API. If you write GTK+ in OCaml, how easy is it to make Python and Ruby bindings? You'd need to use an OCaml -> C bridge and then use the Ruby -> C bridge to interface. Most languages have a C interoperability API, but the same can't necessarily be said of Lisp or ML. How easy is it to bridge between Perl and Python?
Or, maybe the people who wrote GTK just like C better. Personally, I like C better than C++, although I like Lisps and so on better.
Also, C is more widely known than ML and Lisp.
Also, perhaps GTK (which came about before Gnome, b the way) was once designed to be speedy. That doesn't mean that everything built with it needs to be speedy, or that it can't go off in another direction. Gnome was a lot more lightweight in the 1.x days, but it's changed. That doesn't mean it makes sense to rewrite all the libraries in a different language and make all old applications port to a new language.
I'm not saying that the Gnome people don't want to be as speedy as possible, but their primary objective is to build a comprehensive desktop. If you want lightweight stuff, go use lightweight stuff. Saying "Gnome uses C so it therefore is trying to be speedy" is not really logical.
Yes, you need to support new hardware and stuff but it's not sustainable to keep adding it into the core product.
What do you base this on? Do you have any evidence?
You can choose to compile drivers into the kernel or as modules or not at all. Some people use the Linux kernel in embedded devices. It isn't necessary to compile most things into the kernel.
Third parties can also maintain drivers that are separate from the kernel and are loaded as modules. Do you think Windows shouldn't include drivers for hardware on its install CD?
He was referring to MS Word
Do you have any Linux examples?
Printing doesn't happen immediately because it's necessary to convert the data into a format the printer understands. Printers don't understand Word files or PDFs or whatever. They can't magically see what's on you screen and begin printing immediately.
Maybe back in the days of daisy wheel printers it was easy to generate stuff the printer would understand, but these days you need to produce postscript or something else. Unless you want to edit all your documents as postscript, it's going to take some processing beforehand.
But you know, personally I think the output of modern printers actually looks good. You can have good looking printouts, you can have speedy printouts, and you can edit in whatever document format you want, but you can only choose two of the above.
I know you were implying more generality, but I think you're wrong. Yes, you can't use brand new Gnome on old hardware. But that's because it's made to take advantage of new hardware. It does more stuff than old fast applications did.
If you want a speedy desktop, you can use XFCE or Fluxbox. Or you could use bugfixed versions of older apps/Gnome/KDE if they exist. Gnome isn't designed to be super speedy.
Perhaps you should use something like XFCE that aims to be a slimmed down desktop environment. Or Fluxbox that's just a slim window manager.
Gnome aims to be a fully featured desktop environment, with all the apps a user needs (more or less). If that's not what you want, then you probably shouldn't use Gnome (or you could refrain from installing all the applications).
And how is the Linux kernel too bloated? Would you rather they not support any new hardware drivers or something? Do you have specific examples?
I didn't mean to say I think Gnome should go away. Even if it alienates its advanced users, it still has a place. And in any case, there are many advanced users who do like the changes in Gnome.
I said elsewhere in this story that I think having both Gnome and KDE is good, as they're targeting different things and have different principles. People who want KDE and Gnome and so on to all merge together into one project are misguided, in my opinion. There are many sides to this issue, and with one project, you can't satisfy both (you can't even satisfy everyone with two projects).
I just meant to say that that there are many people who complain about Gnome's lack of features, or whatever, and wouldn't even use it if it followed more of their ideals, since they use KDE or whatever. I don't see the point in that. But I agree, the two camps can learn from each other.
Yes, many people like Gnome. Many people liked Gnome 1 and don't like Gnome 2. Many people never liked Gnome.
Do I write articles complaining about Gnome? Or were you talking about other people. I think the reason they write such articles is because they had something they liked, and the people who made it decided to make it worse in their opinion. Wouldn't you be angry if the Gnome designers decided to turn it into a "poorly designed eye-candy infested piece of shit?"
And part of the loud people's frustration is that when they ask for changes back to the way they liked it, they get ignored or flamed or told they're ideas are stupid or whatever.
I wasn't saying Gnome will or should change. It's the decision of the maintainers which parts of their user base they want to listen to, and which they ignore. I was merely explaining the situation and its consequences. Okay?
Slashdot Mirror
Why would he give a 1000 word description of everything he did when two sentences works fine? And why do you (or any one else) need to know that stuff?
If your friend treats you to dinner because he 'got lucky in the stock market,' do you ask for his bank-book and a record of his portfolio?
Lay off the caffeine.
Really? That's strange because I am having a hell of a time playing back video on my 286. Where is the lighening fast program that will play it at the same speed I do now?
So, in other words, you're not running the same programs faster; you're running bigger, fancier programs that do things you couldn't do before.
Just like he said.
Is that what he said? It sounded to me like he was asking if Sybase is significantly better. If the performance/features are the same, and you can get the source to one (not to mention use it on multiple processors or with databases large than 5 GB), then then why not use the latter? You get more. Was he demanding that Sybase open source their stuff? No.
But, by all means, let's all jump down his throat for asking if and what significant advantages it has over OSS solutions.
technophobic grandmas
Sounds like a good name for a band.
That's not really the same thing. It doesn't do the smooth image scaling and such. In its previous form, it just took screenshots, slowly, and eventually showed a picture of them all.
However, if you want skippy to work better, you need to get skippyXD, which uses the XDamage extension, new in 6.8. That allows for live updates of windows while they're being displayed by skippy.
In other words, you can't do it well without things like composite and damage.
According to the text of the article, they switched the default theme to winter, so it looks more like this by default. Those screenshots on the E website are (if you look close) from E 16.0, and this is 16.7.
Also, I don't know what you heard, but E16 is just a window manager, like Fluxbox or KWin or Metacity. It isn't an never was trying to be a whole desktop environment. In fact, it used to be the default window manager for Gnome before Sawfish replaced it.
In other words, if you want panels and desktop icons and stuff, then you need to run Gnome or KDE with Enlightenment as the window manager, or you need to use iDesk or something like that to provide that extra functionality. E by itself is closer to the minimalist window managers.
E17 will be more like Gnome, KDE or XFCE, but that's been years in the making and may yet be years before it's released. But E16 was never trying to be like that. What you're doing is sort of like complaining that Fluxbox doesn't do everything that KDE does. E isn't designed to do fancy stuff out of the box and be GUI configurable in all aspects. That's what KDE and Gnome are for.
I don't know if I agree with you.
I don't have a Gmail account either, so I'll speculate too. I don't see why you couldn't take your sorting methods, and instead of outputting a web page, just output some XML or some other information on the threading structure of that mail box. Then you can display it as a web page, or as a folder in Kmail or whatever. You might need to store your mail in some Gmail database or whatever, but it could be separated. Who's to say they don't already do it this way?
The problem with your example is that the interface is simple enough to not be much of an issue. It's just a list of e-mails sorted some way, and probably a sidebar or something. All the interesting stuff is in the sorting and searching algorithms, and the interface is little more than dumping the output to the screen.
So in that case, it's easy to outsource the gui, but it's hard to make/design the backend. That doesn't mean the backend is inherently part of the interface.
If you have a bunch of data points, you can display that data as a graph or a chart or a table. But the underlying data is the same. In my mind, your example is similar. You have a bunch of e-mails. You can sort and relate those e-mails however you want without actually displaying anything. But if you want to work with them, you need an interface. That can be a text-based list or a gui with a bunch of different panels, or a bunch of 3d rendered stacks of boxes. The underlying data is the same. Interface design is about making sure the presentation of that data is effective for the user.
Maybe sorting e-mails is a bit of a crossover, but I still think it falls more on the functionality side. Sure, you can have visionaries design the algorithms and have a monkey code them. But you can have visionaries design user interfaces and then have a monkey code those as well. What I and the original poster are saying is that the ui visionaries should work on ui stuff, and the functionality visionaries should work on the functionality stuff.
I don't think you understood what the original poster said.
He didn't say, "Usability is for losers. People should learn overly complex interfaces." What he said is that you should separate 'what it does' from 'how it looks.'
Take k3b, for example. People usually say it's the best burner for Linux. I just click on a button to open a project, and then drag the files I want to burn onto a window. Or I can drag songs directly out of my media player, juk. Then I click a burn button and it makes the cd.
But, does k3b implement its own cd burning functionality? No, it uses cdrecord and cdrdao and whatever other command line utilities are out there. People who know what they're doing write the cdr writing stuff, and then the k3b people use those to make a nice interface for it.
Take gift as another example. It runs a file sharing client as a daemon, and then you use various gui programs to interact with the daemon. Don't like KDE? Use the Gnome interface. Or vice versa. Don't like any of the interfaces? Make up some totally new filesharing user interface. But underneath it all is the same piece of code providing the actual functionality, written by people who know how to implement it well.
When functionality and interface are separate, you can choose your interface. People can develop new, better interfaces while preserving the existing functionality, and without having to rewrite it, and you don't force people who aren't good at interface design to make all of the decisions for it.
That's what the original poster was saying. Not "usability doesn't matter."
While the point about not straight-cloning an interface may be a good one, what do you suggest?
"OSS should stop cloning MS and do something completely different, only it should be more usable." That's great, do you have any visionary ideas? How many ways are there to make a word processor or e-mail client work?
As for your window buttons example, you can put them wherever you want on the window. By default they're where they are on Windows presumably to take advantage of user familiarity. You can have as many panels as you want and put them wherever you want, and have buttons or start menus or whatever. You can even remove all desktop icons and have a right-click menu like fluxbox in KDE. Or you can turn off the desktop all together.
So, is your suggestion that the default setup should be arbitrarily different from MS? Or should people magically come up with totally revolutionary ideas? Then the projects will get flamed because "it's too hard for people to learn to use."
This is exactly what holds Slashdot back. Arrogant people spouting the same old stuff labeled as 'insightful.'
The original poster complained about some stuff. The reply pointed out some places where his complaints are inconsistant with reality (i.e. you don't need to recompile the kernel for nVidia drivers).
This person is not necessarily representative of the Linux community. Yes, he came off elitist. However, we're also reading Slashdot, where every elitist nerd comes to post his tripe. There are plenty of friendly people in the Linux community ready to help out newbies. Slashdot is not the place they hang out to do tech support, though.
Also, the "not worthy of my time" type comment was aimed at pointing out the other arguments the original poster made that aren't consistant with reality, not with providing tech support.
But, posting "blah blah Linux elitism blah blah will never succeed," is guaranteed to be modded insightful, even though it's bullshit. You think there aren't elitist Windows or Mac users? If you do, you're deluding yourself.
There's tons of "fucking terrible" software for windows, too. Go into any computer store and there will be racks of CD burning software, and most of it sucks.
Then go look for their rack of "Windows Software for only $10." That stuff probably sucks, too. Look in the games section. It's like 10% good, well known games, and then 90% crappy knock-offs.
Most commercial software comes with huge disclaimers that say they aren't responsible when your computer blows up or whatever, just like open source software.
In other words, open source software isn't very different from commercial software, from an average buyer standpoint (I agree, there probably isn't support for some of the big stuff, like CAD). There's some really great stuff, and then there's tons of crap.
The only difference is that with open source, there's no $200 difference in price between the good stuff and the shit.
It's not that simple.
In order to damage chromosomes, you have to have enough energy to do something to them, like break various types of bonds.
For example, 60 Hz radiation from powerlines, which is oft seen in studies resembling this one, is, if I recall, below the amount of energy needed to even start a molecule of water rotating, and needless to say it's far below that needed to break the hydrogen and covalent bonds in DNA.
I think AM radio signals are similarly below this threshold, so if it does cause cancer, you need to come up with something more complex than "radiation damages DNA."
I don't think there is an "AGP camp."
PCI Express is a replacement for PCI and AGP on desktop class motherboards (I guess PCI-X might be better for servers, but I don't know).
Its advantages are that it has switched uplinks, so, if I understand correctly, each device can have its maximum bandwidth between any other component. PCI shares its bandwidth between all devices.
PCI Express 16x replaces AGP, and roughly doubles the bandwidth, I think. Then there's 8x, 4x, 2x and 1x for devices with lower bandwidth requirements. And you could probably expand to 32x if you really need more bandwidth than 16x. It's all about the number of "lanes" you devote to a card.
Someone here has a link to an article on this stuff, in case you want a description from someone who actually knows what they're talking about.
Hmm...
That does not serve your point at all. We both know that the common user will have a choice between WinXP and XP Pro
Most people don't really have a choice at all, other than perhaps a checkbox on Dell.com. And if Linux were that way, it wouldn't matter either. Most people will have that decision made for them by a knowledgable person, and won't be installing it themselves.
But, assuming they are installing it themselves...
Also, even in friendly distros like Mandrake they don't always just set you up with a "clear" default unless you know enough to choose that during the install. Normally there is a list of several window managers you can choose to log into, include KDE, GNOME, IceWM, and a few others.
I'm pretty sure this isn't true. I don't know about Mandrake, but Red Hat does Gnome by default. If you want KDE, you need to find the "Switch to KDE" menu option somewhere. SuSE, I think, standardizes on KDE. It's not true that people have to choose this unless they want to (say, if they choose the expert installation).
Oh, and the fact that you can run GNOME apps on a KDE desktop and vice versa is great, but it is still confusing and totally non-obvious.
No it's not. They're installed. They're in the "start menu." You click the button and they run. The distro themes make Qt and GTK look the same. What's not obvious?
Your saying that GTK and QT apps look out of place together shows that you haven't used a new Linux distribution in a while (or if you have, you haven't payed any attention). In Fedora, for example, the KDE and Gnome desktops look almost exactly the same (when I had my friend switch from one to the other, I thought nothing had actually changed. It had, but it just looked exactly the same).
We all have this type of blindness to varying degrees.
People say that a lot. How do you know it's right? You think you're blind, but you don't really know, because you can't go back to seeing to test it out.
Modern Linux distributions are just as GUI-clickable/usable as Windows is for the great majority of people.
And for the more admin oriented people, there are plenty of distributions out there that implement clicky network card configuration and so on, just like Windows. And it hides just as much of what's really going on as Windows, so I have just as little of an idea of how to fix it when it breaks. I guess if I used the GUI tools all the time I'd know.
That's how people want it, I guess. But your assertations that we are somehow blind to the staggering complexity of Linux is, basically, horsefeathers. Sure, we may be used to command lines, but modern distributions let you do almost all the common stuff with the GUI, which is good enough for most people. Anything more complex they'd need to call their Windows guru friend for anyway (and I don't even know what you'd be dealing with then, either on Linux or Windows).
If you got Linux in the same situation as Windows now is (preinstalled and so on), there's no reason most people couldn't use it fine, including those who need to do a little tweaking once in a while.
That's a good idea.
You should probably study up and become an investment banker, and then convince every company selling Linux distributions to join in a big merger and become one company. I'm sure all of them will be pleased, and you'll make a lot of money.
First of all, which Linux? Redhat, slack, debian, Mandrake or maybe SuSE? That alone is reason enough for a customer to just pass it by and pick up a copy of windows.
They'll use whichever one comes installed on their PC, or whichever one their local geek installs. Just like Windows.
Oh then you need to pick a desktop, KDE, gnome, Windowmaker, etc etc etc etc.
Most distributions choose one desktop as standard, and then they use themes to make GTK and QT look the same (RedHat and Mandrake do, at least). Most people won't have any idea which they're running. I had a friend who installed Fedora, and switching between Gnome and KDE left the desktop almost identical.
Most people won't know Windowmaker exists.
99.9% of people will just pass and go for the system that just does it for them.
You mean Linux, which comes with 99.9% of the software most people use preinstalled? Or do you mean Windows where I have to run to Tucows to get a free CD ripper, or an IRC client, and so on. Oh, and which of the 4 different Office boxes should I get? Do I need FrontPage? This is so confusing!
Linux is 4 times more bloated with multple EVERTHING.
Yeah, it sure is a pain that Linux requires me to install 5 of everything. I sure wish I didn't have to go through and select every checkbox and install every software package on the install CDs.
It's a good thing that Windows doesn't have people building 58 ftp clients for it, or I'd have to go install every single one of those, too.
Unified interfaces are better, it's why windows and OSX are successful on the desktop and Linux is not.
Oh, you mean how Windows uses a different widget set for Office than it does for IE? And how Windows Media Player doesn't look the same either?
Yeah, that's much better than Redhat's Bluecurve theme making all the GTK and QT applications look identical (and that probably covers 99% of the applications that get installed/used).
Everything attempts to look like windows
I see. When Linux is different from Windows, it's hard to use, but when it's the same, it's just unimaginitive geeks mindlessly parroting Windows. You're totally right.
and the xservers today are no different from the ones from yesteryear.
Yeah, you've obviously been following the XOrg development lists. You're so informed. I wish several of Keith Packard's improvements were slated to be included in the next release in 2 1/2 weeks. And maybe after that they could plan on implementing hardware acceleration and such.
I see no forward movement
Yeah. It'd probably help you see if you opened your eyes, but that's too much trouble.
Boy, you sure told me.
The 2.4 kernels (and possible 2.2, I'm not sure) still have new releases coming out.
How do you want to support new hardware and new features without adding to the kernel? Also, the kernel is modular, so you can choose to either not compile certain features at all, or to compile certain features as modules so that you don't use memory unless you need it.
And if you don't like how much stuff is in Gnome, you can choose not to compile some of it, or you can use one of the other 4 examples he gave that aren't old. KDE is generally regarded as faster, but there are other applications that slim down more. XFCE is a minimalist DE, and Fluxbox (and tons of others) are minimalist window managers.
The purpose of Gnome is to have a whole bunch of applications that integrate and work together to provide for most/all of a user's needs. If you don't want that, then, yes, use something else.
Not on my version of Windows (W2k) you can't.
Right click the start menu/folder you want to add to, and select "Open". Then add your shortcuts there.
Yeah, you certainly don't use the functionality. :)
Some of the KDE stuff is kind of nice. For example, kmplayer wraps Xine/mplayer in a KPart so you can view videos in web pages (I know, there are similar plugins for Mozilla).
There are other interesting things you could do with it as well. for example, you could theoretically specify a vim/emacs kpart (only vim exists, as far as I know) as the default text editor part, and then kwrite and kate and so on would all automatically have vi/emacs behavior.
But I won't claim that you'll gain much with all of that. It's interesting and nice to see all applications integrating together so well, but if you have specific apps you prefer, you're probalby not going to get enough out of switching to make it worth it.
I don't use Gnome, so I can't tell you about it. I can give some KDE examples, though.
Lots of KDE apps are designed as kparts with wrappers around them to be actual applications. For example, KGhostview can view PDFs. But since it's written as a kpart, Konqueror can load up that kpart in itself and display PDFs. In fact, Konqueror isn't much but a wrapper for a whole bunch of kparts, which is how it can work as a browser (khtml) and also display images, text files, PDFs and so on.
Another example is kwrite and kate. Both use the text editor KPart, but kwrite just has a toolbar, while kate has a console built in (probably another kpart), and a file browser sidebar (another kpart), so it's more of a programmer's text editor.
Kontact is another example. It's a lot like Evolution/Outlook, but for mail it uses a KMail part, and for your datebook it uses a korganizer part (or something like that).
I believe Gnome has a similar API (bonobo, CORBA based). I don't think it's used quite as much as in KDE (a lot of "Gnome" apps are simply GTK apps, after all, and don't really make use of this kind of stuff. True Gnome apps could, though), but the functionality is there.
Then why's it written in C?
I don't have any definitive answer to that question. But there are reasons besides speed to write in C.
For example, it's easier to make language bindings for other languages from a C API. If you write GTK+ in OCaml, how easy is it to make Python and Ruby bindings? You'd need to use an OCaml -> C bridge and then use the Ruby -> C bridge to interface. Most languages have a C interoperability API, but the same can't necessarily be said of Lisp or ML. How easy is it to bridge between Perl and Python?
Or, maybe the people who wrote GTK just like C better. Personally, I like C better than C++, although I like Lisps and so on better.
Also, C is more widely known than ML and Lisp.
Also, perhaps GTK (which came about before Gnome, b the way) was once designed to be speedy. That doesn't mean that everything built with it needs to be speedy, or that it can't go off in another direction. Gnome was a lot more lightweight in the 1.x days, but it's changed. That doesn't mean it makes sense to rewrite all the libraries in a different language and make all old applications port to a new language.
I'm not saying that the Gnome people don't want to be as speedy as possible, but their primary objective is to build a comprehensive desktop. If you want lightweight stuff, go use lightweight stuff. Saying "Gnome uses C so it therefore is trying to be speedy" is not really logical.
Yes, you need to support new hardware and stuff but it's not sustainable to keep adding it into the core product.
What do you base this on? Do you have any evidence?
You can choose to compile drivers into the kernel or as modules or not at all. Some people use the Linux kernel in embedded devices. It isn't necessary to compile most things into the kernel.
Third parties can also maintain drivers that are separate from the kernel and are loaded as modules. Do you think Windows shouldn't include drivers for hardware on its install CD?
He was referring to MS Word
Do you have any Linux examples?
Printing doesn't happen immediately because it's necessary to convert the data into a format the printer understands. Printers don't understand Word files or PDFs or whatever. They can't magically see what's on you screen and begin printing immediately.
Maybe back in the days of daisy wheel printers it was easy to generate stuff the printer would understand, but these days you need to produce postscript or something else. Unless you want to edit all your documents as postscript, it's going to take some processing beforehand.
But you know, personally I think the output of modern printers actually looks good. You can have good looking printouts, you can have speedy printouts, and you can edit in whatever document format you want, but you can only choose two of the above.
I know you were implying more generality, but I think you're wrong. Yes, you can't use brand new Gnome on old hardware. But that's because it's made to take advantage of new hardware. It does more stuff than old fast applications did.
If you want a speedy desktop, you can use XFCE or Fluxbox. Or you could use bugfixed versions of older apps/Gnome/KDE if they exist. Gnome isn't designed to be super speedy.
Perhaps you should use something like XFCE that aims to be a slimmed down desktop environment. Or Fluxbox that's just a slim window manager.
Gnome aims to be a fully featured desktop environment, with all the apps a user needs (more or less). If that's not what you want, then you probably shouldn't use Gnome (or you could refrain from installing all the applications).
And how is the Linux kernel too bloated? Would you rather they not support any new hardware drivers or something? Do you have specific examples?
I didn't mean to say I think Gnome should go away. Even if it alienates its advanced users, it still has a place. And in any case, there are many advanced users who do like the changes in Gnome.
I said elsewhere in this story that I think having both Gnome and KDE is good, as they're targeting different things and have different principles. People who want KDE and Gnome and so on to all merge together into one project are misguided, in my opinion. There are many sides to this issue, and with one project, you can't satisfy both (you can't even satisfy everyone with two projects).
I just meant to say that that there are many people who complain about Gnome's lack of features, or whatever, and wouldn't even use it if it followed more of their ideals, since they use KDE or whatever. I don't see the point in that. But I agree, the two camps can learn from each other.
I'm sorry, did I do something to insult you?
Yes, many people like Gnome. Many people liked Gnome 1 and don't like Gnome 2. Many people never liked Gnome.
Do I write articles complaining about Gnome? Or were you talking about other people. I think the reason they write such articles is because they had something they liked, and the people who made it decided to make it worse in their opinion. Wouldn't you be angry if the Gnome designers decided to turn it into a "poorly designed eye-candy infested piece of shit?"
And part of the loud people's frustration is that when they ask for changes back to the way they liked it, they get ignored or flamed or told they're ideas are stupid or whatever.
I wasn't saying Gnome will or should change. It's the decision of the maintainers which parts of their user base they want to listen to, and which they ignore. I was merely explaining the situation and its consequences. Okay?