Do you think that "Joe Computer Geek" is going to be able to get his hands on this to stream his own DVDs? Probably not. There will probably be some sort of DRM built in because the MPAA (as well as the RIAA) is too busy focusing on a few potential lost sales vs. the big revenues that could be had if they just opened their stuff up to internet distribution. They are looking at everything through an outdated selling concept. Not everyone thinks this way though... Peter Gabriel has his entire new album (UP) available to listen to in a streamable format as well as the video for his first single. The quality is low, so it encourages people to buy the real deal, but it's the entire album, so it allows for "try before you buy". The same could be applied to DVD pre-release and this technology would be great for it. But, it's still not going to be something that you or I can legitimately use to stream our own DVDs unless there are a LOT of restrictions. I for one am no longer sure of the legality of me streaming my MP3s to myself at work with icecast and not paying the RIAA those stupid broadcaster's fees. Discuss amongst yourselves.
I should have been more specific. All of the "terrorism" that has been "averted" since 9/11/2001 has nothing to do with Iraq. The man in the Whitehouse who does NOT represent me doesn't even have anything to do with what's going on there right now. It's all his father (Bush Sr.) working through the Commander in Theif's cabinet. I'll ask you this: IF George Bush Sr. had totally removed Saddam and the regime in Iraq the first time around, do you think that would have prevented Sept. 11th? Hmm??? I didn't think so.
"Face your fears NOW, because we are all doomed to confront them eventually." - Me
I'm an American too, but I am not proud of this country on many counts (but I'll leave that for another discussion). The fact is, the United States is the adolecent country in a world of middle-agers (Europe) and seniors (Asia). Think about the personality of each region (not their people, but the region itself) in world affairs. The US has only 200+ years of experience with being an entity, whereas the others have thousands. The US is currently throwing a temper tantrum at the rest of the world because it doesn't want to lose control over oil. (Iraq as a target has nothing to do with "terrorism" and everything to do with a Bush family grudge.) Think about the way our country acts overall. We are "the best", "the coolest", "the trendiest consumers", "indestructable" etc... (As a culture we emphasize youth especdially when it's profitable) How many mature adults really think this way about themselves and express it so outwardly? A mature adult silently revels in the fact that they have "been there" and "done that" and waits for the adolecent to eventaully ask for some help. When the 9/11 tragedy occurred, it was a wake up call letting our citizens know that we aren't indestructible. This could have been the start of young adulthood if we took the lesson seriously. But it appears that we didn't. Sadly, we probably have a few more lessons in store for us before the US really starts growing up and joining the rest of the world. We'll always be the youngest though. (Not always a bad thing)
I've noticed this as well. Having to run the artsdsp command in a shell every time I want to use a non-KDE app is not very friendly. A combination of the best of GNOME and KDE would be excellent. To determine what is best though... ask users, not developers.
Boxen-(n) Plural of box. Refers to multiple computers. Derived from the term of endearment for a computer "box" combined with the plural form of ox.
A play on words with relevance to the concept of a server farm. In the same way that you can have your farm filled with oxen, you can have your server farm filled with boxen.
If you don't understand this, you are a humorless gonad with subhuman mental abilities. Go back under your lonely bridge troll. Now...
I have been corrected it seems. Gnome IS included in SuSE 8.1. My experience was with 8.0 and Gnome was not given as a default option (although I haven't seen it, I would guess this is true of SuSE 8.1 since they seem to be KDE centric). I, *personally*, find KDE too limiting. It's an environment for doing the most basic things posible on a computer, not creating art and expressing one's individuality. I never said I hated KDE, it just doesn't suit my needs.
For the guy who suggested that there is no need to log out of a machine... you must not use your machine to it's fullest capacties much. The need to log out is the very nature of a multiuser OS. Machines are meant to be shared. My wife and I as well as my friends each have our own accounts on my network. Thus the need to log in. Why not make that experience more enjoyable rather than presenting a dull interface like XDM.
For the short-sighted person who complains about themes taking away CPU cycles: Like you're really going to notice that much of a difference. I run pretty heavy eye-candy in Enlightenment and Gnome 2 with no noticeable difference in performance vs. twm. If you have at least a P III, this stuff does not impact you in the least. If you don't have a P III or higher, I can understand your gripe.
I just wish you KDE loving folks would settle down a little. (I'm sure there are KDE users out there who are perfectly happy and have no need to fiercely defend their choice) The KDE environment is not bad and I never said it was. It's great for certain types of people (people who use their computer for basic work). All I was saying was that for those of us who enjoy a little more aesthetic flexibility, KDE is probably not the best choice. The fact that you guys get so bent out of shape only makes it look like the KDE camp is extremely sensitive to criticism. Here's a few bones: KDE has some great applets. The built in software synth is really cool and DOES address the creative factor with regards to music. Gnome really needs to improve in that arena and I'm sure they will. The fact that KDE comes with a music sequencer as a standard component puts it above everyone else where music is concerned.
I am primarily an artist (sight and sound) who happens to code a little and loves working with computers. For someone like me, the computer is not just a boring tool to get work done. It is THE end. It can be THE work of art. Coding is art. Writing music is art. Creating a decent desktop theme is art. So for someone like *me*, KDE only addresses the musical side, but not the visual side.
Once again, I should remind all of you that this stuff is highly subjective. Just like I hate bands like Train, Dashboard Confessional and Alien Ant Farm, and I love all things electronica (ATB, Tiesto, Oakenfold), some of you are bound to feel the same about the aesthetics of KDE vs. Gnome. And there is also inevitably the group that doesn't care about aesthetics and concentrates solely on just doing their work. None of these views is wrong, they just indicate different preferences. My original post was directed at people who are like me. I tried to specify this by stating that if you use your computer in a certain way, that you may not like SuSE. Anyway, this is getting way off topic, but since it's apparently a fairly inflammatory subject I'll close with this OPINION: KDE is decent, but it doesn't address the needs of someone like me. Gnome 2 does. If SuSE 8.1 will allow for Gnome 2 to be the default environment with KDE taking a backseat, then it would probably satisfy me. Otherwise, it won't.
To be fair, my experience with KDE stopped at 3.0. I just found that it didn't do a lot of what I like about Gnome. But this is all personal preference. I was only throwing out a warning for others like me. I lived with KDE 3.0 for about six months at work. No I'm trying Windows XP. But I still like Gnome.
I have to agree with the reviewer's sentiment that they should have included Gnome in SuSE. While KDE is pretty good for newbies from the Windows and Mac world, it's still missing the eye-candy that even basic Gnome 2.x has. Of course this is all my opinion, and highly subjective. But, I will say that if you happen to be a fan of the Gnome environment, you're going to feel a bit restricted by KDE. Some major features are missing in KDE:
-themeable login manager -flexible bitmap themes that allow you to tweak window behavior into something that you want -granular control over the look-and-feel of the environment (multiple toolbars, drag and drop launchers, etc...) -a more standarized approach to where binaries go: '/usr/local/bin' rather than '/opt/kde' (Of course it would be better if things were more like '/usr/local/kde', but thene again I compile everything I use.)
If the only thing you do with your computer is read e-mail, browse the web, word processing, and balancing your checkbook, then KDE should fit nicely. But if you like to express yourself creatively and customize your system for ease of use, KDE is not going to make you too happy.
RedHat's Blue Curve approach is probably a little stronger than SuSE's version of KDE. I've only seen screenshots, but it's much prettier.
What about borrowing the concept of the "Installation Wizard" from that OTHER OS? Not a package manager (like RPM, apt-get, etc) that's included with the distribution, but a single installation program that includes the source and related libraries as one downloadable binary. Yes, the file would be much bigger, but this is not for the hardcore Linux user, it's for the average user. The way it would work is like this:
Distributor: 1. The distributor takes the source, and includes a single level of dependencies (libs other apps). There is also a list of higher level dependencies in the wizard as well in case the system that this installer is being run on doesn't have them. 2. The distributor runs the wizard generator to put the source, libraries and list together along with any graphical niceties (branding, Tux images, the GNU mascot, etc...), license text, and scripting for generating instuctional messages.
User: 1. The user downloads the binary installer and executes it. 2. They are instructed to "click next" unless they wish to set their own preferences, etc... 3. If any of the higher level dependencies are missing, the list is consulted and a central download site (preferrably the location of the base tarball) is connected to for source retrieval. 4. If there is no connectivity, a note indicating missing dependencies and with the proper installer's URL is presented to the user with an option to bookmark (or maybe a submission to a download manager) in their chosen browser for later download. 5. If any of the first level dependencies are missing, the sources are unpacked from the installer, compiled and then installed. 6. The application is then compiled and installed. 7. Program launchers (and uninstall scripts for each app) are dropped into a common location that the various GUI environments can poll for new apps. 8. Insert uninstall info into the common package manager so that when the user wants to uninstall, they can still use the package manager.
Perhaps some of the options during installation can be things like:
-Install in the user's directory under a "Programs" folder for just that user. -Advanced options like CPU arch selection, compile with statically linked libs, any optimization options, etc... (For the low-level power user, assuming that the newbies start getting more interested in learning)
This doesn't change the current scheme of dependencies, but it hides it very well from the user. As far as they are concerned (on most systems) the installer is the only thing they need. Again, assume this is Joe User who hasn't upgraded his gcc, libs, or anything.
The package managers are nice, but they are certainly confusing to the average user. Especially when they start telling you that you need to download other stuff. Put as much as you can into one installer with a common interface, and you can minimize "dependency hell".
You are right. Geeks like to learn new things. But... Joe User will want to learn something new if it's better than the competing product. Lindows isn't there yet because Joe User doesn't care about stability and security (which Lindows appears to be lacking since it runs as root by default). In general, there aren't any Linux distribs that off the average user any tangible advantages beyond Windows. Windows XP DOES offer tangible (but to some pointless) advantages:
1. The ability to have multiple users logged in with applications running for each one. (Stateful sessions) 2. A VERY simple user interface. GDM was close with the photo browser, but XPs login screen is just a simple button click to the desktop. On top of that, each session icon indicates whether a user is logged in with apps running or not. 3. Clear and consistent status indicators (as a task tray app) for things like IM, Wireless NIC, Wired NIC, etc. As well as the OS hiding the least used task tray apps in a simple hide button. (Similar to GNOME's panel hide buttons, but within the task tray area) 4. The five or six most commonly accessed apps show up as the first apps in the Start bar. This order changes depending on how often you use and app. How hard can this be to implement in GNOME or KDE?
The point is that Windows XP offers benefits that a user can see over Windows 2000/9x AND any Linux distrib. I have confidence that the Linux distribs will catch up, but that's not enough. The real goal should be to offer USERS (especially in a distrib like Lindows) something that they can actually percieve as a benefit over Windows. Some ideas:
- Using SSHD and SSH as a backend, how about automatically encrypting all WLAN traffic and having this be set up from a simple GUI control panel. In fact GUI front ends to ALL openssh functionality would be awesome. And SIMPLE, not complicated. Why couldn't the key files be used in a similar fashion to the Security Dynamics RSA soft tokens? Setting up encrypted tunnels with a simple radio button app with options like "Forward web traffic from a remote location to this computer" or "Forward remote desktop (vnc) to this computer".
- How hard could it be to check for the most recently used applications and just start putting then on a Gnome panel or set up KDE launchers after X number of executions?
- Proxy root access. Gnome sort of does this when it prompts you for the root password to access certain administrative tasks. But, this is not organized enough. Why not group all administrative tasks that require root access into one front end app that any user can run. When IT is launched, they are prompted for the root password once, then they can do any of those tasks without having to type the root PW again and again.
- A GUI based 'su'. Windows XP has "Run As" now which basically does the same thing as 'su'. BUt make it better in Linux. Just have a K menu or Gnome menu "Run Command with..." item with a radio button to select between "Current user priveleges" or "Run with administrative priveleges". This obviates the need for a "Run As" since it would be availabel to any command a user would type. Or... how about having the middle click on a launcher run the app with root privs, but prompt for the password first. I have run into several instances where I wanted to run Nautilus, GIMP, Mozilla, etc... as root.
The whole point is to make Linux distribs better overall. I think that can only be done by taking advantage of what Linux does best and using those capabilities in more ingenious ways that a USER WILL percieve.
My experience with RedHat 7.1 and 7.2 on the Compaq wasn't horrible, but it wasn't great either. I haven't tried 7.3 yet, so I can't say anything about it, but I do hope that WLAN support improves. A friend of mine with a fairly new HP latptop (2001 model) tried SuSE 8.0 on it. He is a Windows guy who wants to get into Linux, so he has some stuff to learn. But, the things that didn't work out of the box for him were WLAN and CD-R/CD-RW burning as well as DVD playback. I had to recompile his kernel to add support for DVD (ufs). But, I couldn't get the WLAN stuff to work (ORINOCO based). (I also couldn't get his CD/RW drive to work either.) The WLAN card not working seemed to have less to do with the WLAN driver than the PCMCIA driver. Oddly enough his NIC and modem (also PCMCIA worked.
In my case, I got wireless working on my Compaq, but I had to recompile the kernel as well as PCMCIA support and the orinoco driver. Once it was working, it wasn't totally reliable. It would pick up signal when it felt like it whic was about 20% of the time, and there was no GUI applet to let me know that there was a signal. I could use iwconfig, but I have to keep refeshing it every second to see if there is a signal or not. Under XP with the identical hardware, the signal comes in immediately no matter where I am in the house. The only thing I can chalk it up to is that the drivers for orinoco are still in development. In my friend's case, it was a PCMCIA problem. Unfortunately, he re-installed XP because he couldn't wait for me to get some free time to look at his SuSE installation. His opinion of SuSE was, "Almost there... but needs a little more work. It should be able to support standard (hah!) hardware right out of the box."
So far, my XP laptop has been running for a few days with no problems. I've even been able to keep the uptime going with suspend. I know it's possible to set up suspend in Linux too, but it doesn't happen out of the box with RedHat. So uptimes in Linux on the same laptop were never very high. I have yet to see XP crash on my laptop though... Maybe after running with it for a few months, it'll happen. (Remember, I'm primarily a Linux dude! I just have an open mind...) Haven't tried DVD under XP, but it probably works too. I used Ogle on that laptop before and it played DVDs just fine, so I don't expect much of a difference. MPlayer ruled as well, but I only used it for MPEGs and DiVX (nothing in the Windows world can compete with MPlayer)
As far as "work environment", I'm a net admin by day, so I don't have a whole lot of use for devel tools, etc... All I need is a mail client, web browser, help desk management software and NT admin tools. There are plenty of things I wish Windows had (I did install Cygwin so I could use the X server in XP), but to do my day to day stuff, Windows XP works just fine. A lot of my peers still prefer the classic mode, but I don't. I like the new XP mode since it's very logical and puts pretty much everything I need right at my fingertips for a day's work. I have been using Linux at work for the past two years with Win2K in VMWare for managing the NT domain, but XP really hasn't been that bad. As I said before, there's a lot less to tweak though. And I hate the licensing. I think it's braindead to tie the OS to the machine for life.
I know I'll go back to Linux at work within the next few months because I still tire of Windows in general, but I have to take a look and see where things are in the other camp from time to time. I think a lot of you who are totally ignoring Windows XP are still thinking it's Windows 95. Well... it isn't. But don't let me tell you that, just take a deep look at it yourself if you can. You won't love it (I don't), but you will find yourself thinking that there are some good features there that really need a counterpart in Linux. If you really look at it with an objective mindset that is. Think of it this way, when you made your transition from Linux to Windows, some of you must have felt that "oh... it doesn't do X" feeling. You'll feel the same way if you try Windows XP. Especially if you've been avoiding using Windows on a regular basis for as long as I have. As TurdFerguson said in another post on another day, "Know your enemy". That's what a lot of us have been avoiding.
Here... here's a little troll snack: Don't let your inability to learn newer, streamlined interfaces obscure your judgement of an improved OS. Windows XP Pro is easier to install, use and configure than Windows 2K. It's too bad that it's not fair to consumers and has DRM all over it. Linux distros will have a little catch up to do again. But... this is a constant state in the OS market (the software industry overall). One side is always slightly ahead of the other in some way. Linux still has stability and security, but MS will catch up there too. That's why distro makers should focus on end user things like usability, eye-candy, "geewhiz" stuff, etc... Who can refute that there is a need for something like the stateful sessions that Windows XP has in the Linux world? Go back under your bridge now troll.
Hmmm... Before you misread me as another MS shill, I will state up front that I am a big Linux supporter/user. All my systems at home run Linux only. And all of them are pretty much built by me from the ground up with custom compiled kernels and apps. But, I've been using XP at work and on my laptop for the past month. I have to say, regretfully, that MS got a LOT of things right in this version of Windows.
My laptop is a Compaq Armada D500 (PIII 600/w 128Megs of RAM). The system seems to run a lot faster under XP than it did under RedHat 7.2 or SuSE 8.0. Even compared to when I had a custom compiled kernel and apps on it. The wireless PCMCIA worked with no need to grab drivers (my Windows 2000 experience on this laptop) or recompile anything. All apps load quickly. The suspend feature works exactly as expected. The environment is much more organized and task oriented. Etc... The bottom line is that Linux distro makers can't rest on the old laurels (Linux is more stable, secure, you can tweak the code, etc...) and ignore the MS camp. Take a look at what Windows is today. I mean a REAL look. Most of today's Linux distro's are great alternatives to Windows 2k, but they leave something to be desired when compared with Win XP.
I have been forcing myself to use XP here at work for the past month and it really does blow most Linux distros away in terms of a basic work environment. The only problem I've had so far is that I can't get under the hood and tweak as much, but I haven't found that there is a need to either...
As far as your experience goes, what make/model of laptop were you trying to install Windows XP on? That could be the key to understanding why it didn't work.
I still won't use it at home, not because it isn't as good as Linux, but because I can't afford it and the licensing sucks. Joe User doesn't think that way though...
Neither view has been proved or disproved. This means there can be other explanations of origin. Since I personally don't care about either view or find either one satisfactory, I just choose to ignore both. But labeling someone who doesn't believe in evolution (or it's associated concepts) a fool or saying they are unscientific is just as bad as labeling someone who doesn't believe in creationism a heretic and a blasphemer. The creationist have chosen their religion, and you have chosen yours apparently. Log in, so your part in this discussion isn't set down to 0.
I agree wholeheartedly. Evolution and "Intelligent Design" are not the only ways to explain the origins of living things. For instance, why does the "creator" in the intelligent design model HAVE to be "God"? What if some ancient race from another part of the universe decided to experiment with creating new forms of life and chose to do it here? Since we currently have no true understanding of who or what we are, how can we say this didn't happen?
That example aside, my point is that evolution can't be the only answer since there is as much a lack of data there as there is for proving intelligent design. There may be more data to support it than the creationists have, but it's still not complete. Until all the data is in, it remains a theory in the traditional sense of the word: an unproven hypothesis.
When you exist in a universe that is so much larger that you or your ego are, you need to accept that fact that there are things that you are never going to understand or see. You aren't large enough to see the whole picture. If you refute this, then you only comfort yourself in your own ignorance and will never develop futher in either intelligence or courage. Being small and less important can be scary.
You are correct sir, I hadn't thought about endianess and I am not sure what AltiVec does other than provide MMX-like features that perform better than MMX.:)
Here is (very simplified) the way the layers to get apps to run on OS X work, starting at the bottom:
Darwin Kernel (BSD Licensed)
|
Cocoa, Carbon (API) application libraries (Closed)
|
Aqua/Apps (Aqua and most apps-Closed)
The Kernel (as always) only provides the system calls from APIs to the underlying hardware.
The APIs are what stand between the apps and the kernel. The killer app here is Aqua.
Aqua stands between the user and the other apps in the system (Finder, the Launchbar, iMovie, etc...)
So... you can run the kernel on x86, but you don't have any ability to recompile the GUI based apps to run even if the application source is available. What you would need would be an open source re-implementation of Cocoa, Carbon, Aqua, etc... Kinda-sorta like W.I.N.E. Probably with about the same level of compatibility. Hope this clears it up a bit...
Fooling OS X into thinking that it's running on a Motorola chip wouldn't be effective either. You run into the same problems you did with old Apple emulation... There are chips in the original hardware that provide things that are usually in libs on other platforms. You would have to emulate Mac motherboard, chipset and proc to get this to work. That WOULD be difficult and unsightly. Not impossible though. Just impractical.
Regarding the potential port of Mac OS X to Intel, I won't say it will never happen. I think all it would take is for Apple's new servers to actually take off. Their greatest selling point is that they don't require a license for each client that connects and they provide decent support for Windows clients. If they get enough market share in the server market, and they can sell the OS without having to rely on the hardware, then there is an open door to doing a port. I think this is why Steve Jobs said that AFTER the migration to OS X, anything is possible. Of course I can't see Steve Jobs himself approving a switch to x86 Intel. 4th gen Itanium? Maybe. At least there, you're dealing with what used to be Alpha to a certain extent. I think Steve Jobs still doesn't like dealing with the companies that he percieves as being "inferior", and I applaud him for that. Only the best things come out of clear direction and tight control: Macintosh, MPlayer, the Linux Kernel, etc...;)
Just like dishing out catty comments like the above has always been the realm of fag0rtz. Pot. Kettle. Black... You must be a homo too. Stop posting anonymously and come out of the closet, it'll make your life so much better.
OK... totally off-topic here, but I wanted to respond to this with some thinking I've been doing lately:
Interesting point that you make at the end: "This is an education problem not an operating system problem." But, this is wrong. The real answer is that computers are still too hard to use. You are applying this towards the admin end of things, but I am going to shift to the users in a little bit.
As far as the admins go, I think we need to work towards a time when there is no need for an admin. WHAT!!? Heresy you say? No, practicality. Technicians that repair/swap out broken hardware should be around, but OSes and networks should pretty much run themselves.
The original automobiles were hard to use and generally operated by people with a sharper intellect and an ability to work on their cars. Today, any monkey can (almost) operate a car with a minimum of education. And when it breaks down, they take it to a technician to fix the problem. The end-user does not need to know anything about how it operates to get it to move and take them places. That's the direction computers should go in, from both the admin and user perspectives.
Again... the car example: you don't see people getting certified to drive different cars by make, do you? Going from one OS to another should not be as difficult as it is. There are many reasons for the difficulty: monopolies putting up roadblocks, open source programmers who aren't writing for Joe Average, an outdated platform (x86, ISA, PCI...), etc...
As far as users go, I think that we expect too much of them. Even in the Windows world, things that should be simple procedures are difficult if you don't have the "right" software.
Over the weekend, my dad asked me how he could record himself onto a CD with his burner (in Windows). I thought... Easy! I spent the better part of the afternoon trying to show him how to use his limited edition of Roxio and Audacity in conjunction to do this. (He can't afford to buy anything better and he can't use Linux) It turned out not to be so easy after all. The concept of using two different apps to access/create data for each other is an alien and confusing concept to most people since they don't get the abstraction and have problems thinking in a hierarchical fashion. My dad was just hoping to be able to plug in a mic, press record and get his voice on a CD. He didn't want to mess with creating folders and files and editing his recordings. Sure, there are some programs that would keep the user in one application throughout the process, but it's still alien to them. This is why I still say computers are not easy to use.
Just like cars, there will always be a small population of people who understand computing devices pretty thoroughly, and there will always be others who just want to turn the key and drive. So, it's not an education issue, it's a platform, OS, application issue. Ditch the old way of thinking and look at it this way... Someday computing should be happening all around us without any need for us to directly interact. The entire world around us should be one big user interface. Forget "desktop" environment and think about getting computing into the physical realm... Then we won't have what appears to currently be an education problem. And... if we apply this thinking to open source projects now, we'll be right on track with where the world is going.
Just a loose set of thoughts with regard to this article. I don't have time to arrange them into something more cohesive. I use the term Linux to generically apply to complete distros with a desktop environment:
The quote from Ballmer at the very end of the article may be a harbinger of things to come: software is not consistently profitable. The very fact that Ballmer considers it weird that IBM would tell a company to buy software from someone else indicates that the "playing field" is changing. Sure, IBM isn't at the top of the game anymore, but I think you may start to see more and more companies abandoning the software business for more profitable fields like embedded devices and other dedicated systems that we haven't yet dreamed of. The whole problem with computers right now is that people actually have to "interface" with them in non-intuitive ways. But that's a different topic...
In Neal Stephenson's "In the Beginning there was Command Line", he says that it is the fate of software to become free. Commercial UNIX gave way to free UNIX, Microsoft Word now has a respectable challenger in OpenOffice.org, etc... Or at a deeper level (the concept level as opposed to the product level), GUIs have become so inexpensive compared to the original Xerox systems that some are free: XFree86 + GNOME or KDE. I believe his observations are correct. The OS market will continue to become less profitable if the "movers and shakers" aren't always looking for the next "great thing".
The only thing propping up Microsoft right now is the Office suite and to some extent Internet Explorer. To take this crutch down would only require the provision of a application that uses a totally new and better approach to achieving the same results. No one has done it yet. But again, I digress... (;) )
My point is... that Ballmer's comment about "Added Value" above Linux should really be about finding the next "killer app" that Windows has and Linux does not. This ensures that more people who follow that path of least resistance will choose Windows every time.
These victories are short-lived however. As soon as a concept is out in the open, it's fate is to have reproductions and innovations built around it. Witness: Apple popularizes the GUI that Xerox couldn't move. Microsoft immediately responds with their first release of Windows. Mosaic begat Netscape who begat Internet Explorer... (at the concept level, not the business/profit level).
Look at the music industry. In 1994 the Spice Girls came on the scene and were hugely successful (opinions about their music aside). So what happened immediately after that? Knock offs. Tons of them. None with a chance of making it as big as the original, even if the original was not as good as the newer acts. To a certain extent, this happens in the GNU\Linux\Open Source world more than it should. But, undeniably, there are some ideas that just can't be improved on. So, what do we do? Look ahead and occasionally check the other runners next to you. When I say look ahead, I mean look for new approaches at the user level not the system level. These are real differences that the user can see, feel and experience. Of course, this is assuming that you are interested in moving Linux out to "Joe Average".
Microsoft can't outsmart "Linux" since there isn't any one model to take down without some heavy handed help from the governments of the world. At the moment, they aren't doing to well in that arena either... Linux will be around until something better comes along. That "something better" has to be completely different compared to Linux and provide features that Linux doesn't have. However, it should also still be free. That is where Microsoft will never be able to compete.
Do you think that "Joe Computer Geek" is going to be able to get his hands on this to stream his own DVDs? Probably not. There will probably be some sort of DRM built in because the MPAA (as well as the RIAA) is too busy focusing on a few potential lost sales vs. the big revenues that could be had if they just opened their stuff up to internet distribution. They are looking at everything through an outdated selling concept. Not everyone thinks this way though... Peter Gabriel has his entire new album (UP) available to listen to in a streamable format as well as the video for his first single. The quality is low, so it encourages people to buy the real deal, but it's the entire album, so it allows for "try before you buy". The same could be applied to DVD pre-release and this technology would be great for it. But, it's still not going to be something that you or I can legitimately use to stream our own DVDs unless there are a LOT of restrictions. I for one am no longer sure of the legality of me streaming my MP3s to myself at work with icecast and not paying the RIAA those stupid broadcaster's fees. Discuss amongst yourselves.
Whoa! Can someone mod this guy down below -1? That was intolerable. ;P
Someone please mod the parent up as +3 Funny. This was just too damn funny.
I should have been more specific. All of the "terrorism" that has been "averted" since 9/11/2001 has nothing to do with Iraq. The man in the Whitehouse who does NOT represent me doesn't even have anything to do with what's going on there right now. It's all his father (Bush Sr.) working through the Commander in Theif's cabinet. I'll ask you this: IF George Bush Sr. had totally removed Saddam and the regime in Iraq the first time around, do you think that would have prevented Sept. 11th? Hmm??? I didn't think so.
"Face your fears NOW, because we are all doomed to confront them eventually." - Me
I'm an American too, but I am not proud of this country on many counts (but I'll leave that for another discussion). The fact is, the United States is the adolecent country in a world of middle-agers (Europe) and seniors (Asia). Think about the personality of each region (not their people, but the region itself) in world affairs. The US has only 200+ years of experience with being an entity, whereas the others have thousands. The US is currently throwing a temper tantrum at the rest of the world because it doesn't want to lose control over oil. (Iraq as a target has nothing to do with "terrorism" and everything to do with a Bush family grudge.) Think about the way our country acts overall. We are "the best", "the coolest", "the trendiest consumers", "indestructable" etc... (As a culture we emphasize youth especdially when it's profitable) How many mature adults really think this way about themselves and express it so outwardly? A mature adult silently revels in the fact that they have "been there" and "done that" and waits for the adolecent to eventaully ask for some help. When the 9/11 tragedy occurred, it was a wake up call letting our citizens know that we aren't indestructible. This could have been the start of young adulthood if we took the lesson seriously. But it appears that we didn't. Sadly, we probably have a few more lessons in store for us before the US really starts growing up and joining the rest of the world. We'll always be the youngest though. (Not always a bad thing)
I've noticed this as well. Having to run the artsdsp command in a shell every time I want to use a non-KDE app is not very friendly. A combination of the best of GNOME and KDE would be excellent. To determine what is best though... ask users, not developers.
Boxen-(n) Plural of box. Refers to multiple computers. Derived from the term of endearment for a computer "box" combined with the plural form of ox.
A play on words with relevance to the concept of a server farm. In the same way that you can have your farm filled with oxen, you can have your server farm filled with boxen.
If you don't understand this, you are a humorless gonad with subhuman mental abilities. Go back under your lonely bridge troll. Now...
For a desktop distribution that expects to reach out to Joe User, a GUI is NOT optional. It should be integrated into the OS very tightly.
I have been corrected it seems. Gnome IS included in SuSE 8.1. My experience was with 8.0 and Gnome was not given as a default option (although I haven't seen it, I would guess this is true of SuSE 8.1 since they seem to be KDE centric). I, *personally*, find KDE too limiting. It's an environment for doing the most basic things posible on a computer, not creating art and expressing one's individuality. I never said I hated KDE, it just doesn't suit my needs.
For the guy who suggested that there is no need to log out of a machine... you must not use your machine to it's fullest capacties much. The need to log out is the very nature of a multiuser OS. Machines are meant to be shared. My wife and I as well as my friends each have our own accounts on my network. Thus the need to log in. Why not make that experience more enjoyable rather than presenting a dull interface like XDM.
For the short-sighted person who complains about themes taking away CPU cycles: Like you're really going to notice that much of a difference. I run pretty heavy eye-candy in Enlightenment and Gnome 2 with no noticeable difference in performance vs. twm. If you have at least a P III, this stuff does not impact you in the least. If you don't have a P III or higher, I can understand your gripe.
I just wish you KDE loving folks would settle down a little. (I'm sure there are KDE users out there who are perfectly happy and have no need to fiercely defend their choice) The KDE environment is not bad and I never said it was. It's great for certain types of people (people who use their computer for basic work). All I was saying was that for those of us who enjoy a little more aesthetic flexibility, KDE is probably not the best choice. The fact that you guys get so bent out of shape only makes it look like the KDE camp is extremely sensitive to criticism. Here's a few bones: KDE has some great applets. The built in software synth is really cool and DOES address the creative factor with regards to music. Gnome really needs to improve in that arena and I'm sure they will. The fact that KDE comes with a music sequencer as a standard component puts it above everyone else where music is concerned.
I am primarily an artist (sight and sound) who happens to code a little and loves working with computers. For someone like me, the computer is not just a boring tool to get work done. It is THE end. It can be THE work of art. Coding is art. Writing music is art. Creating a decent desktop theme is art. So for someone like *me*, KDE only addresses the musical side, but not the visual side.
Once again, I should remind all of you that this stuff is highly subjective. Just like I hate bands like Train, Dashboard Confessional and Alien Ant Farm, and I love all things electronica (ATB, Tiesto, Oakenfold), some of you are bound to feel the same about the aesthetics of KDE vs. Gnome. And there is also inevitably the group that doesn't care about aesthetics and concentrates solely on just doing their work. None of these views is wrong, they just indicate different preferences. My original post was directed at people who are like me. I tried to specify this by stating that if you use your computer in a certain way, that you may not like SuSE. Anyway, this is getting way off topic, but since it's apparently a fairly inflammatory subject I'll close with this OPINION: KDE is decent, but it doesn't address the needs of someone like me. Gnome 2 does. If SuSE 8.1 will allow for Gnome 2 to be the default environment with KDE taking a backseat, then it would probably satisfy me. Otherwise, it won't.
To be fair, my experience with KDE stopped at 3.0. I just found that it didn't do a lot of what I like about Gnome. But this is all personal preference. I was only throwing out a warning for others like me. I lived with KDE 3.0 for about six months at work. No I'm trying Windows XP. But I still like Gnome.
I have to agree with the reviewer's sentiment that they should have included Gnome in SuSE. While KDE is pretty good for newbies from the Windows and Mac world, it's still missing the eye-candy that even basic Gnome 2.x has. Of course this is all my opinion, and highly subjective. But, I will say that if you happen to be a fan of the Gnome environment, you're going to feel a bit restricted by KDE. Some major features are missing in KDE:
-themeable login manager
-flexible bitmap themes that allow you to tweak window behavior into something that you want
-granular control over the look-and-feel of the environment (multiple toolbars, drag and drop launchers, etc...)
-a more standarized approach to where binaries go: '/usr/local/bin' rather than '/opt/kde' (Of course it would be better if things were more like '/usr/local/kde', but thene again I compile everything I use.)
If the only thing you do with your computer is read e-mail, browse the web, word processing, and balancing your checkbook, then KDE should fit nicely. But if you like to express yourself creatively and customize your system for ease of use, KDE is not going to make you too happy.
RedHat's Blue Curve approach is probably a little stronger than SuSE's version of KDE. I've only seen screenshots, but it's much prettier.
What about borrowing the concept of the "Installation Wizard" from that OTHER OS? Not a package manager (like RPM, apt-get, etc) that's included with the distribution, but a single installation program that includes the source and related libraries as one downloadable binary. Yes, the file would be much bigger, but this is not for the hardcore Linux user, it's for the average user. The way it would work is like this:
Distributor:
1. The distributor takes the source, and includes a single level of dependencies (libs other apps). There is also a list of higher level dependencies in the wizard as well in case the system that this installer is being run on doesn't have them.
2. The distributor runs the wizard generator to put the source, libraries and list together along with any graphical niceties (branding, Tux images, the GNU mascot, etc...), license text, and scripting for generating instuctional messages.
User:
1. The user downloads the binary installer and executes it.
2. They are instructed to "click next" unless they wish to set their own preferences, etc...
3. If any of the higher level dependencies are missing, the list is consulted and a central download site (preferrably the location of the base tarball) is connected to for source retrieval.
4. If there is no connectivity, a note indicating missing dependencies and with the proper installer's URL is presented to the user with an option to bookmark (or maybe a submission to a download manager) in their chosen browser for later download.
5. If any of the first level dependencies are missing, the sources are unpacked from the installer, compiled and then installed.
6. The application is then compiled and installed.
7. Program launchers (and uninstall scripts for each app) are dropped into a common location that the various GUI environments can poll for new apps.
8. Insert uninstall info into the common package manager so that when the user wants to uninstall, they can still use the package manager.
Perhaps some of the options during installation can be things like:
-Install in the user's directory under a "Programs" folder for just that user.
-Advanced options like CPU arch selection, compile with statically linked libs, any optimization options, etc... (For the low-level power user, assuming that the newbies start getting more interested in learning)
This doesn't change the current scheme of dependencies, but it hides it very well from the user. As far as they are concerned (on most systems) the installer is the only thing they need. Again, assume this is Joe User who hasn't upgraded his gcc, libs, or anything.
The package managers are nice, but they are certainly confusing to the average user. Especially when they start telling you that you need to download other stuff. Put as much as you can into one installer with a common interface, and you can minimize "dependency hell".
You are right. Geeks like to learn new things. But... Joe User will want to learn something new if it's better than the competing product. Lindows isn't there yet because Joe User doesn't care about stability and security (which Lindows appears to be lacking since it runs as root by default). In general, there aren't any Linux distribs that off the average user any tangible advantages beyond Windows. Windows XP DOES offer tangible (but to some pointless) advantages:
1. The ability to have multiple users logged in with applications running for each one. (Stateful sessions)
2. A VERY simple user interface. GDM was close with the photo browser, but XPs login screen is just a simple button click to the desktop. On top of that, each session icon indicates whether a user is logged in with apps running or not.
3. Clear and consistent status indicators (as a task tray app) for things like IM, Wireless NIC, Wired NIC, etc. As well as the OS hiding the least used task tray apps in a simple hide button. (Similar to GNOME's panel hide buttons, but within the task tray area)
4. The five or six most commonly accessed apps show up as the first apps in the Start bar. This order changes depending on how often you use and app. How hard can this be to implement in GNOME or KDE?
The point is that Windows XP offers benefits that a user can see over Windows 2000/9x AND any Linux distrib. I have confidence that the Linux distribs will catch up, but that's not enough. The real goal should be to offer USERS (especially in a distrib like Lindows) something that they can actually percieve as a benefit over Windows. Some ideas:
- Using SSHD and SSH as a backend, how about automatically encrypting all WLAN traffic and having this be set up from a simple GUI control panel. In fact GUI front ends to ALL openssh functionality would be awesome. And SIMPLE, not complicated. Why couldn't the key files be used in a similar fashion to the Security Dynamics RSA soft tokens? Setting up encrypted tunnels with a simple radio button app with options like "Forward web traffic from a remote location to this computer" or "Forward remote desktop (vnc) to this computer".
- How hard could it be to check for the most recently used applications and just start putting then on a Gnome panel or set up KDE launchers after X number of executions?
- Proxy root access. Gnome sort of does this when it prompts you for the root password to access certain administrative tasks. But, this is not organized enough. Why not group all administrative tasks that require root access into one front end app that any user can run. When IT is launched, they are prompted for the root password once, then they can do any of those tasks without having to type the root PW again and again.
- A GUI based 'su'. Windows XP has "Run As" now which basically does the same thing as 'su'. BUt make it better in Linux. Just have a K menu or Gnome menu "Run Command with..." item with a radio button to select between "Current user priveleges" or "Run with administrative priveleges". This obviates the need for a "Run As" since it would be availabel to any command a user would type. Or... how about having the middle click on a launcher run the app with root privs, but prompt for the password first. I have run into several instances where I wanted to run Nautilus, GIMP, Mozilla, etc... as root.
The whole point is to make Linux distribs better overall. I think that can only be done by taking advantage of what Linux does best and using those capabilities in more ingenious ways that a USER WILL percieve.
From the article:
> "Imagine trying to communicate with another
> human being using just a mouse and a keyboard.
> It works, but it is slow and tedious."
Tell me about it. I post to Slashdot, write e-mail, IRC and IM all the time. What else is new?
My experience with RedHat 7.1 and 7.2 on the Compaq wasn't horrible, but it wasn't great either. I haven't tried 7.3 yet, so I can't say anything about it, but I do hope that WLAN support improves. A friend of mine with a fairly new HP latptop (2001 model) tried SuSE 8.0 on it. He is a Windows guy who wants to get into Linux, so he has some stuff to learn. But, the things that didn't work out of the box for him were WLAN and CD-R/CD-RW burning as well as DVD playback. I had to recompile his kernel to add support for DVD (ufs). But, I couldn't get the WLAN stuff to work (ORINOCO based). (I also couldn't get his CD/RW drive to work either.) The WLAN card not working seemed to have less to do with the WLAN driver than the PCMCIA driver. Oddly enough his NIC and modem (also PCMCIA worked.
In my case, I got wireless working on my Compaq, but I had to recompile the kernel as well as PCMCIA support and the orinoco driver. Once it was working, it wasn't totally reliable. It would pick up signal when it felt like it whic was about 20% of the time, and there was no GUI applet to let me know that there was a signal. I could use iwconfig, but I have to keep refeshing it every second to see if there is a signal or not. Under XP with the identical hardware, the signal comes in immediately no matter where I am in the house. The only thing I can chalk it up to is that the drivers for orinoco are still in development. In my friend's case, it was a PCMCIA problem. Unfortunately, he re-installed XP because he couldn't wait for me to get some free time to look at his SuSE installation. His opinion of SuSE was, "Almost there... but needs a little more work. It should be able to support standard (hah!) hardware right out of the box."
So far, my XP laptop has been running for a few days with no problems. I've even been able to keep the uptime going with suspend. I know it's possible to set up suspend in Linux too, but it doesn't happen out of the box with RedHat. So uptimes in Linux on the same laptop were never very high. I have yet to see XP crash on my laptop though... Maybe after running with it for a few months, it'll happen. (Remember, I'm primarily a Linux dude! I just have an open mind...)
Haven't tried DVD under XP, but it probably works too. I used Ogle on that laptop before and it played DVDs just fine, so I don't expect much of a difference. MPlayer ruled as well, but I only used it for MPEGs and DiVX (nothing in the Windows world can compete with MPlayer)
As far as "work environment", I'm a net admin by day, so I don't have a whole lot of use for devel tools, etc... All I need is a mail client, web browser, help desk management software and NT admin tools. There are plenty of things I wish Windows had (I did install Cygwin so I could use the X server in XP), but to do my day to day stuff, Windows XP works just fine. A lot of my peers still prefer the classic mode, but I don't. I like the new XP mode since it's very logical and puts pretty much everything I need right at my fingertips for a day's work. I have been using Linux at work for the past two years with Win2K in VMWare for managing the NT domain, but XP really hasn't been that bad. As I said before, there's a lot less to tweak though. And I hate the licensing. I think it's braindead to tie the OS to the machine for life.
I know I'll go back to Linux at work within the next few months because I still tire of Windows in general, but I have to take a look and see where things are in the other camp from time to time. I think a lot of you who are totally ignoring Windows XP are still thinking it's Windows 95. Well... it isn't. But don't let me tell you that, just take a deep look at it yourself if you can. You won't love it (I don't), but you will find yourself thinking that there are some good features there that really need a counterpart in Linux. If you really look at it with an objective mindset that is. Think of it this way, when you made your transition from Linux to Windows, some of you must have felt that "oh... it doesn't do X" feeling. You'll feel the same way if you try Windows XP. Especially if you've been avoiding using Windows on a regular basis for as long as I have. As TurdFerguson said in another post on another day, "Know your enemy". That's what a lot of us have been avoiding.
Titties!
Here... here's a little troll snack: Don't let your inability to learn newer, streamlined interfaces obscure your judgement of an improved OS. Windows XP Pro is easier to install, use and configure than Windows 2K. It's too bad that it's not fair to consumers and has DRM all over it. Linux distros will have a little catch up to do again. But... this is a constant state in the OS market (the software industry overall). One side is always slightly ahead of the other in some way. Linux still has stability and security, but MS will catch up there too. That's why distro makers should focus on end user things like usability, eye-candy, "geewhiz" stuff, etc... Who can refute that there is a need for something like the stateful sessions that Windows XP has in the Linux world? Go back under your bridge now troll.
Hmmm... Before you misread me as another MS shill, I will state up front that I am a big Linux supporter/user. All my systems at home run Linux only. And all of them are pretty much built by me from the ground up with custom compiled kernels and apps. But, I've been using XP at work and on my laptop for the past month. I have to say, regretfully, that MS got a LOT of things right in this version of Windows.
My laptop is a Compaq Armada D500 (PIII 600/w 128Megs of RAM). The system seems to run a lot faster under XP than it did under RedHat 7.2 or SuSE 8.0. Even compared to when I had a custom compiled kernel and apps on it. The wireless PCMCIA worked with no need to grab drivers (my Windows 2000 experience on this laptop) or recompile anything. All apps load quickly. The suspend feature works exactly as expected. The environment is much more organized and task oriented. Etc... The bottom line is that Linux distro makers can't rest on the old laurels (Linux is more stable, secure, you can tweak the code, etc...) and ignore the MS camp. Take a look at what Windows is today. I mean a REAL look. Most of today's Linux distro's are great alternatives to Windows 2k, but they leave something to be desired when compared with Win XP.
I have been forcing myself to use XP here at work for the past month and it really does blow most Linux distros away in terms of a basic work environment. The only problem I've had so far is that I can't get under the hood and tweak as much, but I haven't found that there is a need to either...
As far as your experience goes, what make/model of laptop were you trying to install Windows XP on? That could be the key to understanding why it didn't work.
I still won't use it at home, not because it isn't as good as Linux, but because I can't afford it and the licensing sucks. Joe User doesn't think that way though...
Neither view has been proved or disproved. This means there can be other explanations of origin. Since I personally don't care about either view or find either one satisfactory, I just choose to ignore both. But labeling someone who doesn't believe in evolution (or it's associated concepts) a fool or saying they are unscientific is just as bad as labeling someone who doesn't believe in creationism a heretic and a blasphemer. The creationist have chosen their religion, and you have chosen yours apparently. Log in, so your part in this discussion isn't set down to 0.
I agree wholeheartedly. Evolution and "Intelligent Design" are not the only ways to explain the origins of living things. For instance, why does the "creator" in the intelligent design model HAVE to be "God"? What if some ancient race from another part of the universe decided to experiment with creating new forms of life and chose to do it here? Since we currently have no true understanding of who or what we are, how can we say this didn't happen?
That example aside, my point is that evolution can't be the only answer since there is as much a lack of data there as there is for proving intelligent design. There may be more data to support it than the creationists have, but it's still not complete. Until all the data is in, it remains a theory in the traditional sense of the word: an unproven hypothesis.
When you exist in a universe that is so much larger that you or your ego are, you need to accept that fact that there are things that you are never going to understand or see. You aren't large enough to see the whole picture. If you refute this, then you only comfort yourself in your own ignorance and will never develop futher in either intelligence or courage. Being small and less important can be scary.
You are correct sir, I hadn't thought about endianess and I am not sure what AltiVec does other than provide MMX-like features that perform better than MMX. :)
Here is (very simplified) the way the layers to get apps to run on OS X work, starting at the bottom:
;)
Darwin Kernel (BSD Licensed)
|
Cocoa, Carbon (API) application libraries (Closed)
|
Aqua/Apps (Aqua and most apps-Closed)
The Kernel (as always) only provides the system calls from APIs to the underlying hardware.
The APIs are what stand between the apps and the kernel. The killer app here is Aqua.
Aqua stands between the user and the other apps in the system (Finder, the Launchbar, iMovie, etc...)
So... you can run the kernel on x86, but you don't have any ability to recompile the GUI based apps to run even if the application source is available. What you would need would be an open source re-implementation of Cocoa, Carbon, Aqua, etc... Kinda-sorta like W.I.N.E. Probably with about the same level of compatibility. Hope this clears it up a bit...
Fooling OS X into thinking that it's running on a Motorola chip wouldn't be effective either. You run into the same problems you did with old Apple emulation... There are chips in the original hardware that provide things that are usually in libs on other platforms. You would have to emulate Mac motherboard, chipset and proc to get this to work. That WOULD be difficult and unsightly. Not impossible though. Just impractical.
Regarding the potential port of Mac OS X to Intel, I won't say it will never happen. I think all it would take is for Apple's new servers to actually take off. Their greatest selling point is that they don't require a license for each client that connects and they provide decent support for Windows clients. If they get enough market share in the server market, and they can sell the OS without having to rely on the hardware, then there is an open door to doing a port. I think this is why Steve Jobs said that AFTER the migration to OS X, anything is possible. Of course I can't see Steve Jobs himself approving a switch to x86 Intel. 4th gen Itanium? Maybe. At least there, you're dealing with what used to be Alpha to a certain extent. I think Steve Jobs still doesn't like dealing with the companies that he percieves as being "inferior", and I applaud him for that. Only the best things come out of clear direction and tight control: Macintosh, MPlayer, the Linux Kernel, etc...
Just like dishing out catty comments like the above has always been the realm of fag0rtz. Pot. Kettle. Black... You must be a homo too. Stop posting anonymously and come out of the closet, it'll make your life so much better.
OK... totally off-topic here, but I wanted to respond to this with some thinking I've been doing lately:
Interesting point that you make at the end: "This is an education problem not an operating system problem." But, this is wrong. The real answer is that computers are still too hard to use. You are applying this towards the admin end of things, but I am going to shift to the users in a little bit.
As far as the admins go, I think we need to work towards a time when there is no need for an admin. WHAT!!? Heresy you say? No, practicality. Technicians that repair/swap out broken hardware should be around, but OSes and networks should pretty much run themselves.
The original automobiles were hard to use and generally operated by people with a sharper intellect and an ability to work on their cars. Today, any monkey can (almost) operate a car with a minimum of education. And when it breaks down, they take it to a technician to fix the problem. The end-user does not need to know anything about how it operates to get it to move and take them places. That's the direction computers should go in, from both the admin and user perspectives.
Again... the car example: you don't see people getting certified to drive different cars by make, do you? Going from one OS to another should not be as difficult as it is. There are many reasons for the difficulty: monopolies putting up roadblocks, open source programmers who aren't writing for Joe Average, an outdated platform (x86, ISA, PCI...), etc...
As far as users go, I think that we expect too much of them. Even in the Windows world, things that should be simple procedures are difficult if you don't have the "right" software.
Over the weekend, my dad asked me how he could record himself onto a CD with his burner (in Windows). I thought... Easy! I spent the better part of the afternoon trying to show him how to use his limited edition of Roxio and Audacity in conjunction to do this. (He can't afford to buy anything better and he can't use Linux) It turned out not to be so easy after all. The concept of using two different apps to access/create data for each other is an alien and confusing concept to most people since they don't get the abstraction and have problems thinking in a hierarchical fashion. My dad was just hoping to be able to plug in a mic, press record and get his voice on a CD. He didn't want to mess with creating folders and files and editing his recordings. Sure, there are some programs that would keep the user in one application throughout the process, but it's still alien to them. This is why I still say computers are not easy to use.
Just like cars, there will always be a small population of people who understand computing devices pretty thoroughly, and there will always be others who just want to turn the key and drive. So, it's not an education issue, it's a platform, OS, application issue. Ditch the old way of thinking and look at it this way... Someday computing should be happening all around us without any need for us to directly interact. The entire world around us should be one big user interface. Forget "desktop" environment and think about getting computing into the physical realm... Then we won't have what appears to currently be an education problem. And... if we apply this thinking to open source projects now, we'll be right on track with where the world is going.
Just a loose set of thoughts with regard to this article. I don't have time to arrange them into something more cohesive. I use the term Linux to generically apply to complete distros with a desktop environment:
;) )
The quote from Ballmer at the very end of the article may be a harbinger of things to come: software is not consistently profitable. The very fact that Ballmer considers it weird that IBM would tell a company to buy software from someone else indicates that the "playing field" is changing. Sure, IBM isn't at the top of the game anymore, but I think you may start to see more and more companies abandoning the software business for more profitable fields like embedded devices and other dedicated systems that we haven't yet dreamed of. The whole problem with computers right now is that people actually have to "interface" with them in non-intuitive ways. But that's a different topic...
In Neal Stephenson's "In the Beginning there was Command Line", he says that it is the fate of software to become free. Commercial UNIX gave way to free UNIX, Microsoft Word now has a respectable challenger in OpenOffice.org, etc... Or at a deeper level (the concept level as opposed to the product level), GUIs have become so inexpensive compared to the original Xerox systems that some are free: XFree86 + GNOME or KDE. I believe his observations are correct. The OS market will continue to become less profitable if the "movers and shakers" aren't always looking for the next "great thing".
The only thing propping up Microsoft right now is the Office suite and to some extent Internet Explorer. To take this crutch down would only require the provision of a application that uses a totally new and better approach to achieving the same results. No one has done it yet. But again, I digress... (
My point is... that Ballmer's comment about "Added Value" above Linux should really be about finding the next "killer app" that Windows has and Linux does not. This ensures that more people who follow that path of least resistance will choose Windows every time.
These victories are short-lived however. As soon as a concept is out in the open, it's fate is to have reproductions and innovations built around it. Witness: Apple popularizes the GUI that Xerox couldn't move. Microsoft immediately responds with their first release of Windows. Mosaic begat Netscape who begat Internet Explorer... (at the concept level, not the business/profit level).
Look at the music industry. In 1994 the Spice Girls came on the scene and were hugely successful (opinions about their music aside). So what happened immediately after that? Knock offs. Tons of them. None with a chance of making it as big as the original, even if the original was not as good as the newer acts. To a certain extent, this happens in the GNU\Linux\Open Source world more than it should. But, undeniably, there are some ideas that just can't be improved on. So, what do we do? Look ahead and occasionally check the other runners next to you. When I say look ahead, I mean look for new approaches at the user level not the system level. These are real differences that the user can see, feel and experience. Of course, this is assuming that you are interested in moving Linux out to "Joe Average".
Microsoft can't outsmart "Linux" since there isn't any one model to take down without some heavy handed help from the governments of the world. At the moment, they aren't doing to well in that arena either... Linux will be around until something better comes along. That "something better" has to be completely different compared to Linux and provide features that Linux doesn't have. However, it should also still be free. That is where Microsoft will never be able to compete.