I oversimplified things a bit. The current syntax has yet to be decided, for for most metadata attributes the current plan (subject to change without notice blah de blah) is to prefix metadata attributes with double dots, so it'd be
$ cat $HOME/..owner
Alternatively, standard UNIX attributes may be placed in a subsubdir, so:
I don't think the average user understands that Linux is the underlying "kernel" driving KDE or Gnome or whatever, that Linux is the underlying kernel in a RH, SuSe or Yellow Dog distro. Thus, Linux is interoperable no matter what platform you are on.
Tru dat. But Linux is largely interoperable between distros. As much as cars are interoperable with each other, for instance.
I think I am referring to too many prefs in the GUIs, but that is where 99% of the home market wants to do their configuration.
Ah, I think perhaps you misunderstand. It's better to have prefs stored in some kind of database (text files in/etc, gconf registry) as then GUIs can be layered over them. If you use OS X you must know how many small utilities there are that simply adjust settings that Apple (deliberately) do not provide GUIs for - primarily reducing graphics overhead so it can run on lower spec machines. There are lots of TweakUI style programs around, that are basically just front ends to registry keys.
Heck, my girlfriend tells me that Windows had been out a few years before she could actually LEAVE DOS, but now she would never want to go back to all that command line and function key stuff.
That is not evidence that the GUI is superior to the command line. Considering that DOS had no support even for basic command line stuff like tab completion, I'm not surprised she doesn't want to go back. Yet here we are, most Linux users are of course ex-Windows users, but the cli is still a very popular interface. Why? Because the Linux CLI doesn't suck.
She's also probably confusing the fact that most DOS programs were text based, rather than GUI based. So to her, DOS doesn't mean "cd", "dir" etc, it means text mode word processors. It works quite well on Linux, really. And no, it's not the same as in OS X, which doesn't really provide much power via the command line, it inherited one from unix but never really made it a central feature.
Lack of polish is a problem. If you followed my post with my locked-up GUI, the solution was simply plugging my mouse back into my laptop. That's the kind of stuff that infuriates the average user and keeps them away from Linux.
I didn't see that post, but I've seen enough to know that you consider Yellow Dog to be an easy to use Linux distro. Sorry, but Mac support for Linux has never been a high priority for the dev teams, and using on a Mac is a second rate experience compared to an Intel PC. The reasons for this should be pretty clear, PCs are open, Macs are not, and PCs are vastly more popular, so they get priority. I dunno what caused that mouse problem, but I've never heard of anything like that on the PC. Even the PPC versions of mainstream distros (of which YellowDog is not one) are usually harder to setup and are buggier than the PC ones. It should be noted that installing SuSE 8 is easier than installing Windows (98). It detects all my hardware and installs the drivers for it perfectly, and it installs and configures all my software. It'll even resize partitions and auto-detect networks. Setting up Windows with all the drivers and apps takes a day. Linux takes a few hours. Installation is no longer a problem mate. I suggest you find a modern PC and try the latest Intel distros (suse ppc is not the same btw).
While I appreciate that the way a distro essentially makes money is by making you pay for support, it seems that a little ridiculous to have to call tech support for every little hiccup. Maybe I'll feel differently when I have a better handle on Linux, but right now, I truly am a Linux newbie.
Don't worry, we all were once. This is another misconception, that companies like RedHat/SuSE make money out of private individuals phoning up to ask how to install their CD burner. When we say "support" in reference to distros, we mean corporate level support. This isn't for fixing the odd problem, it's for having 24/7 engineers on callout, consulting etc. You might want to check out the #linuxhelp channel on FreeNode IRC. It's where I got all my tech support, didn't have to ring SuSE once.
Linux definiately wins on being able to tear the entire system apart to find a problem and make however you want. But then again, as I mentioned, that is why I think, for now, the best of both worlds is OS X.
If you believe that you can tear apart and change OS X in the same way as Linux then you need to think again. Even Windows is vastly more configurable than OS X, you can alter almost any setting, it's themable, it can also have multiple user interfaces (check out talisman). OS X is a dog when it comes to customization, you need to buy 3rd party programs just to turn off graphics effects you don't want by and large.
Finally, it should be noted that the biggest strength Linux has is of course it's licensing. It's a gift to society, not a mechanism for extracting money from locked-in customers (that applies to apple too). Enough said. Seeing why Linux is the way forward despite appearances takes time, but many do get there, hence the number of people working on improving it.
Is this correct? Will the VFS also be extended so that you can make use of extended attributes in XFS?
Cooler, if I read the tea leaves right. I believe some time ago now there was a thread on lkml about whether it'd be possible to have files as also directories (and vice-versa). The reasoning behind this was simple: we want flexible filing system attributes, but not at the expense of API bloat. You want ACLs? That'll be another API then. Extended Attributes? Another API. What, you want heirarchical extended attributes too? Well you've just created another version of the filing system API haven't you.
The theory goes (and Hans Reiser, top guy, explains it much better than I can) that by altering one of the rules of the filing system, we can get lots more power and expressiveness without having to invent lots of new APIs. Let's say you want to find out the owner of file foo. You can just read/home/user/foo/owner. You can edit ACLs by doing similar operations. Now you can have something more powerful than extended attributes, but you can also manipulate that data using the standard command line tools too! Coupled with a more powerful version of locate, you can have very interesting searching and indexing facilities.
This has implications beyond just string attributes. Now throw in plugins, so for instance the FS layer interprets JPEGs and adds extra attributes. Now you can read the colour depth of an image by doing "cat photo.jpg/colour_depth" or whatever. You can get the raw, uncompressed version of the file by doing "cp photo.jpg/raw > photo.raw". Noticed something yet? You no longer need a new API for reading JPEG data, because you are reusing the filing system API.
But the FS is not a powerful enough concept, I hear you cry! Have no fear, for with new storage mechanisms comes new syntax too, to allow for BeFS style live queries. If you want more info, you should really read up on this stuff at Reisers site.
That's why ReiserFS is so good at small files as well as large files. Have you ever wondered why that is? It's not just a quirk of its design, it was very deliberate. One day, Hans wants to see us store as much information as possible in a souped up version of the filing system, so reducing interfaces and increasing interconnectedness. Or something. It sounds cool anyway:) That's one thing that RFS has that the other *FSs don't - the ReiserFS team has vision.
The Linux "collective" needs to start making Linux less configurable (yes I said LESS) for the average user. The average user just doesn't care about all the configurability of Linux... heck, I'm a slightly above average user and I don't care!!! When the learning curve is less steep for Linux, or when somebody offers a distro (Lycoris?) that has less of a learning curve (a sort of Linux Home edition), then Linux will make some major inroads into the home.
A few points.
a) There is no Linux "collective". There are distros, there are projects. Some are making things more configurable, some less.
b) There is indeed a recognition that too many prefs make bad GUIs (which is what you're actually referring to I think). For instance, GNOME2 has far simpler configuration and in some cases fewer features than GNOME 1.4 - this was painful for some but deliberate. Often, you can access prefs via the GConf system anyway like in the Windows registry if you must change settings that don't have GUIs.
c) There are lots of reasons Linux has yet to make big inroads onto the desktop. Too much configurability is only a very minor one when compared to the just general lack of polish it has. For instance, it's too hard to install software often.
Finally, although it's a common truism that Joe User does not want complexity or configurability, my own experience with non-geeks are that they often love the higher flexibility of Linux. "You can have multiple user interfaces, cool!", "You can change the window borders - sweet!", "You can make an auto updated web cam link on your desktop - amazing!". Yes, of course there are some people who cannot stand choice, probably because they are not used to it, but many many others like it. What can you do?
Well, having said that, looks are indeed an issue, but clearly something is being done about that (thanks RedHat!). The real problem however is that both desktops provide their own APIs to do essentially the same thing. That's OK.... when those APIs are compatible, though it does use up more memory than strictly necessary.
But all too often these APIs aren't compatible. Theming APIs are one good example as they are related to the problem that people either bitch about or get used to (depending on how long they use Linux for;) Both GTK and Qt have different ways of writing pluggable theming engines. What RedHat have done is create several themes that look the same. OK. Good start. What we really need to do though is have a standard for plugin theming engines, so I can use the GTK ThinIce theme in KDE and vice-versa. Then if I run a program in KDE, it adopts the theme I chose for KDE, and if I run it in GNOME, it adopts the theme I chose for gnome.
That's only one simple example. Sound servers are another pet hate. When I run Konqueror inside of GNOME2, it won't play sounds. Why? Because Konqueror uses aRts, and GNOME uses eSound. In about 6 months gnome is switching to GStreamer, and KDE is looking interested too, which will hopefully provide a unified API for Linux multimedia. The work Havoc Pennington is doing at freedesktop.org is great too.
We're getting there, but it's a lot of effort. Which brings me on to my final point. There's a reason there are so few unified APIs - it's very, very hard to create code that is easily used by anybody. You often have to manually create bindings to umpteen different languages, and even different "feels" to make all the different programmers happy. Then you have to campaign for adoption. It's a lot of work.
What's needed is a decent object model. Something as easy as KParts, as powerful as CORBA and as ubiqitous as COM. Something that lets me write an object in any language (including C), and then "export" it to the OS, so that it can be used in any other app. This would not only increase code sharing (so reducing memory usage), but would also lay the groundwork for many more technologies to be standardised. I wish I had time to do this, but it's far from easy.
What's this talk of a golden age? An age where we are all so paranoid that we encrypt our mail routinely? Sounds like a world ruled by fear more than anything. I for one have nothing to hide, and want no part in it.
What's that? My iMac runs Linux just fine, and I can install any of 10,000 packages from the Debian archive. Well over 99% of commonly-used Linux software builds and runs equally well on any hardware platform. './configure && make'
Ah yes, you're missing the point. Most Linux software is open source, so can be rebuilt for any CPU architecture. Most Windows/commercial software is not, so you need CPU emulation.
since Linux is looking more and more like Windows every day. I'm amazed at how much KDE tries to ape Windows rather than trying to adddress the problems of the Windows interface.
I fear that if Linux continues in this problem we may end the problem of being weighed down by a monopolistic regime but we will still not have bettered the PC computing environment.
Well, GNOME2 is trying to come up with new interface ideas. Take a look at some of jimmacs screenshots
Unfortunately there's a very good reason that these projects are similar to Windows - namely that they have to be. Most people are NOT willing to learn anything in order to use a new system. That's why software meant for joe public rarely comes with user manuals. Online training can go some way towards fixing this, but you've still got the problem that you've got to be similar to Windows in order to get people to switch.
Note that doesn't mean you always have to be like Windows. KDE has a truckload of cool stuff that Windows doesn't have, and Linux also has usability features Windows doesn't (no, really) like single-click. There was a huge thread on the KDE lists about this: namely some people double click on icons when they don't need to. Which is better, single or double? Usability says single is more consistant, makes more sense and is easier to do, but people are used to double.
KDE is a great launchpad for Windows converts. They can always migrate to GNOME, Enlightenment, FluxBox or whatever later.
ok, are there ANY numbers that back this up? I keep hearing AGAIN and AGAIN about how people are migrating to OS X in droves or something, but I never hear numbers to back it up - it's just gushing from Mac fanatics.
You are correct, and I am tired of it too. Here are some facts. These statistics are not made up.
The total Macintosh market share is rougly 4% (these figures come from Apple itself by the way, so will probably be slightly biased in favour of the Mac). 90% of Mac users still run OS9, only 1 in 10 have upgraded/bought new Macs since then. As such, the market share of OS X is about 0.4%
Half a percent market share is pitiful even compared to the estimated 2% market share of desktop Linux. Where do I get this figure for Linux from? Google, which I think must be a pretty good tally. Their zeitgeist page doesn't round by the way, it truncs, I asked them once and was told the true figure was closer to 1.85/1.9% so I call it 2%. IDC estimate it as being higher than that, take your pick.
Therefore we can see that takeup of MacOS has been pathetic. It has only a quarter of the market share of Linux even on the desktop, and this article was about servers anyway. We see large numbers of Apple converts here on slashdot because they are geeks going over to it because they are willing to spend lots on technology (note the theme of that article is, IT depts are poor) and because it's supposedly UNIX but looks good. This doesn't reflect the real world, hence the huge upgrade fee for Jaguar.
Here is another fact for you. MacOS X apps are not UNIX apps. They use virtually no UNIX apis as standardised by POSIX, they are virtually all Apple proprietary APIs. Given this, I can't see how the MacOS is UNIX anymore than Windows XP with Cygwin is UNIX.
While there are probably a lot of corps out there thinking about switching to linux from unix/windows, there are also an increasing amount of home users searching for an alternate desktop environment.
I wonder how this might tally if things such as linux firewalls, mp3 servers, and other more custom uses were considered?
I don't know, but my experiences in this are as follows:
I tried Linux first in 1995, but it wasn't ready for me, and I wasn't ready for it. It got deleted.
Ditto in 1998. That was when KDE was in Alpha.
I was finally wooed (by screenshots and happy tales from people I met online) into buying SuSE in January of this year. I switched over.
My friends watched this process with interest. They came round, toyed with Linux etc. My Machead friend experimented with it for running his old iMac as OS X was too much of a dog performance wise for it. He tried lots of distros, but didn't really do any research (he tried Debian first;) and he wasn't prepared to do any learning, and Linux still has a sharp learning curve for home desktop usage. Another friend of mine decided to turn an old box of his into a router/firewall for to share his home network - and also to use it to play with Linux on the desktop. As far as I know, he still uses XP on the desktop, but Linux is happily running a small server.
Another friend wanted to try it, but was prevented by the fact that Linux can't resize XP NTFS partitions yet. Finally, Hugh had a brother who was into it, and so he's tried it as well.
So far, I'm the only one who stuck with it, probably because I'm the most technical and everybody has a "switch" threshold, the point at which they are confident enough and Linux is easy/compatible enough for them to make the leap.
For most people it isn't there yet, hence the tiny (2-3%) market share it has on the desktop. BUT... the server end is often a way for people to get into it, as Ken is doing.
The corporate desktop would come first I think. Really we should be concentrating on that first, as the entry barrier for the corporate desktop is lower and the demand is higher (MS licensing etc). Home desktops will follow naturally after that.
I haven't used gnome or kde yet (do they run on OSX? ) but I think there are a lot of things wrong with that screenshot's interface..
I believe you can get them running on OSX, but why bother? They are seriously crippled on that platform due to lack of full POSIX compliance, and people buy OS X largely for the GUI. Anyway. Let's deal with your criticisms
First of all, the desktop icons... they look... weird. It's some weird angle on the icons. Also, the shoadow on the bottom of the icons don't make sense. The shadow on the front of the folders should be eliminated.
These folders are, fyi, SVG icons so they can be scaled to any size. Because this shot was of an early beta of GNOME2, not all the icons had been converted, hence the 2 different styles (both of which I think look quite nice really). I don't know what you mean about the shadows, the icons look fine to me. This really seems to be merely a personal aesthetic issue, which is fine, that's why we have icon themes.
Also, in the 'gfx' icon, I don't see what the slide should mean. The toolbar icons should be made a bit larger, or the toolbar itself should be made a bit smaller. Users are too apt to miss the button.
The slide - who knows? It was a demo of the different folder decorations you can have. There's no need to use it, if you don't think it's appropriate. The toolbars are fine, I use them all the time and you don't have to click directly on the image, as long as you're close it works OK.
Finally, the text rendering on the taskbar(which is not unique, but win95ish) looks weird. Look at 'Keyboard Accessibility Control'. The text should probably be squeezed or truncrated, but instead it clashes with the icon.
That was a minor bug that has since been fixed. I'm using it now, and the buttons are fine.
Oh yeah, the 'applications' menu, sounds macos-classic-ish, but I can't tell since I haven't actually seen it.
I've used Classic a bit, I don't recall any Applications menu. At least on the mac I used, everything was started either from the finder, or DragThing.
The point was not to hold up that particular screenshot as UI perfection itself, but to demonstrate that gnome has a fairly unique look to it.
Also, you don't need to emulate Windows using KDE/GNOME. Their default configurations just include a panel at the bottom of the screen with a K or a foot where the Windows Start Button is.
Actually this isn't true of GNOME any more. The default layout is as far as I can tell unique. You can see a screenshot of it here.
There is a bar across the top Mac style (though that's the only similarity to the mac), and a taskswitcher with desktop switcher at the bottom. There are two menus, Applications and Actions (which makes sooo much sense I can't help but think, why did nobody think of this before?). You can add your own applets to the panels, or add your own. It's not only extremely flexible, but very intuitive as well.
Having said that, I wouldn't recommend GNOME2 for new users yet. I'm using it now, and it's interesting as much for what it lacks as for what it's got. It's clearly a fantastic foundation to build on, and I await the 2.2 release with interest. Hopefully Nautilus won't suck then.
But anyway. The old bottom panel with the foot system has gone (unless you use RedHat) - I suggest you check out those screenshots of Jimmacs. They are too sweet. And nothing like Windows or the Mac, or KDE.
This is good news, but without a big corp behind it I'm sceptical about whether hardware manufacturers will adopt it. What I'm more worried about now is Xiph themselves though. They've done a great job, and given us this fantastic gift, but now how do they make money? If the library was originally proprietary, then what do they have now?
I'm mainly referring to the guys at Webb who started taking over the business, back when the subsidiary was getting bigger than the parent company itself.
Also, it should be noted that I've often talked with employees of the company. Of course they may just be pointlessly doom talking or whatever, but they told me it was nothing like the company it once was.
Why can't you people* realize that a device that has an ugly user interface can't possibly be "better" in any meaningful sense of the word than any other device? I don't care if it runs off of the moral power of virginity and ends world hunger. If it's ugly to look at, it sucks.
Indeed, as the engine of your car is so ugly, it can't possibly be any better than, say, a lava lamp. Clearly, as an engine is ugly, it sucks, so it should be replaced with a lava lamp, which isn't ugly, so it doesn't suck. Please tell me how you get on when you've replaced your car engine with a lava lamp.
We have talked to the jabber.org and jabber.com multiple times. It's always been difficult to figure out how this all fits together from an open source point of view. You see, jabber.com has patents on stuff you need to implement jabber. At the Jabber BOF at IETF I specifically asked them if they would make this IPR available in a way that worked for open source people. They answered that people had implements this stuff and they weren't suing them. This is like yah, DUH, of course when we are trying to get people addicted to the drug we don't sue them. They have NEVER made any commitment to allow this IPR to be used in open source products. They are a desperate company looking for a way to make a business model out of jabber. If you think jabber is the best for open source - give that some careful thought.
As far as I know, this is just FUD. I've been tracking Jabber since the cows came home, as has Adam, and neither of us has heard anything about any trouble from (enforceable) patents on Jabber.
Having said that, it's probably true. Bear in mind any non trivial program will have patents on it within the US. There are hundreds of patents that affect Linux for instance. What can these companies say, except we won't use them?
Finally, Jabber Inc does have a business model. They sell industrial strength commercial servers and support. This does make money. The question is, will they make enough money to survive.
Unfortunately, J inc has sort of lost its way lately. Andre Durand who set it up has left, and it was taken over by his former bosses who didn't really have much of a business before. The company has been going downhill since then.... I hope they do survive, but it's not the company it once was. Oh, and a quick disclaimer : I now work for Andre, on an unrelated project;)
If a good public server was available, I would have been running jabber years ago.
Well, it's a plug, but here goes:)
I help run the theoretic.com with Adam Theo, which is most definately not a public testing server.
First up, we don't have AOL or ICQ transports, as we got too popular to remain unnoticed ages ago so it's firewalled now. The MSN transport still works fine though, and Yahoo, well, nobody used that:) You can always use an external transport (and many do)
Now for the good news. There are 2 admins in 2 separate timezones (London and Florida time), so there's normally somebody about if you want to chat to an admin. We don't track the latest releases unless we need a bugfix, so the server is stable. We haven't rebooted it since February as far as I can remember.
We also have some features other servers don't. For instance, the SMTP transport is mondo cool. If you send an email to user@im.theoretic.com, it'll be turned into a jabber message. If you reply, it'll be turned back into an email. You can add email addresses to your roster as well if you want to fire off quick messages.
Finally, we try to be as polite as possible. If we're restarting a transport for whatever reason, or need to reboot the server (rare now), we'll send a public broadcast message about 10 minutes before so you can finish up conversations etc.
RhymBox, which is a great little Windows client already, and the new version will totally rock, tied with theoretic should hopefully ease most of the pains that have put off non geeks from using jabber up until now.
... but on the desktop, GUI is the only thing that really matters :
Actually you are wrong. On the home user desktop maybe, but we're talking corporates here. To IT administrators, things like cost, stability, security, CPU efficiency, open source and unix underpinnings do in fact matter a great deal.
Especially the open source thing of course, as it means "no lockin" - virtually every IT using company today has been bitten on the backside by this, so that's a major win.
As an aside where do you see companies with 2ghz chips that just run Office? Most company machines I've seen are fairly low end by todays standard.
You are the reason that people are working on Wine. The chances of Macromedia ports to Linux are slim: not really through lack of will, Macromedia are friendly to Linux/open source, more because the massive investment that would be required could not be justified.
One day of course there will be Dreamweaver/Linux, but until that day, you can run Dreamweaver under Wine. It works just like in Windows, and gets some additional capabilities as well (like being able to read Linux disks). I believe Dreamweaver has a 4/5 star rating at the moment, so it's not perfectly there yet, but CodeWeavers are working hard on popular apps like these.
And yeah there are nice areas. For the privileged few. The top 5 percent. Yeah I two ISDN lines in my apartment. And the montly cost would have fed a family of five.
Show me where opensource benefits latin america. Medical records, state agencies, but until then this announcement has all the weight of Pam Anderson announcing her new fashion line.
I'm sorry, it sounds harsh, but as others have pointed out this is probably more important for the west than Latin America. If other countries see that Venezuela made the switch OK, then even if Venezueala doesn't save money (as they weren't paying for it before anyway) they will see that there is an opportunity for them to save money.
Can somebody point out the areas where Linux isn't yet as good as Windows? Granted, you can't run as many Windows Apps on Linux as you can on Windows, but then different versions of Windows have the same problem. IMHO, 2002 will go down in history as the year Linux ease of use surpassed that of Windows. What else do we still need to fix?
Here's a quick list off the top of my head:
Usability: GNOME2 kicks ass in this dept, but KDE is still (rightly) in the lead as it has more features and is generally more robust. It's usability effort isn't ramped up yet though. It'll be interesting to see what effect the competition from gnome has on them
Multimedia. We have various incompatible sound servers. OSS sucks, but ALSA0.9 (development version) which is what I'm stuck with sucks more. Even then, we don't have a MM framework, though hopefully kde and gnome will adopt GStreamer
Package management. It's still too hard to install/uninstall packages. I'm working on this in my small amount of spare time with autopackage - 0.1 will be released some time this month hopefully.
Configuration - not an issue for corporate desktops of course, but a big deal for the home user. Where GUIs do exist, they are split between KDE/GNOME/Distro control centeres. Yuck.
Hardware support: not really a huge issue anymore, if anything Linux is now better than Windows in terms of easy autodetection and driver installation. WinModems need better support, I'm using one right now but support is still primitive.
Consistant theming. This is a tricky one, as once you get used to it which happens almost within 5 minutes of starting to use Linux, you no longer care. It's a big deal for new users though.
Better windows compatability. Wine has been making huge progress lately, but nothing less than virtual perfection in this area will do. In particular, the inability to resize NTFS partitions is an acute problem - it prevents WinXP users from trying Linux out.
Games. Need I say more?
Shared object model. Not being able to embed a Dia diagram into OpenOffice is dumb. We need to make something as ubiquitous as COM/ActiveX, as simple as KParts but as powerful as CORBA/Bonobo. Anybody up for it?
Online training. We haven't even started this one yet. Too many users freeze at the sight of something that doesn't have a start button. The best way to get people over the learning curve (considering many can't be bothered learning themselves) is to have online training. Interactive tutorial engines etc.
So I think you can see.... we have a lot of work cut out for us:)
Linux users have been saying this for years, yet the mythical next version of Linux, KDE, or Gnome never seems to completely deliver on this promise. Plus you are assuming that Windows' "goodness" is frozen. You are mistaken if you think Microsoft will not spend a portion of its $40 billion to improve Windows.
Sure. A lot of that was hype though. I mean the remaining problems are all toughies - stuff like good desktop integration, package management etc. When these problems are solved, Linux will be as good as (if not better) than Windows 98 in terms of user interface and consistancy. What is Windows 2000? To the average user, it's 98 that doesn't crash. What is XP? To the average user, it's 2000 with knobs on. 4 years on, MS has improved the reliability of Windows no end, as well as added lots of features (especially on the corporate end). For the large majority of desktop users though, the difference is minor - hence the slow upgrade rate. Most people still use Windows 98, and they'll upgrade to make it suck less, not to get the killer features (hehe) available in XP.
So I think it's absolutely possible to catch up to Windows. I know Windows won't be static. If Longhorn delivers, then once more Windows will be ahead of us in some areas. But we'll have caught up with XP by that point, so it's a straight race.
Finally, remember that the scale of Windows development scales linearly (ie they can only hire X new developers a year, if that), whereas Linux development scales exponentially. Especially as it's so easy to work on, the number of man hours being invested in Linux already easily outstrips that being put into Windows 2000 at its peak (about 5000 engineers). I think we can catch up.
Apparently they have practical rather than philosophical motivations: keeping cash in the country and promoting local software development.
When you're rich, the time and sacrifices needed for philosophy are cheap. When you're poor, the practical rules the roost.
Pithy comments aside, this only reinforces what I have come to believe in recent months: that the eventual dominance of Linux/open source is an economic inevitability.
The reasoning behind this is very simple, and has nothing to do with blind zealotry. Capitalism does not tolerate inefficiency. If you can do something better than your competitors, or if you can do the same thing but cheaper, you will have an advantage and the natural selection of the free market will elevate you above the rest.
Linux is more efficient in economic terms. Right now of course, it's "efficiency" is being held back by the number of rough edges that need polishing, the huge resources needed to overcome Microsoft lockin and so on. However, these are becoming less and less all the time. Eventually (like within a few years) Linux will be as good as Windows, as well as compatible with it thanks to the efforts of the wine/samba/OpenOffice/NTFS crews. At that point, you can be better and cheaper at the same time by using it. The result? Market dominance.
It has another advantage as well - multiple vendors. History shows that economics favours systems with multiple vendors: witness Macs vs PCs, or VHS vs Betamax.
Re:Socialism is all that works for information
on
Want Freedom?
·
· Score: 2
Here I agree completely. I am all for minimal government involvement in business, but law enforcement should be ever present, which, in this case, means prosecution of fraud.
Unfortunately too many countries take the minimal government approach too far. Capitalism optimizes for efficiency and little else. Social justice, the environment, humanity - it is all secondary to efficiency: ie producing the greatest amount of profit.
Minimal government involvement in business is a bad idea, because capitalism only serves society when it is bracketed and constrained by that society and directed towards a greater good. It's similar to electricity - it has raw power that should be respected and harnessed, but that's not an excuse for minimal insulation or lack of safety features, though they do make it harder to use.
Business, by itself, looks out only for itself. No, this is not good, this is useless. If we accept the proposition that the economy is not the be-all and end-all of society, but in fact is meant to serve us by sustaining and sometimes increasing our standard of living, then we can see that the trick is not to get minimal government interference, it's to get the right amount of government interference.
Only then can capitalism serve us, rather than the other way around.
A laugh or two at some American excess is one thing, but to portray America as malign is just the opposite of reality.
Firstly, before I start, don't get me wrong. I'm not saying America is evil or anything. But I'd like to make a few points, America is hardly saintly either:
Guilty of hypocrisy over Iraq? Little known fact is that despite all the bitching from Washington about the lack of inspectors in Iraq etc, they don't allow UN inspectors into some American sites either - and we KNOW that America has weapons of mass destruction, so shouldn't they be monitored too?
Since Bush came to power, America has ripped up more international treaties and conventions, and ignored more UN conventions than the rest of the world combined in the last 20 years. Examples: kyoto is the biggie of course, but america also attempted to dismember many other international bodies
America is steadily isolating every single ally it has. Even the British government is seriously split on the issue. It apparently refuses to listen to anybody except big business, and has firsthand told the world that it will not do anything for the environment if that might mean harm comes to the American economy. They acknowledge the science and even agree with most of it (the rest of the world long ago agreed with all of it, but hey) - yet they still refuse to take action.
Bush preaches free trade as the means of saving the world. Yet he has imposed massive trade barriers to protect inefficient American industries.
The administration is prepared to invade another country pretty much without evidence, and without any political backing from anybody else at all. If the US can invade Iraq because Bush doesn't like Hussein, why is Iraq invading Kuwait bad?
Evil? No. But where do you draw the line between deliberate damage through rampant self interest and evil anyway?
Sorry, but had to get that off my chest. Politics is important, games or no. I'm not saying other countries are perfect either, far from it.
Finally, political views are frequently expressed through comedy as well. Is that infantile?
Slashdot Mirror
I oversimplified things a bit. The current syntax has yet to be decided, for for most metadata attributes the current plan (subject to change without notice blah de blah) is to prefix metadata attributes with double dots, so it'd be
:
$ cat $HOME/..owner
Alternatively, standard UNIX attributes may be placed in a subsubdir, so
$ cat $HOME/..metadata/owner
Nobodies entirely sure yet.
Tru dat. But Linux is largely interoperable between distros. As much as cars are interoperable with each other, for instance.
I think I am referring to too many prefs in the GUIs, but that is where 99% of the home market wants to do their configuration.
Ah, I think perhaps you misunderstand. It's better to have prefs stored in some kind of database (text files in /etc, gconf registry) as then GUIs can be layered over them. If you use OS X you must know how many small utilities there are that simply adjust settings that Apple (deliberately) do not provide GUIs for - primarily reducing graphics overhead so it can run on lower spec machines. There are lots of TweakUI style programs around, that are basically just front ends to registry keys.
Heck, my girlfriend tells me that Windows had been out a few years before she could actually LEAVE DOS, but now she would never want to go back to all that command line and function key stuff.
That is not evidence that the GUI is superior to the command line. Considering that DOS had no support even for basic command line stuff like tab completion, I'm not surprised she doesn't want to go back. Yet here we are, most Linux users are of course ex-Windows users, but the cli is still a very popular interface. Why? Because the Linux CLI doesn't suck.
She's also probably confusing the fact that most DOS programs were text based, rather than GUI based. So to her, DOS doesn't mean "cd", "dir" etc, it means text mode word processors. It works quite well on Linux, really. And no, it's not the same as in OS X, which doesn't really provide much power via the command line, it inherited one from unix but never really made it a central feature.
Lack of polish is a problem. If you followed my post with my locked-up GUI, the solution was simply plugging my mouse back into my laptop. That's the kind of stuff that infuriates the average user and keeps them away from Linux.
I didn't see that post, but I've seen enough to know that you consider Yellow Dog to be an easy to use Linux distro. Sorry, but Mac support for Linux has never been a high priority for the dev teams, and using on a Mac is a second rate experience compared to an Intel PC. The reasons for this should be pretty clear, PCs are open, Macs are not, and PCs are vastly more popular, so they get priority. I dunno what caused that mouse problem, but I've never heard of anything like that on the PC. Even the PPC versions of mainstream distros (of which YellowDog is not one) are usually harder to setup and are buggier than the PC ones. It should be noted that installing SuSE 8 is easier than installing Windows (98). It detects all my hardware and installs the drivers for it perfectly, and it installs and configures all my software. It'll even resize partitions and auto-detect networks. Setting up Windows with all the drivers and apps takes a day. Linux takes a few hours. Installation is no longer a problem mate. I suggest you find a modern PC and try the latest Intel distros (suse ppc is not the same btw).
While I appreciate that the way a distro essentially makes money is by making you pay for support, it seems that a little ridiculous to have to call tech support for every little hiccup. Maybe I'll feel differently when I have a better handle on Linux, but right now, I truly am a Linux newbie.
Don't worry, we all were once. This is another misconception, that companies like RedHat/SuSE make money out of private individuals phoning up to ask how to install their CD burner. When we say "support" in reference to distros, we mean corporate level support. This isn't for fixing the odd problem, it's for having 24/7 engineers on callout, consulting etc. You might want to check out the #linuxhelp channel on FreeNode IRC. It's where I got all my tech support, didn't have to ring SuSE once.
Linux definiately wins on being able to tear the entire system apart to find a problem and make however you want. But then again, as I mentioned, that is why I think, for now, the best of both worlds is OS X.
If you believe that you can tear apart and change OS X in the same way as Linux then you need to think again. Even Windows is vastly more configurable than OS X, you can alter almost any setting, it's themable, it can also have multiple user interfaces (check out talisman). OS X is a dog when it comes to customization, you need to buy 3rd party programs just to turn off graphics effects you don't want by and large.
Finally, it should be noted that the biggest strength Linux has is of course it's licensing. It's a gift to society, not a mechanism for extracting money from locked-in customers (that applies to apple too). Enough said. Seeing why Linux is the way forward despite appearances takes time, but many do get there, hence the number of people working on improving it.
Cooler, if I read the tea leaves right. I believe some time ago now there was a thread on lkml about whether it'd be possible to have files as also directories (and vice-versa). The reasoning behind this was simple: we want flexible filing system attributes, but not at the expense of API bloat. You want ACLs? That'll be another API then. Extended Attributes? Another API. What, you want heirarchical extended attributes too? Well you've just created another version of the filing system API haven't you.
The theory goes (and Hans Reiser, top guy, explains it much better than I can) that by altering one of the rules of the filing system, we can get lots more power and expressiveness without having to invent lots of new APIs. Let's say you want to find out the owner of file foo. You can just read /home/user/foo/owner. You can edit ACLs by doing similar operations. Now you can have something more powerful than extended attributes, but you can also manipulate that data using the standard command line tools too! Coupled with a more powerful version of locate, you can have very interesting searching and indexing facilities.
This has implications beyond just string attributes. Now throw in plugins, so for instance the FS layer interprets JPEGs and adds extra attributes. Now you can read the colour depth of an image by doing "cat photo.jpg/colour_depth" or whatever. You can get the raw, uncompressed version of the file by doing "cp photo.jpg/raw > photo.raw". Noticed something yet? You no longer need a new API for reading JPEG data, because you are reusing the filing system API.
But the FS is not a powerful enough concept, I hear you cry! Have no fear, for with new storage mechanisms comes new syntax too, to allow for BeFS style live queries. If you want more info, you should really read up on this stuff at Reisers site.
That's why ReiserFS is so good at small files as well as large files. Have you ever wondered why that is? It's not just a quirk of its design, it was very deliberate. One day, Hans wants to see us store as much information as possible in a souped up version of the filing system, so reducing interfaces and increasing interconnectedness. Or something. It sounds cool anyway :) That's one thing that RFS has that the other *FSs don't - the ReiserFS team has vision.
A few points.
a) There is no Linux "collective". There are distros, there are projects. Some are making things more configurable, some less.
b) There is indeed a recognition that too many prefs make bad GUIs (which is what you're actually referring to I think). For instance, GNOME2 has far simpler configuration and in some cases fewer features than GNOME 1.4 - this was painful for some but deliberate. Often, you can access prefs via the GConf system anyway like in the Windows registry if you must change settings that don't have GUIs.
c) There are lots of reasons Linux has yet to make big inroads onto the desktop. Too much configurability is only a very minor one when compared to the just general lack of polish it has. For instance, it's too hard to install software often.
Finally, although it's a common truism that Joe User does not want complexity or configurability, my own experience with non-geeks are that they often love the higher flexibility of Linux. "You can have multiple user interfaces, cool!", "You can change the window borders - sweet!", "You can make an auto updated web cam link on your desktop - amazing!". Yes, of course there are some people who cannot stand choice, probably because they are not used to it, but many many others like it. What can you do?
But all too often these APIs aren't compatible. Theming APIs are one good example as they are related to the problem that people either bitch about or get used to (depending on how long they use Linux for ;) Both GTK and Qt have different ways of writing pluggable theming engines. What RedHat have done is create several themes that look the same. OK. Good start. What we really need to do though is have a standard for plugin theming engines, so I can use the GTK ThinIce theme in KDE and vice-versa. Then if I run a program in KDE, it adopts the theme I chose for KDE, and if I run it in GNOME, it adopts the theme I chose for gnome.
That's only one simple example. Sound servers are another pet hate. When I run Konqueror inside of GNOME2, it won't play sounds. Why? Because Konqueror uses aRts, and GNOME uses eSound. In about 6 months gnome is switching to GStreamer, and KDE is looking interested too, which will hopefully provide a unified API for Linux multimedia. The work Havoc Pennington is doing at freedesktop.org is great too.
We're getting there, but it's a lot of effort. Which brings me on to my final point. There's a reason there are so few unified APIs - it's very, very hard to create code that is easily used by anybody. You often have to manually create bindings to umpteen different languages, and even different "feels" to make all the different programmers happy. Then you have to campaign for adoption. It's a lot of work.
What's needed is a decent object model. Something as easy as KParts, as powerful as CORBA and as ubiqitous as COM. Something that lets me write an object in any language (including C), and then "export" it to the OS, so that it can be used in any other app. This would not only increase code sharing (so reducing memory usage), but would also lay the groundwork for many more technologies to be standardised. I wish I had time to do this, but it's far from easy.
What's this talk of a golden age? An age where we are all so paranoid that we encrypt our mail routinely? Sounds like a world ruled by fear more than anything. I for one have nothing to hide, and want no part in it.
Ah yes, you're missing the point. Most Linux software is open source, so can be rebuilt for any CPU architecture. Most Windows/commercial software is not, so you need CPU emulation.
I fear that if Linux continues in this problem we may end the problem of being weighed down by a monopolistic regime but we will still not have bettered the PC computing environment.
Well, GNOME2 is trying to come up with new interface ideas. Take a look at some of jimmacs screenshots
Unfortunately there's a very good reason that these projects are similar to Windows - namely that they have to be. Most people are NOT willing to learn anything in order to use a new system. That's why software meant for joe public rarely comes with user manuals. Online training can go some way towards fixing this, but you've still got the problem that you've got to be similar to Windows in order to get people to switch.
Note that doesn't mean you always have to be like Windows. KDE has a truckload of cool stuff that Windows doesn't have, and Linux also has usability features Windows doesn't (no, really) like single-click. There was a huge thread on the KDE lists about this: namely some people double click on icons when they don't need to. Which is better, single or double? Usability says single is more consistant, makes more sense and is easier to do, but people are used to double.
KDE is a great launchpad for Windows converts. They can always migrate to GNOME, Enlightenment, FluxBox or whatever later.
You are correct, and I am tired of it too. Here are some facts. These statistics are not made up.
The total Macintosh market share is rougly 4% (these figures come from Apple itself by the way, so will probably be slightly biased in favour of the Mac). 90% of Mac users still run OS9, only 1 in 10 have upgraded/bought new Macs since then. As such, the market share of OS X is about 0.4%
Half a percent market share is pitiful even compared to the estimated 2% market share of desktop Linux. Where do I get this figure for Linux from? Google, which I think must be a pretty good tally. Their zeitgeist page doesn't round by the way, it truncs, I asked them once and was told the true figure was closer to 1.85/1.9% so I call it 2%. IDC estimate it as being higher than that, take your pick.
Therefore we can see that takeup of MacOS has been pathetic. It has only a quarter of the market share of Linux even on the desktop, and this article was about servers anyway. We see large numbers of Apple converts here on slashdot because they are geeks going over to it because they are willing to spend lots on technology (note the theme of that article is, IT depts are poor) and because it's supposedly UNIX but looks good. This doesn't reflect the real world, hence the huge upgrade fee for Jaguar.
Here is another fact for you. MacOS X apps are not UNIX apps. They use virtually no UNIX apis as standardised by POSIX, they are virtually all Apple proprietary APIs. Given this, I can't see how the MacOS is UNIX anymore than Windows XP with Cygwin is UNIX.
I don't know, but my experiences in this are as follows:
I tried Linux first in 1995, but it wasn't ready for me, and I wasn't ready for it. It got deleted.
Ditto in 1998. That was when KDE was in Alpha.
I was finally wooed (by screenshots and happy tales from people I met online) into buying SuSE in January of this year. I switched over.
My friends watched this process with interest. They came round, toyed with Linux etc. My Machead friend experimented with it for running his old iMac as OS X was too much of a dog performance wise for it. He tried lots of distros, but didn't really do any research (he tried Debian first ;) and he wasn't prepared to do any learning, and Linux still has a sharp learning curve for home desktop usage. Another friend of mine decided to turn an old box of his into a router/firewall for to share his home network - and also to use it to play with Linux on the desktop. As far as I know, he still uses XP on the desktop, but Linux is happily running a small server.
Another friend wanted to try it, but was prevented by the fact that Linux can't resize XP NTFS partitions yet. Finally, Hugh had a brother who was into it, and so he's tried it as well.
So far, I'm the only one who stuck with it, probably because I'm the most technical and everybody has a "switch" threshold, the point at which they are confident enough and Linux is easy/compatible enough for them to make the leap.
For most people it isn't there yet, hence the tiny (2-3%) market share it has on the desktop. BUT... the server end is often a way for people to get into it, as Ken is doing.
The corporate desktop would come first I think. Really we should be concentrating on that first, as the entry barrier for the corporate desktop is lower and the demand is higher (MS licensing etc). Home desktops will follow naturally after that.
I believe you can get them running on OSX, but why bother? They are seriously crippled on that platform due to lack of full POSIX compliance, and people buy OS X largely for the GUI. Anyway. Let's deal with your criticisms
First of all, the desktop icons... they look... weird. It's some weird angle on the icons. Also, the shoadow on the bottom of the icons don't make sense. The shadow on the front of the folders should be eliminated.
These folders are, fyi, SVG icons so they can be scaled to any size. Because this shot was of an early beta of GNOME2, not all the icons had been converted, hence the 2 different styles (both of which I think look quite nice really). I don't know what you mean about the shadows, the icons look fine to me. This really seems to be merely a personal aesthetic issue, which is fine, that's why we have icon themes.
Also, in the 'gfx' icon, I don't see what the slide should mean. The toolbar icons should be made a bit larger, or the toolbar itself should be made a bit smaller. Users are too apt to miss the button.
The slide - who knows? It was a demo of the different folder decorations you can have. There's no need to use it, if you don't think it's appropriate. The toolbars are fine, I use them all the time and you don't have to click directly on the image, as long as you're close it works OK.
Finally, the text rendering on the taskbar(which is not unique, but win95ish) looks weird. Look at 'Keyboard Accessibility Control'. The text should probably be squeezed or truncrated, but instead it clashes with the icon.
That was a minor bug that has since been fixed. I'm using it now, and the buttons are fine.
Oh yeah, the 'applications' menu, sounds macos-classic-ish, but I can't tell since I haven't actually seen it.
I've used Classic a bit, I don't recall any Applications menu. At least on the mac I used, everything was started either from the finder, or DragThing.
The point was not to hold up that particular screenshot as UI perfection itself, but to demonstrate that gnome has a fairly unique look to it.
Actually this isn't true of GNOME any more. The default layout is as far as I can tell unique. You can see a screenshot of it here.
There is a bar across the top Mac style (though that's the only similarity to the mac), and a taskswitcher with desktop switcher at the bottom. There are two menus, Applications and Actions (which makes sooo much sense I can't help but think, why did nobody think of this before?). You can add your own applets to the panels, or add your own. It's not only extremely flexible, but very intuitive as well.
Having said that, I wouldn't recommend GNOME2 for new users yet. I'm using it now, and it's interesting as much for what it lacks as for what it's got. It's clearly a fantastic foundation to build on, and I await the 2.2 release with interest. Hopefully Nautilus won't suck then.
But anyway. The old bottom panel with the foot system has gone (unless you use RedHat) - I suggest you check out those screenshots of Jimmacs. They are too sweet. And nothing like Windows or the Mac, or KDE.
This is good news, but without a big corp behind it I'm sceptical about whether hardware manufacturers will adopt it. What I'm more worried about now is Xiph themselves though. They've done a great job, and given us this fantastic gift, but now how do they make money? If the library was originally proprietary, then what do they have now?
Also, it should be noted that I've often talked with employees of the company. Of course they may just be pointlessly doom talking or whatever, but they told me it was nothing like the company it once was.
Well, whatever. I never worked for Jinc myself.
Indeed, as the engine of your car is so ugly, it can't possibly be any better than, say, a lava lamp. Clearly, as an engine is ugly, it sucks, so it should be replaced with a lava lamp, which isn't ugly, so it doesn't suck. Please tell me how you get on when you've replaced your car engine with a lava lamp.
As far as I know, this is just FUD. I've been tracking Jabber since the cows came home, as has Adam, and neither of us has heard anything about any trouble from (enforceable) patents on Jabber.
Having said that, it's probably true. Bear in mind any non trivial program will have patents on it within the US. There are hundreds of patents that affect Linux for instance. What can these companies say, except we won't use them?
Finally, Jabber Inc does have a business model. They sell industrial strength commercial servers and support. This does make money. The question is, will they make enough money to survive.
Unfortunately, J inc has sort of lost its way lately. Andre Durand who set it up has left, and it was taken over by his former bosses who didn't really have much of a business before. The company has been going downhill since then.... I hope they do survive, but it's not the company it once was. Oh, and a quick disclaimer : I now work for Andre, on an unrelated project ;)
I help run the theoretic.com with Adam Theo, which is most definately not a public testing server.
First up, we don't have AOL or ICQ transports, as we got too popular to remain unnoticed ages ago so it's firewalled now. The MSN transport still works fine though, and Yahoo, well, nobody used that :) You can always use an external transport (and many do)
Now for the good news. There are 2 admins in 2 separate timezones (London and Florida time), so there's normally somebody about if you want to chat to an admin. We don't track the latest releases unless we need a bugfix, so the server is stable. We haven't rebooted it since February as far as I can remember.
We also have some features other servers don't. For instance, the SMTP transport is mondo cool. If you send an email to user@im.theoretic.com, it'll be turned into a jabber message. If you reply, it'll be turned back into an email. You can add email addresses to your roster as well if you want to fire off quick messages.
Finally, we try to be as polite as possible. If we're restarting a transport for whatever reason, or need to reboot the server (rare now), we'll send a public broadcast message about 10 minutes before so you can finish up conversations etc.
RhymBox, which is a great little Windows client already, and the new version will totally rock, tied with theoretic should hopefully ease most of the pains that have put off non geeks from using jabber up until now.
Actually you are wrong. On the home user desktop maybe, but we're talking corporates here. To IT administrators, things like cost, stability, security, CPU efficiency, open source and unix underpinnings do in fact matter a great deal.
Especially the open source thing of course, as it means "no lockin" - virtually every IT using company today has been bitten on the backside by this, so that's a major win.
As an aside where do you see companies with 2ghz chips that just run Office? Most company machines I've seen are fairly low end by todays standard.
One day of course there will be Dreamweaver/Linux, but until that day, you can run Dreamweaver under Wine. It works just like in Windows, and gets some additional capabilities as well (like being able to read Linux disks). I believe Dreamweaver has a 4/5 star rating at the moment, so it's not perfectly there yet, but CodeWeavers are working hard on popular apps like these.
Show me where opensource benefits latin america. Medical records, state agencies, but until then this announcement has all the weight of Pam Anderson announcing her new fashion line.
I'm sorry, it sounds harsh, but as others have pointed out this is probably more important for the west than Latin America. If other countries see that Venezuela made the switch OK, then even if Venezueala doesn't save money (as they weren't paying for it before anyway) they will see that there is an opportunity for them to save money.
Domino effect. Latin America may be the start.
Here's a quick list off the top of my head:
So I think you can see .... we have a lot of work cut out for us :)
Sure. A lot of that was hype though. I mean the remaining problems are all toughies - stuff like good desktop integration, package management etc. When these problems are solved, Linux will be as good as (if not better) than Windows 98 in terms of user interface and consistancy. What is Windows 2000? To the average user, it's 98 that doesn't crash. What is XP? To the average user, it's 2000 with knobs on. 4 years on, MS has improved the reliability of Windows no end, as well as added lots of features (especially on the corporate end). For the large majority of desktop users though, the difference is minor - hence the slow upgrade rate. Most people still use Windows 98, and they'll upgrade to make it suck less, not to get the killer features (hehe) available in XP.
So I think it's absolutely possible to catch up to Windows. I know Windows won't be static. If Longhorn delivers, then once more Windows will be ahead of us in some areas. But we'll have caught up with XP by that point, so it's a straight race.
Finally, remember that the scale of Windows development scales linearly (ie they can only hire X new developers a year, if that), whereas Linux development scales exponentially. Especially as it's so easy to work on, the number of man hours being invested in Linux already easily outstrips that being put into Windows 2000 at its peak (about 5000 engineers). I think we can catch up.
When you're rich, the time and sacrifices needed for philosophy are cheap. When you're poor, the practical rules the roost.
Pithy comments aside, this only reinforces what I have come to believe in recent months: that the eventual dominance of Linux/open source is an economic inevitability.
The reasoning behind this is very simple, and has nothing to do with blind zealotry. Capitalism does not tolerate inefficiency. If you can do something better than your competitors, or if you can do the same thing but cheaper, you will have an advantage and the natural selection of the free market will elevate you above the rest.
Linux is more efficient in economic terms. Right now of course, it's "efficiency" is being held back by the number of rough edges that need polishing, the huge resources needed to overcome Microsoft lockin and so on. However, these are becoming less and less all the time. Eventually (like within a few years) Linux will be as good as Windows, as well as compatible with it thanks to the efforts of the wine/samba/OpenOffice/NTFS crews. At that point, you can be better and cheaper at the same time by using it. The result? Market dominance.
It has another advantage as well - multiple vendors. History shows that economics favours systems with multiple vendors: witness Macs vs PCs, or VHS vs Betamax.
Unfortunately too many countries take the minimal government approach too far. Capitalism optimizes for efficiency and little else. Social justice, the environment, humanity - it is all secondary to efficiency: ie producing the greatest amount of profit.
Minimal government involvement in business is a bad idea, because capitalism only serves society when it is bracketed and constrained by that society and directed towards a greater good. It's similar to electricity - it has raw power that should be respected and harnessed, but that's not an excuse for minimal insulation or lack of safety features, though they do make it harder to use.
Business, by itself, looks out only for itself. No, this is not good, this is useless. If we accept the proposition that the economy is not the be-all and end-all of society, but in fact is meant to serve us by sustaining and sometimes increasing our standard of living, then we can see that the trick is not to get minimal government interference, it's to get the right amount of government interference.
Only then can capitalism serve us, rather than the other way around.
A laugh or two at some American excess is one thing, but to portray America as malign is just the opposite of reality.
Firstly, before I start, don't get me wrong. I'm not saying America is evil or anything. But I'd like to make a few points, America is hardly saintly either:
Evil? No. But where do you draw the line between deliberate damage through rampant self interest and evil anyway?
Sorry, but had to get that off my chest. Politics is important, games or no. I'm not saying other countries are perfect either, far from it.
Finally, political views are frequently expressed through comedy as well. Is that infantile?