I was under the impression that the equations of relativity predict that nothing with positive mass-squared can go faster than light, and that nothing with negative mass-squared could go slower, or something like that.
That's where you come up with tachyons; particles with imaginary mass that always go faster than light.
Then again, the question remains whether such particles actually exist (imaginary numbers for 'real' quantities aren't well regarded in physics), and if you can make them if they do.
I think negative mass just makes you accellerate in the 'opposite' direction in a gravitational field. Feel free to correct me, though. It's been a while.
P.S.: By your criteria, quantum entanglement also transfers information faster than c, doesn't it? However, since you can't get that information without other information that travels at or slower than c, it effectively doesn't travel faster than c. I think that's what most people mean: you can't transfer information faster than c such that you can view it sooner than information travelling at c. Or are tachyons different?
That's hardly the fault of open source developers, is it? I'm sure Dell could make a Linux PC where the user could just plug-in and go. They'd know what all was in the PC, so they could configure it totally. But, I don't think they do this. From the looks of their site, their "Linux PCs" come with a copy of FreeDOS installed, with Linux in the box, so Microsoft doesn't destroy them, or something.
That doesn't mean it can't be done, it just means it isn't on a large scale. But how much can the Linux Desktop advocates do to put this kind of thing in place? They don't have the resources to start up a company the size of Dell.
I agree with you for the most part. Linux probably can't be used by the lowest common denominator of people. There's plenty of people out there who don't even want it to be (or at least don't care). However, the fact that it can't be currently isn't totally the fault of the "open source community" if you go by the LCD standard. If Dell computers came installed with FreeDOS and Windows XP in a box, people wouldn't be able to use it either.
I'd say Linux really fails usability when you look at (what, apparently, are) "moderate" users. People who can already use computers and do stuff by themselves, and configure things the way they want. Configuration is different on Linux, for the most part. That's not to say it's inherently harder (how is reading and writing a line in a file harder than searching through tons of configuration panels for the one you want?), but it's different and scary for people who've only done it one way and don't care to learn (I'm not saying they should have to, either).
And if so does your child know how to run linux? And if so does your child know the intricacies of linux and OSS in general? And if so, in all honesty, how normal is your child?
Does your child know how to "run" Windows? Does he know all the intricacies of Windows and closed-source software in general? And what does "run" mean?
Seriously, what do kids do on computers? I've watched my nephew and some young cousins, and they click on the buttons for their games and play them. How would this be any different in Linux today? I can click buttons and visit disney.com or whatever and it's not any harder than it would be in Windows.
I think the only real argument there is the fact that so many kids games might not exist on Linux.
Yes, programming at 7 or 8 is abnormal. I agree there.
Oh god shut up "average user". Are you also offended that your tv remote came with instructions to install the batteries? I mean, an average user who was as above average as you could figure out how to install the batteries, no?
You think the average user of a remote control doesn't know how to put the batteries in? I doubt that. Many companies write documentation for below average users. "Don't get in the tub with your hair drier," "Don't stick a fork in your eye." Stuff like that.
if you use, or know how to use, or have heard of anything but Microsoft Windows, then you're NOT a typical user.
If you use the computer for anything but writing e-mail and browsing the web, than you're probably not a 'typical user' by your definition. I doubt your 'average user' could do any configuration of anything on Windows either. They'd have as much luck adding a line to a configuration file as they would poking around all the different tabs and menus in the Windows control panel programs.
I'll say that I don't even know if Linux _should_ try to target the average user. It works fine for me and I don't care if it gets onto everyone's desktop. However, if typical users are how you represent, them, then they probably have trouble using anything that isn't totally set-up for them to begin with, and how is a totally set-up Linux box more difficult than a totally set-up Windows box? Either way, all I have to do is click on stuff.
True, but I don't really think the hobbyist crowd can do anything about what the advocacy crowd says. Neither one really reports to the other. Not to mention that both crowds are made up of individuals who do whatever the hell they want. It's not like the "Linux grandma desktop advocacy group" has meetings and such.:)
The duality won't go away any time soon. Some people will want Linux on the desktop, and some people won't care. What could go away is people assuming everyone feels the same way. Understanding this situation explains a lot about why the state of Linux usability is what it is, and what the 'solution' is, from the advocists' point of view -- the solution isn't necessarily to rely on the original author for grandma usability, that's more likely to come from boxed desktop distribution companies that collect money (in general), if it comes from anyone at all.
He didn't say that his experiences are the majority. You made that up yourself.
His point, and the original point of this thread, was that kids don't necessarily need to be sheltered from all possible complexity for them to use computers. In fact, I'd wager that kids could learn to edit configuration files and handle all sorts of nasty stuff faster than a similarly experienced adult could be taught.
1. There are more web browsers and mail programs that aren't of corporate origin than those that are. Some people may even claim that some can be "ready for the desktop" (KMail, Konqueror...). OpenOffice may be more compatible with MSWord than other offerings (KOffice, AbiWord...), but non-corporate programs exist there, too.
2. Perhaps.
3. Incorrect. You can criticize Slashdot or Mandrake or Red Hat for pushing it that way, but the original and possibly current intentions of most of the nebulous "Linux community" might not fall in the same line, so you're criticizing people based on principles they don't ascribe to. "This ice cream is okay, but it's not soupy."
I think people aren't bitching so much trying to explain to the insular Unix programmers out there what the expectations are.
So if you take what your parent said as given, they're trying to explain how to make programs for mass consumption to people who don't care to make programs for mass consumption?
The grandparent says that 'we need to focus on the basics.' What he fails to realize is that Calculus and Algebra (among other things) _are_ the basics when it comes to math. Lots of statistics is based on Calculus underneath. And if you don't know algebra, I don't know how you'll follow much of anything of importance in a discrete math course.
Sure, you could teach people the results of discrete math and statistics. "When we have this distribution (which you don't really understand) and you take the average, you get this." Then you're students will just have to say, "Okay, I believe you," because without Calculus and Algebra and such, there's no way they can derive that result for themselves.
We'll end up with a society where some people know how to derive the results, and the rest have just memorized them. And I don't remember the last time I've heard anyone say that memorizing a long list of facts was more fun than understanding a little of why those facts are true.
I wasn't attempting to do gymnastics. My statements are simply the way I read those posts.
You, on the other hand, seem to me to _want_ to see any post that doesn't agree with MDI as religiously biased. I could take almost any discussion where there are two opposite opinions, and interpret either side as religiously committed to their side.
In the purest sense of the word, yes, MDI is a religious issue -- on both sides. What does this mean? It means that you can only have an opinion on the subject. It's subjective. Your view that MDI is good is no less religious than their view that MDI is bad, because it's only your opinion, and it's only their opinion.
However, among the anti-MDI opinions on that site, there are explanations why MDI is not currently feasible to implement. Further there are some explanations of why MDI in general is against what I can only call the Unix-GUI-philosophy. This is also a religious issue, I suppose, but it means that without a lot of work, and a partial redefinition of the purpose of some large pieces of software, you _can't_ implement MDI (Note, I'm talking exclusively about GTK here, as there may be some toolkits that allow WiW on Linux).
The only real gymnastics I've done is saying things more nicely than the original posters, which is probably part of what you interpret as religious intolerance.
Maybe some people here were saying "This is what you should prefer." However, most of the people on the GIMP site were sayng "I don't prefer this" and "I don't support this" and "I don't think enough people like this to make it worth while to do." At least the ones you cited were, for the most part.
My initial point was that the story's view was flawed. Open Source projects are being held back from what? Their purpose is not necessarily to displace Windows and etc. etc. The GIMP is very successful. The jobs of the GIMP developers aren't to cater to every need, or to be usable by every Tom, Dick and Harry, and this is true of any open source project. They won't be hurt by ignoring some people. If you want something, you'll have to do it yourself, and if they don't want your work, you'll have to fork their project. If you don't want to do that work, then you need to pay someone to do it. Mandrake exists so that you can pay someone to make Linux more gui-user friendly.
By the way, you never explained why MDI is better than GIMP on a virtual desktop. Could you, because I honestly don't see it. It keeps all windows for the GIMP separate from others. Minimizing all windows at once is similar to switching desktops. Is it that there's one menu bar at the top of the screen? I could do that if it had some sort of KDE integration (which it doesn't, I know). I honestly don't see much of a difference, could you explain?
I'm quite sure I'm reading the posts you listed in your message. Reverse order:
28) This discusses a possible alternative to window-in-window, ending with, "I just wanted to point that there could be a third way..." He was proposing a possible way to please both crowds. That's not religious anti-MDI.
26) "And what exactly would be better about this interface? IMO it has nothing but disadvantages." IMO means In My Opinion. He's asking why WiW would be better than the current system, which he finds fine.
21) "IMO the full-screen WiW MDI interface should be discussed on the mailing-list before you start working on it. You risk to waste your time hacking on a feature that might not be included at all." He doesn't say it "probably won't be added." He says it might not be added, and you should discuss it on the mailing list, presumably to see if there's enough interest to make it worth your while. Of course, if you want it badly enough, I guess you could do it all for yourself, but I imagine most people wouldn't, as it's a lot of work.
15) Does state that "WiW is evil," but I think that can only be taken as hyperbole. He goes on to talk about how he doesn't think window managing should be the job of the toolkit, it's the WM's job and they should focus on proper window manager hints to give it some of the benefits that come from WiW without actually having WiW.
I already talked about the rest, so I won't again. I honestly don't see why you think these posts are religiously anti-MDI/WiW. Yes, they may state that they don't think it's worth doing, and they think people should spend their time elsewhere, but that is their opinion. People other than you are entitled to have opinions, right?
Most of those posts talk constructively about possible alternatives/improvments so that WiW isn't needed. Why is that totally unacceptable? If you won't accept possible alternatives, doesn't that make you as religiously pro-MDI as you say these people are against it? Can you say definitively that "Nothing but MDI will do" until you've actually tried the alternatives they're suggesting?
Also, might I ask what operating system you're using the GIMP on? If it's Windows, I understand that having Windows underneath your workspace might be a problem (and some of the comments you referenced actually admit this, and try to propose non-WiW solutions that will make both sides happy). However, if you're on Linux, how is a window-in-window interface different from running the GIMP on a separate virtual desktop? That is, after all, the main purpose of window-in-window.
Anyhow, people having a different opinion than you isn't religious dogmatism. Offering constructive arguments to try and change your mind (like, "GTK isn't designed to do this, and it probably shouldn't be for these reasons...") isn't religious dogmatism. It's people discussing their opinions, and everyone's entitled to do that (even if some people are assholes about it).
I didn't say that you don't contribute to free software. However, the fact that you aren't pleased with the GIMP's interface isn't inherently the problem of its primary developers. If _you_ want an alternative, unpopular interface, then it is likely that _you_ will have to do something about it. Whether that means making it yourself, paying someone to do it, or getting a group that shares your interests to do it, I can't say. Obviously it isn't the first, since you have more important things to be doing. However, the main GIMP developers also probably have better things to do than maintain multiple, separate interfaces for their program, especially if one is rather unpopular.
Specifically, I was referring to the fact that my argument wasn't religiously oriented. However, I went and read the comments you mention, let me comment on some of them:
6) It starts off with opinion, but is backed up by facts. GTK+ has no window-in-window widget support. The best they could do is tabs (I don't know if that's acceptable for you). This is a current practical limitation.
11) Similar to 6, but with less opinion. There is no Window-in-window support in GTK+, and there probably won't be in the future. Apparently the dockable suggestion won't work in its current form. These aren't religious arguments.
13) Is more religious
21) Simply suggests that you discuss it on the mailing list to make sure you aren't wasting your time coding something that won't actually be included. Although you could always fork with the MDI interface, I suppose, or even just keep it to yourself.
However, I'd say that none of the posts are any more religious than your view is. You say, "I think MDI is useful." They counter, "I don't think MDI is useful." Some may say it less eloquently than others, but ultimately they're just expressing their opinion. They aren't saying, "You're wrong, MDI isn't more useful for you," they're saying, "MDI isn't more useful for me." Saying otherwise is misinterpreting them.
They also say, "We can't easily implement MDI." So in addition to it being relatively unpopular, it's actually a significantly difficult problem. They try to give some examples why (GTK doesn't support Window-in-window; window management is the job of the window manager, which the GIMP isn't), and some possible 'work arounds' so to speak (multiple windows with separate desktops is similar to MDI).
There are some people who express their opinion as "MDI is terrible and no one wants it," which seems religious, but ultimately it's an exaggeration of their opinion. Many people there, though, give legitimate explanations of why it can't yet, and in their opinion shouldn't be done, and it's not all because there's a big religious conspiracy against MDI.
Well, since you lke MDI so much, perhaps you should write MDI interfaces to all your favorite applications. After all, it's very easy to totally restructure the GUI of a program to a different interface model. The developers just refuse to do so because of their religious intolerance toward your preferred method of interface. If they can do it so easily, it's probably easy for you to do.
Most open source developers write software for their own use. Some others (on big projects), write for other people to use. However, it would seem that most people don't share your particular interface preferences. Why should these projects take a lot of time to write an alternative interface for 2% of their potential user base, when they themselves won't gain anything by it?
That's not a religious argument. It's just and argument of practicality. If you want an application to behave some way, and other people don't, then you're either going to have to write it yourself, or pay someone to write it for you. Open source developers don't write software for you, they write it for themselves.
Your analogy is also flawed. Your friend just builds the engine, he doesn't have to design the interface. That's already been built. His projects are automatically usable because he doesn't have to build the usable part.
What your friend does would be more like building an open source Windows kernel for all the rest of the software in Windows to run on. Then it'd be just as usable as Windows. Or perhaps more correctly, it's like OS X. The underlying system is open source, but the interface is made by a big-money company.
As many people here have said, open source developers often write software for their own use, and release it for pubic use because some people might find it useful. They don't _need_ more people to use it. The only people they _need_ to use it are themselves, and they're usually technically advanced, so they don't need the hand holding that an average computer user needs, so they don't make it.
There are people who have world domination goals for open source software. These aren't necessarily the same people. If it's not the goal of some open source programmer to have the perfect newbie interface, he probably won't make it, so it needs to be done by one of two groups:
1) People who want open source domination
2) Companies that deal in open source software
If they want, they can pay people to make things usable to Joe Average, and the original programmers may even adopt the better interface. But the original developer may not care one way or another about it.
Bottom line is, it doesn't matter how many users there are (it does to some people, but they may be doing more preaching than coding). People who write L33T software for their own use don't need you to use it. And when you re-write their application from scratch so even a chid can use it, for your own use, maybe you should benefit other people by open sourcing it.
Personally, I find open software plenty usable for me, so I don't really care if it's usable for Joe Average. In fact, many changes to make it more usable to Joe Average (Gnome stripping out features because they're "confusing") just gets in my way, so in some ways I hope it never gets to being usable by every Joe, or at least that there are alternatives for people who aren't stupid.
Note also that the people who actually are concerned about usability (Red Hat, Mandrake, etc.) _are_ doing things about it, and they're doing a decent job of it. However, this isn't the responsibility of every random open source programmer. They just make stuff they can use.
Interesting? Perhaps the moderators fell off a truck.
Algol descendants aren't exactly "horrible abominations," but they're not inherently better than functional languages either. If you believe they are, you probably haven't spent enough time with functional languages.
You claim that the syntax is barely readable. I'll admit, Lisp is hard to get used to at first. However, I suspect that's because most people focus on the parentheses. If you actually get into using Lisp, and get a good editor that matches parentheses and indents for you, it becomes much easier. How is:
if(condition) {
stmt1; } else {
stmt2; }
so much different than:
(if condition
stmt1
stmt2)
? Also, if you don't like all the parentheses, have you looked at Haskell for example? There's more to functional languages than Lisp.
Also, how are the semantics of functional languages any more unclear than any other language? Sure, I don't expect someone to know what callcc does just by looking at it, but how would someone know what *foo does just by looking at it? Go up to a random novice with no pre-existing C knowledge and ask them what atoi does. The semantics of C-like languages are only "more clear" because you've been learning them for years, and you've never bothered to really learn functional languages.
As for non-recursive problems, you can program in a procedural fashion in Lisp. There are macros for loops so you don't have to write tail recursion yourself. Your argument there holds no water.
As a final note, I'd like to point out that as far as "catering to the machine," C is closer to that than modern functional languages. C is a couple steps above assembly; everything executes one line after another; you have variables that map to memory addresses; you have functions that are like blocks of code with labels (and some other magic). Haskell, on the other hand lets you write code like this:
f x = y * y
where y = z + sin z
where z = x * x
Which actually has to be executed in reverse order (more or less). That's not exactly catering to the machine.
This is true, however, on Windows at least (I don't konw MacOS), many times you need to automate GUI apps because there is no command line app to do the same thing.
In most cases, if there's a GUI application to do someting on Linux, there are one or more command line applications to do the same thing, so you _can_ use shell scripts to do many things that you'd have to script a GUI for in Windows.
On another note, Gnome and KDE have a little functionality in this regard with DCop and whatever the Gnome equivalent of that is (Bonobo? I'm not a Gnome guy), so you can access some parts of a GUI application from shell scripts. Other applications also have full-fledged scripting languages as well (Gimp has Scheme and Python as scripting languages, and Mozilla has Javascript, and I believe will be getting Python and possibly others). I'm sure it's a far cry from the level of GUI control you can get with WinBatch and AppleScript, but these things take time (and to refer back to the last point, there's less need for this from the perspective of being able to script every feasible task).
Mandrake, SuSE, and Red Hat all have applications to automatically resolve library dependencies for you. Upgrading is a non-issue if you stick to packages approved by your disto.
You may run into problems if you try to install random-aplication-x.rpm, but then, there's no guarantee that you random-application-x.exe will install all libraries perfectly on Windows either.
Installing printers a normal person would use (usb, parallel port) is similar in difficulty to Windows if you use the KDE printer manager. Setting up a Samba shared printer is also nearly as easy as Windows with this utility.
I don't remember what utility ESR used, or what kind of printer he was using exactly, but in gneral, the problems you mention aren't that difficult anymore.
That's the whole problem. Nothing ever goes away. This is the first time this has become a problem for the OSS development model, but a serious problem it is.
Is it?
I want a desktop OS where I can be sure that any application I develop will run on any desktop and look exactly the same on all of them, regardless of what language I develop it in.
If you develop in GTK+, it will look like GTK+ in every desktop; If you develop in QT, it will look like QT in every desktop; et-cetera. It doesn't matter what language you use, assuming that language has bindings.
I guess what I'm saying is that it's not the user experience of mac/windows that should be copied by linux, it's the whole model of development control that should be adopted to open source development, much like the kernel development process. Less and freedom and less choice.
I want to use KDE. I don't like Gnome/GTK+ as much as I like KDE/QT. You're not going to tell me I can't use it. That's not your decision.
Less pride and zealotry from the supporters of each camp.
The zealotry you see people post on Slashdot isn't necessarily the majority, or even a very significant part of the people who use a particular desktop. They're just vocal. The people who actually work on the projects, rather than sitting around yelling at each other here, are more focused on _doing_ and making a better product, and possibly, helping users. Even if your unify both desktop environments, you'll still have application zeaots and so on, so it won't make things any more newbie friendly. You'll just have removed some choice from people who actually appreciate it.
The linux people need to understand that ONE half-assed product is better than the choice between TWO superb products. This has certainly been realized at redmond.
Better in what way? Marketability? Certainly not usability.
Frankly, you've given little evidence to support your opinions. Will unifying all desktops help newbies? Currently distros configure users to have use one desktop by default and they never have to know the difference. In fact, they've developed GUI themes to make both GTK and QT look the same, so you can't tell which you're using unless you're experienced. How is this any harder or more inconsistent from a new user's perspective?
I don't think your arguments hold water.
Re:From sSomeone who pitches those PHB's...
on
Why PHBs Fear Linux
·
· Score: 1
I'm not sure I quite understand.
Are the Linux zealots coming in and giving presentations to the managers at your company? Or are your managers making business decisions by reading Slashdot?
I imagine that people actually trying to get Linux used within their company are likely to make a more reasonable, fact-based argument than the zealots you think are such a problem. And I'm sure a decent manager can give such a presentation from someone at their company the consideration it deserves.
There are Mac zealots and Microsoft zealots as well, and they can't take any sort of constructive criticism of their respective products. They're less visible to you because most probably don't post here. This is one of several centers for Linux zealots, so you're probably more acquainted with them, but that doesn't mean zealots are unique to Linux.
If your boss, or anyone for that matter, bases all their opinions off of what they read in the comments section of Slashdot or the linux-rulz mailing list, then they are, quite frankly, not very smart. There are plenty of level headed people out there in the position to influence real decisions. The opinions that zealots post on Slashdot and linux advocacy message boards are of little consequence (and likely aren't very visible to your average PHB type).
In conclusion, the fact that there are Linux zealots isn't going to destroy Linux or halt its adoption. There are zealots for other systems and it hasn't destroyed them. You're overreacting.
Help Java's greatest supporters face the competition of C#?
I imagine Java's best supporters are out getting work done with Java, not sitting on slashdot or javalobby arguing over which VM is best, which, I imagine, is where most debates like that occur.
Do fights break out at JavaOne because people are arguing over which VM is best? Most likely not.
And since when are two programming languages in competition with one another? Use and convince your employer to use whichever works best for you and your employer. Otherwise it's probably best to let Sun and MS fight marketing duels.
Seriously, I think I've only seen two valid arguments for open sourcing Java. One is that bugs will probably get fixed a little faster (Sun is pretty slow with that), and the other is the statement "I believe all software should be open source." At least with the latter, you're sticking to some principles.
I seriously doubt that unifying the Java bickerers on slashdot will have much of an effect. People who actually _use_ Java probably don't waste their time fighting over trivial things like whose JVM is better.
It's poor form to add to java.lang. Only approved, core stuff goes in there. Official extention stuff is javax.whatever. Stuff from microsoft would likely be com.microsoft.whatever.
If Microsoft writes its own extentions for current Java, it will be known as Microsoft's unofficial extention libraries to Java to most people, I'm sure.
If Java is GPLed, and MS forks it, and changes all kinds of stuff to be Windows only, and distributes it with every copy of Windows and Visual Studio, then I'm sure for about 80% of the computing world, it will be known as Java. That is, they have the ability to make their incompatible, Windows-only version more common than the legit version. How does the GPL solve this? It doesn't matter if the code is released if you don't want it in the main version (if you don't want to use it at all).
There are many open source languages, so to speak.
Ruby, Python, Perl, several flavors of Lisp, Haskell, I believe, OCaml, ML and Prolog, just to name a few. Some compile to machine code, some are interpreted. Perl is getting its own VM with Perl 6 (although I wouldn't hold my breath until it's released).
Really, there are two things behind Java/.Net. They have huge standard libraries, which would be difficult to duplicate. You may say that we have tons of open source programmers out there, but you don't want too many cooks spoiling the language. I'd say that in general, you want more of a core group working on your language and standard library, otherwise you'll run into inconsistencies. Getting the amount of libraries and such that Java/.Net have would take a long time (just look at the project trying to make Java class libraries from scratch, GNU Classpath).
Second, Java/.Net have huge marketing behind them. That's why Java is where it is today. It has backing from a major corporation and that sounds safe to other businesses. You're unlikely to get major commercial support for your homebrew, Java-like, open source language, so it'll end up being much smaller time.
Ruby, Haskell, etc. are all great languages (in fact, I'd rate most of the languages on my list higher than Java), but they generally lack these two properties (there's also the issue that most aren't suitable for they type of development that Java is used for, but that's another story:)).
So you can make another language (it happens all the time), but the odds of it being as widespread as Java are slim. And that's the issue, really. We have lots of great languages, but we can't use them for anything other than hobby programming.
(Not that I think Java needs to be open sourced. I don't see the utility myself. I tend to think most people here have a "just because" attitude towards the issue)
Strictly speaking, even if their changes have to be open source, they could still add all kinds of Windows only stuff, break compatibility with the main Java branch, and distribute it with every copy of Windows.
Then you'd be forced to write two implementations, one for real Java and one for MS Java, assuming the main branch doesn't want MS's arbitrary changes.
I'm not saying this would happen, but the "MS will have to release their code" people's arguments aren't without error.
"As opposed to value systems carefully crafted for you in the modern era by Madison Avenue?"
No, I meant more like the value systems that come from learning and thinking for myself.
MTV and the like just push a lot of bullshit to turn people into consumers, I agree. Perhaps if your children are seeing this in school, it's a good idea to discuss it with them and enlighten them to the attempted brainwashing their being exposed to.
However, never exposing them to anything you think is controversial isn't any better. They have to go out into the world and lose their innocence some time. Perhaps it's better to have that happen when you can explain what's going on and help them see the bull.
If you can teach your kids to think for themselves, you don't have to worry about MTV turning them into mindless consumers.
The differences between C# and Java are mostly fairly cosmetic (and will probably be even more so after 1.5, whenever that arrives). They are better than C++ in my opinion, but is mono really ready to convert a whole desktop environment to it?
If you really wanted to be innovative you'd couple KDE with a seriously higher level language like Smalltalk or Ruby, but then you'd have trouble compiling down to machine code and there'd likely be performance issues.
Slashdot Mirror
Why would anyone care about a grade on a random paper if it had nothing to do with the final grade?
"Oh, I got an F on this paper that doesn't count for anything and was graded by picking a letter at random."
"Oh, I got an A on this paper that doesn't count for anything and was graded by picking a letter at random."
"I got all A's on my papers and ended up with a D in the class, but Bobby got D's on his papers and got a B in the class."
I don't understand how any of those situations would motivate the student to do well or learn the material.
Don't you mean imaginary mass?
I was under the impression that the equations of relativity predict that nothing with positive mass-squared can go faster than light, and that nothing with negative mass-squared could go slower, or something like that.
That's where you come up with tachyons; particles with imaginary mass that always go faster than light.
Then again, the question remains whether such particles actually exist (imaginary numbers for 'real' quantities aren't well regarded in physics), and if you can make them if they do.
I think negative mass just makes you accellerate in the 'opposite' direction in a gravitational field. Feel free to correct me, though. It's been a while.
P.S.: By your criteria, quantum entanglement also transfers information faster than c, doesn't it? However, since you can't get that information without other information that travels at or slower than c, it effectively doesn't travel faster than c. I think that's what most people mean: you can't transfer information faster than c such that you can view it sooner than information travelling at c. Or are tachyons different?
There's a totally set up Linux box? :)
That's hardly the fault of open source developers, is it? I'm sure Dell could make a Linux PC where the user could just plug-in and go. They'd know what all was in the PC, so they could configure it totally. But, I don't think they do this. From the looks of their site, their "Linux PCs" come with a copy of FreeDOS installed, with Linux in the box, so Microsoft doesn't destroy them, or something.
That doesn't mean it can't be done, it just means it isn't on a large scale. But how much can the Linux Desktop advocates do to put this kind of thing in place? They don't have the resources to start up a company the size of Dell.
I agree with you for the most part. Linux probably can't be used by the lowest common denominator of people. There's plenty of people out there who don't even want it to be (or at least don't care). However, the fact that it can't be currently isn't totally the fault of the "open source community" if you go by the LCD standard. If Dell computers came installed with FreeDOS and Windows XP in a box, people wouldn't be able to use it either.
I'd say Linux really fails usability when you look at (what, apparently, are) "moderate" users. People who can already use computers and do stuff by themselves, and configure things the way they want. Configuration is different on Linux, for the most part. That's not to say it's inherently harder (how is reading and writing a line in a file harder than searching through tons of configuration panels for the one you want?), but it's different and scary for people who've only done it one way and don't care to learn (I'm not saying they should have to, either).
And if so does your child know how to run linux? And if so does your child know the intricacies of linux and OSS in general? And if so, in all honesty, how normal is your child?
Does your child know how to "run" Windows? Does he know all the intricacies of Windows and closed-source software in general? And what does "run" mean?
Seriously, what do kids do on computers? I've watched my nephew and some young cousins, and they click on the buttons for their games and play them. How would this be any different in Linux today? I can click buttons and visit disney.com or whatever and it's not any harder than it would be in Windows.
I think the only real argument there is the fact that so many kids games might not exist on Linux.
Yes, programming at 7 or 8 is abnormal. I agree there.
Oh god shut up "average user". Are you also offended that your tv remote came with instructions to install the batteries? I mean, an average user who was as above average as you could figure out how to install the batteries, no?
You think the average user of a remote control doesn't know how to put the batteries in? I doubt that. Many companies write documentation for below average users. "Don't get in the tub with your hair drier," "Don't stick a fork in your eye." Stuff like that.
if you use, or know how to use, or have heard of anything but Microsoft Windows, then you're NOT a typical user.
If you use the computer for anything but writing e-mail and browsing the web, than you're probably not a 'typical user' by your definition. I doubt your 'average user' could do any configuration of anything on Windows either. They'd have as much luck adding a line to a configuration file as they would poking around all the different tabs and menus in the Windows control panel programs.
I'll say that I don't even know if Linux _should_ try to target the average user. It works fine for me and I don't care if it gets onto everyone's desktop. However, if typical users are how you represent, them, then they probably have trouble using anything that isn't totally set-up for them to begin with, and how is a totally set-up Linux box more difficult than a totally set-up Windows box? Either way, all I have to do is click on stuff.
True, but I don't really think the hobbyist crowd can do anything about what the advocacy crowd says. Neither one really reports to the other. Not to mention that both crowds are made up of individuals who do whatever the hell they want. It's not like the "Linux grandma desktop advocacy group" has meetings and such. :)
The duality won't go away any time soon. Some people will want Linux on the desktop, and some people won't care. What could go away is people assuming everyone feels the same way. Understanding this situation explains a lot about why the state of Linux usability is what it is, and what the 'solution' is, from the advocists' point of view -- the solution isn't necessarily to rely on the original author for grandma usability, that's more likely to come from boxed desktop distribution companies that collect money (in general), if it comes from anyone at all.
He didn't say that his experiences are the majority. You made that up yourself.
His point, and the original point of this thread, was that kids don't necessarily need to be sheltered from all possible complexity for them to use computers. In fact, I'd wager that kids could learn to edit configuration files and handle all sorts of nasty stuff faster than a similarly experienced adult could be taught.
1. There are more web browsers and mail programs that aren't of corporate origin than those that are. Some people may even claim that some can be "ready for the desktop" (KMail, Konqueror...). OpenOffice may be more compatible with MSWord than other offerings (KOffice, AbiWord...), but non-corporate programs exist there, too.
2. Perhaps.
3. Incorrect. You can criticize Slashdot or Mandrake or Red Hat for pushing it that way, but the original and possibly current intentions of most of the nebulous "Linux community" might not fall in the same line, so you're criticizing people based on principles they don't ascribe to. "This ice cream is okay, but it's not soupy."
I think people aren't bitching so much trying to explain to the insular Unix programmers out there what the expectations are.
So if you take what your parent said as given, they're trying to explain how to make programs for mass consumption to people who don't care to make programs for mass consumption?
I'll just say I agree wholely with you.
The grandparent says that 'we need to focus on the basics.' What he fails to realize is that Calculus and Algebra (among other things) _are_ the basics when it comes to math. Lots of statistics is based on Calculus underneath. And if you don't know algebra, I don't know how you'll follow much of anything of importance in a discrete math course.
Sure, you could teach people the results of discrete math and statistics. "When we have this distribution (which you don't really understand) and you take the average, you get this." Then you're students will just have to say, "Okay, I believe you," because without Calculus and Algebra and such, there's no way they can derive that result for themselves.
We'll end up with a society where some people know how to derive the results, and the rest have just memorized them. And I don't remember the last time I've heard anyone say that memorizing a long list of facts was more fun than understanding a little of why those facts are true.
I wasn't attempting to do gymnastics. My statements are simply the way I read those posts.
You, on the other hand, seem to me to _want_ to see any post that doesn't agree with MDI as religiously biased. I could take almost any discussion where there are two opposite opinions, and interpret either side as religiously committed to their side.
In the purest sense of the word, yes, MDI is a religious issue -- on both sides. What does this mean? It means that you can only have an opinion on the subject. It's subjective. Your view that MDI is good is no less religious than their view that MDI is bad, because it's only your opinion, and it's only their opinion.
However, among the anti-MDI opinions on that site, there are explanations why MDI is not currently feasible to implement. Further there are some explanations of why MDI in general is against what I can only call the Unix-GUI-philosophy. This is also a religious issue, I suppose, but it means that without a lot of work, and a partial redefinition of the purpose of some large pieces of software, you _can't_ implement MDI (Note, I'm talking exclusively about GTK here, as there may be some toolkits that allow WiW on Linux).
The only real gymnastics I've done is saying things more nicely than the original posters, which is probably part of what you interpret as religious intolerance.
Maybe some people here were saying "This is what you should prefer." However, most of the people on the GIMP site were sayng "I don't prefer this" and "I don't support this" and "I don't think enough people like this to make it worth while to do." At least the ones you cited were, for the most part.
My initial point was that the story's view was flawed. Open Source projects are being held back from what? Their purpose is not necessarily to displace Windows and etc. etc. The GIMP is very successful. The jobs of the GIMP developers aren't to cater to every need, or to be usable by every Tom, Dick and Harry, and this is true of any open source project. They won't be hurt by ignoring some people. If you want something, you'll have to do it yourself, and if they don't want your work, you'll have to fork their project. If you don't want to do that work, then you need to pay someone to do it. Mandrake exists so that you can pay someone to make Linux more gui-user friendly.
By the way, you never explained why MDI is better than GIMP on a virtual desktop. Could you, because I honestly don't see it. It keeps all windows for the GIMP separate from others. Minimizing all windows at once is similar to switching desktops. Is it that there's one menu bar at the top of the screen? I could do that if it had some sort of KDE integration (which it doesn't, I know). I honestly don't see much of a difference, could you explain?
I'm quite sure I'm reading the posts you listed in your message. Reverse order:
28) This discusses a possible alternative to window-in-window, ending with, "I just wanted to
point that there could be a third way..." He was proposing a possible way to please both crowds. That's not religious anti-MDI.
26) "And what exactly would be better about this interface? IMO it has nothing but disadvantages." IMO means In My Opinion. He's asking why WiW would be better than the current system, which he finds fine.
21) "IMO the full-screen WiW MDI interface should be discussed on the mailing-list before you start working on it. You risk to waste your time hacking on a feature that might not be included at all." He doesn't say it "probably won't be added." He says it might not be added, and you should discuss it on the mailing list, presumably to see if there's enough interest to make it worth your while. Of course, if you want it badly enough, I guess you could do it all for yourself, but I imagine most people wouldn't, as it's a lot of work.
15) Does state that "WiW is evil," but I think that can only be taken as hyperbole. He goes on to talk about how he doesn't think window managing should be the job of the toolkit, it's the WM's job and they should focus on proper window manager hints to give it some of the benefits that come from WiW without actually having WiW.
I already talked about the rest, so I won't again. I honestly don't see why you think these posts are religiously anti-MDI/WiW. Yes, they may state that they don't think it's worth doing, and they think people should spend their time elsewhere, but that is their opinion. People other than you are entitled to have opinions, right?
Most of those posts talk constructively about possible alternatives/improvments so that WiW isn't needed. Why is that totally unacceptable? If you won't accept possible alternatives, doesn't that make you as religiously pro-MDI as you say these people are against it? Can you say definitively that "Nothing but MDI will do" until you've actually tried the alternatives they're suggesting?
Also, might I ask what operating system you're using the GIMP on? If it's Windows, I understand that having Windows underneath your workspace might be a problem (and some of the comments you referenced actually admit this, and try to propose non-WiW solutions that will make both sides happy). However, if you're on Linux, how is a window-in-window interface different from running the GIMP on a separate virtual desktop? That is, after all, the main purpose of window-in-window.
Anyhow, people having a different opinion than you isn't religious dogmatism. Offering constructive arguments to try and change your mind (like, "GTK isn't designed to do this, and it probably shouldn't be for these reasons...") isn't religious dogmatism. It's people discussing their opinions, and everyone's entitled to do that (even if some people are assholes about it).
I didn't say that you don't contribute to free software. However, the fact that you aren't pleased with the GIMP's interface isn't inherently the problem of its primary developers. If _you_ want an alternative, unpopular interface, then it is likely that _you_ will have to do something about it. Whether that means making it yourself, paying someone to do it, or getting a group that shares your interests to do it, I can't say. Obviously it isn't the first, since you have more important things to be doing. However, the main GIMP developers also probably have better things to do than maintain multiple, separate interfaces for their program, especially if one is rather unpopular.
Specifically, I was referring to the fact that my argument wasn't religiously oriented. However, I went and read the comments you mention, let me comment on some of them:
6) It starts off with opinion, but is backed up by facts. GTK+ has no window-in-window widget support. The best they could do is tabs (I don't know if that's acceptable for you). This is a current practical limitation.
11) Similar to 6, but with less opinion. There is no Window-in-window support in GTK+, and there probably won't be in the future. Apparently the dockable suggestion won't work in its current form. These aren't religious arguments.
13) Is more religious
21) Simply suggests that you discuss it on the mailing list to make sure you aren't wasting your time coding something that won't actually be included. Although you could always fork with the MDI interface, I suppose, or even just keep it to yourself.
However, I'd say that none of the posts are any more religious than your view is. You say, "I think MDI is useful." They counter, "I don't think MDI is useful." Some may say it less eloquently than others, but ultimately they're just expressing their opinion. They aren't saying, "You're wrong, MDI isn't more useful for you," they're saying, "MDI isn't more useful for me." Saying otherwise is misinterpreting them.
They also say, "We can't easily implement MDI." So in addition to it being relatively unpopular, it's actually a significantly difficult problem. They try to give some examples why (GTK doesn't support Window-in-window; window management is the job of the window manager, which the GIMP isn't), and some possible 'work arounds' so to speak (multiple windows with separate desktops is similar to MDI).
There are some people who express their opinion as "MDI is terrible and no one wants it," which seems religious, but ultimately it's an exaggeration of their opinion. Many people there, though, give legitimate explanations of why it can't yet, and in their opinion shouldn't be done, and it's not all because there's a big religious conspiracy against MDI.
Well, since you lke MDI so much, perhaps you should write MDI interfaces to all your favorite applications. After all, it's very easy to totally restructure the GUI of a program to a different interface model. The developers just refuse to do so because of their religious intolerance toward your preferred method of interface. If they can do it so easily, it's probably easy for you to do.
Most open source developers write software for their own use. Some others (on big projects), write for other people to use. However, it would seem that most people don't share your particular interface preferences. Why should these projects take a lot of time to write an alternative interface for 2% of their potential user base, when they themselves won't gain anything by it?
That's not a religious argument. It's just and argument of practicality. If you want an application to behave some way, and other people don't, then you're either going to have to write it yourself, or pay someone to write it for you. Open source developers don't write software for you, they write it for themselves.
Your analogy is also flawed. Your friend just builds the engine, he doesn't have to design the interface. That's already been built. His projects are automatically usable because he doesn't have to build the usable part.
What your friend does would be more like building an open source Windows kernel for all the rest of the software in Windows to run on. Then it'd be just as usable as Windows. Or perhaps more correctly, it's like OS X. The underlying system is open source, but the interface is made by a big-money company.
As many people here have said, open source developers often write software for their own use, and release it for pubic use because some people might find it useful. They don't _need_ more people to use it. The only people they _need_ to use it are themselves, and they're usually technically advanced, so they don't need the hand holding that an average computer user needs, so they don't make it.
There are people who have world domination goals for open source software. These aren't necessarily the same people. If it's not the goal of some open source programmer to have the perfect newbie interface, he probably won't make it, so it needs to be done by one of two groups:
1) People who want open source domination
2) Companies that deal in open source software
If they want, they can pay people to make things usable to Joe Average, and the original programmers may even adopt the better interface. But the original developer may not care one way or another about it.
Bottom line is, it doesn't matter how many users there are (it does to some people, but they may be doing more preaching than coding). People who write L33T software for their own use don't need you to use it. And when you re-write their application from scratch so even a chid can use it, for your own use, maybe you should benefit other people by open sourcing it.
Personally, I find open software plenty usable for me, so I don't really care if it's usable for Joe Average. In fact, many changes to make it more usable to Joe Average (Gnome stripping out features because they're "confusing") just gets in my way, so in some ways I hope it never gets to being usable by every Joe, or at least that there are alternatives for people who aren't stupid.
Note also that the people who actually are concerned about usability (Red Hat, Mandrake, etc.) _are_ doing things about it, and they're doing a decent job of it. However, this isn't the responsibility of every random open source programmer. They just make stuff they can use.
Interesting? Perhaps the moderators fell off a truck.
Algol descendants aren't exactly "horrible abominations," but they're not inherently better than functional languages either. If you believe they are, you probably haven't spent enough time with functional languages.
You claim that the syntax is barely readable. I'll admit, Lisp is hard to get used to at first. However, I suspect that's because most people focus on the parentheses. If you actually get into using Lisp, and get a good editor that matches parentheses and indents for you, it becomes much easier. How is:
if(condition) {
stmt1;
} else {
stmt2;
}
so much different than:
(if condition
stmt1
stmt2)
? Also, if you don't like all the parentheses, have you looked at Haskell for example? There's more to functional languages than Lisp.
Also, how are the semantics of functional languages any more unclear than any other language? Sure, I don't expect someone to know what callcc does just by looking at it, but how would someone know what *foo does just by looking at it? Go up to a random novice with no pre-existing C knowledge and ask them what atoi does. The semantics of C-like languages are only "more clear" because you've been learning them for years, and you've never bothered to really learn functional languages.
As for non-recursive problems, you can program in a procedural fashion in Lisp. There are macros for loops so you don't have to write tail recursion yourself. Your argument there holds no water.
As a final note, I'd like to point out that as far as "catering to the machine," C is closer to that than modern functional languages. C is a couple steps above assembly; everything executes one line after another; you have variables that map to memory addresses; you have functions that are like blocks of code with labels (and some other magic). Haskell, on the other hand lets you write code like this:
f x = y * y
where y = z + sin z
where z = x * x
Which actually has to be executed in reverse order (more or less). That's not exactly catering to the machine.
This is true, however, on Windows at least (I don't konw MacOS), many times you need to automate GUI apps because there is no command line app to do the same thing.
In most cases, if there's a GUI application to do someting on Linux, there are one or more command line applications to do the same thing, so you _can_ use shell scripts to do many things that you'd have to script a GUI for in Windows.
On another note, Gnome and KDE have a little functionality in this regard with DCop and whatever the Gnome equivalent of that is (Bonobo? I'm not a Gnome guy), so you can access some parts of a GUI application from shell scripts. Other applications also have full-fledged scripting languages as well (Gimp has Scheme and Python as scripting languages, and Mozilla has Javascript, and I believe will be getting Python and possibly others). I'm sure it's a far cry from the level of GUI control you can get with WinBatch and AppleScript, but these things take time (and to refer back to the last point, there's less need for this from the perspective of being able to script every feasible task).
Mandrake, SuSE, and Red Hat all have applications to automatically resolve library dependencies for you. Upgrading is a non-issue if you stick to packages approved by your disto.
You may run into problems if you try to install random-aplication-x.rpm, but then, there's no guarantee that you random-application-x.exe will install all libraries perfectly on Windows either.
Installing printers a normal person would use (usb, parallel port) is similar in difficulty to Windows if you use the KDE printer manager. Setting up a Samba shared printer is also nearly as easy as Windows with this utility.
I don't remember what utility ESR used, or what kind of printer he was using exactly, but in gneral, the problems you mention aren't that difficult anymore.
That's the whole problem. Nothing ever goes away. This is the first time this has become a problem for the OSS development model, but a serious problem it is.
Is it?
I want a desktop OS where I can be sure that any application I develop will run on any desktop and look exactly the same on all of them, regardless of what language I develop it in.
If you develop in GTK+, it will look like GTK+ in every desktop; If you develop in QT, it will look like QT in every desktop; et-cetera. It doesn't matter what language you use, assuming that language has bindings.
I guess what I'm saying is that it's not the user experience of mac/windows that should be copied by linux, it's the whole model of development control that should be adopted to open source development, much like the kernel development process. Less and freedom and less choice.
I want to use KDE. I don't like Gnome/GTK+ as much as I like KDE/QT. You're not going to tell me I can't use it. That's not your decision.
Less pride and zealotry from the supporters of each camp.
The zealotry you see people post on Slashdot isn't necessarily the majority, or even a very significant part of the people who use a particular desktop. They're just vocal. The people who actually work on the projects, rather than sitting around yelling at each other here, are more focused on _doing_ and making a better product, and possibly, helping users. Even if your unify both desktop environments, you'll still have application zeaots and so on, so it won't make things any more newbie friendly. You'll just have removed some choice from people who actually appreciate it.
The linux people need to understand that ONE half-assed product is better than the choice between TWO superb products. This has certainly been realized at redmond.
Better in what way? Marketability? Certainly not usability.
Frankly, you've given little evidence to support your opinions. Will unifying all desktops help newbies? Currently distros configure users to have use one desktop by default and they never have to know the difference. In fact, they've developed GUI themes to make both GTK and QT look the same, so you can't tell which you're using unless you're experienced. How is this any harder or more inconsistent from a new user's perspective?
I don't think your arguments hold water.
I'm not sure I quite understand.
Are the Linux zealots coming in and giving presentations to the managers at your company? Or are your managers making business decisions by reading Slashdot?
I imagine that people actually trying to get Linux used within their company are likely to make a more reasonable, fact-based argument than the zealots you think are such a problem. And I'm sure a decent manager can give such a presentation from someone at their company the consideration it deserves.
There are Mac zealots and Microsoft zealots as well, and they can't take any sort of constructive criticism of their respective products. They're less visible to you because most probably don't post here. This is one of several centers for Linux zealots, so you're probably more acquainted with them, but that doesn't mean zealots are unique to Linux.
If your boss, or anyone for that matter, bases all their opinions off of what they read in the comments section of Slashdot or the linux-rulz mailing list, then they are, quite frankly, not very smart. There are plenty of level headed people out there in the position to influence real decisions. The opinions that zealots post on Slashdot and linux advocacy message boards are of little consequence (and likely aren't very visible to your average PHB type).
In conclusion, the fact that there are Linux zealots isn't going to destroy Linux or halt its adoption. There are zealots for other systems and it hasn't destroyed them. You're overreacting.
And if he meant fusion, he can look up hydrogen bombs.
It will help focus the Java community?
Help Java's greatest supporters face the competition of C#?
I imagine Java's best supporters are out getting work done with Java, not sitting on slashdot or javalobby arguing over which VM is best, which, I imagine, is where most debates like that occur.
Do fights break out at JavaOne because people are arguing over which VM is best? Most likely not.
And since when are two programming languages in competition with one another? Use and convince your employer to use whichever works best for you and your employer. Otherwise it's probably best to let Sun and MS fight marketing duels.
Seriously, I think I've only seen two valid arguments for open sourcing Java. One is that bugs will probably get fixed a little faster (Sun is pretty slow with that), and the other is the statement "I believe all software should be open source." At least with the latter, you're sticking to some principles.
I seriously doubt that unifying the Java bickerers on slashdot will have much of an effect. People who actually _use_ Java probably don't waste their time fighting over trivial things like whose JVM is better.
It's poor form to add to java.lang. Only approved, core stuff goes in there. Official extention stuff is javax.whatever. Stuff from microsoft would likely be com.microsoft.whatever.
If Microsoft writes its own extentions for current Java, it will be known as Microsoft's unofficial extention libraries to Java to most people, I'm sure.
If Java is GPLed, and MS forks it, and changes all kinds of stuff to be Windows only, and distributes it with every copy of Windows and Visual Studio, then I'm sure for about 80% of the computing world, it will be known as Java. That is, they have the ability to make their incompatible, Windows-only version more common than the legit version. How does the GPL solve this? It doesn't matter if the code is released if you don't want it in the main version (if you don't want to use it at all).
There are many open source languages, so to speak.
:)).
Ruby, Python, Perl, several flavors of Lisp, Haskell, I believe, OCaml, ML and Prolog, just to name a few. Some compile to machine code, some are interpreted. Perl is getting its own VM with Perl 6 (although I wouldn't hold my breath until it's released).
Really, there are two things behind Java/.Net. They have huge standard libraries, which would be difficult to duplicate. You may say that we have tons of open source programmers out there, but you don't want too many cooks spoiling the language. I'd say that in general, you want more of a core group working on your language and standard library, otherwise you'll run into inconsistencies. Getting the amount of libraries and such that Java/.Net have would take a long time (just look at the project trying to make Java class libraries from scratch, GNU Classpath).
Second, Java/.Net have huge marketing behind them. That's why Java is where it is today. It has backing from a major corporation and that sounds safe to other businesses. You're unlikely to get major commercial support for your homebrew, Java-like, open source language, so it'll end up being much smaller time.
Ruby, Haskell, etc. are all great languages (in fact, I'd rate most of the languages on my list higher than Java), but they generally lack these two properties (there's also the issue that most aren't suitable for they type of development that Java is used for, but that's another story
So you can make another language (it happens all the time), but the odds of it being as widespread as Java are slim. And that's the issue, really. We have lots of great languages, but we can't use them for anything other than hobby programming.
(Not that I think Java needs to be open sourced. I don't see the utility myself. I tend to think most people here have a "just because" attitude towards the issue)
Strictly speaking, even if their changes have to be open source, they could still add all kinds of Windows only stuff, break compatibility with the main Java branch, and distribute it with every copy of Windows.
Then you'd be forced to write two implementations, one for real Java and one for MS Java, assuming the main branch doesn't want MS's arbitrary changes.
I'm not saying this would happen, but the "MS will have to release their code" people's arguments aren't without error.
"As opposed to value systems carefully crafted for you in the modern era by Madison Avenue?"
No, I meant more like the value systems that come from learning and thinking for myself.
MTV and the like just push a lot of bullshit to turn people into consumers, I agree. Perhaps if your children are seeing this in school, it's a good idea to discuss it with them and enlighten them to the attempted brainwashing their being exposed to.
However, never exposing them to anything you think is controversial isn't any better. They have to go out into the world and lose their innocence some time. Perhaps it's better to have that happen when you can explain what's going on and help them see the bull.
If you can teach your kids to think for themselves, you don't have to worry about MTV turning them into mindless consumers.
The differences between C# and Java are mostly fairly cosmetic (and will probably be even more so after 1.5, whenever that arrives). They are better than C++ in my opinion, but is mono really ready to convert a whole desktop environment to it?
If you really wanted to be innovative you'd couple KDE with a seriously higher level language like Smalltalk or Ruby, but then you'd have trouble compiling down to machine code and there'd likely be performance issues.