For instance, fresco has subpixel addressing, a scene graph based windowing model etc. These aren't buzzwords, but I can't really explain them all in depth, i'm too tired right now.
Fresco is based on corba which is implicitly multilanguage, so that's not just picogui.
NeWS was an early attempt but it had design issues and never really went anywhere.
Yeah, I'll second the last comment. It's good to see people trying out new ideas. I suspect that soon X will have most of the features people want in the short term, and we'll all be happy, but that doesn't mean X is perfect and cannot be improved upon. It's good to know people are thinking about a future beyond "we want transparency".
X is just now getting anti alaised fonts and everyone is saying X is so great, we are about a year away from the release of Xfree5.0 which is supposed to have the finished Xrender, only one guy is working on Xrender (Keith Packard)
RENDER is basically done. What you're thinking of is, we don't have true transparency yet. Well what do you know, that's because
a) Transparency is useless for virtually anything except screenshots, making a list of places where it enhances usability for instance is very hard.... and
b) X has bigger issues which need resolving first, like on the fly resolution switching (R&R, done), decent cursors (XCursor, done), reducing the amount of configuration work needed (ongoing, the aim is to eliminate the XF86Config file eventually).
And FYI the Matrox drivers already have hardware accelerated alpha blending.
People make such a big deal of having "true" transparency, but you know what? I'd probably turn most of it off. Truly transparent terminal windows I find are harder to use than the desktop wallpaper transparency that KDE and GNOME use. One shows you all the text, lines etc, the latter just gives you a nice working background. Maybe if they were blended as well to make the background less distracting, that'd be cool, but at the end of the day it's just FUN, not USEFUL.
Imagine I'm a Linux and BSD poweruser/sysadmin and the problem is related to Windoze or Mac. I honestly don't know that much about those systems (nor do I care to learn) so there's no 'pretend'.
That's how it is with my family now. A few years ago now when Code Red and Melissa were going around, my father came to me with a concerned look on his face and asked if I could recommend a virus checker. "Why do you want one?" I asked, as he is by no means stupid and knows not to run random programs from strangers etc.
"I've heard there are these new email viruses that you only have to open to get infected with", he said. "Sure, but you don't need a virus checker, just don't use Outlook Express". At the time I was using MozMail, which worked great for me, and I get way more mail than he does, so I thought that would be fine. But no, he didn't even want to hear that idea. "No, I like Outlook Express, I know you don't like Microsoft but I'm more reasonable".
So what happens? Of course he gets a virus and wants me to help remove it. So I do, he's my father and he's helped me god knows enough times. I look up the virus specs on the net, remove it and send him on his way, along with an admonition - "See, I told you using OE was a bad idea. Are you sure you don't want to try the competition". No, he didn't. He didn't even want to experiment, look at the competition even. He was sure Outlook Express was fine, and this was just a one off.
About 6 months later it happens again. By this time I've been using Linux long enough that I don't really understand the virus specs anymore, and can't really be bothered leaving the comfy seat in front of the TV to remove a Windows virus *again* (i'd done it several times in the last few months for brothers/friends etc). Of course he gets pissed off, tells me I should help him because he's my father, tells me to drop the stupid hate of Microsoft just because they are successful etc.
Sorry, but nobody is obliged to give out free tech support time and time again when your advice on how to save money and hassle is ignored. He knows now that I'm happy to help with other stuff, but if he has virus problems he's on his own. And he still uses Outlook Express. He just got an expensive virus scanner, that must be used about once a week, if that often.
Moral of the story: only give out tech support as long as it's appreciated and they are willing to learn from the experience. If they aren't, you're just wasting your time.
Would somebody please explain where people get insulted on IRC? I regularly help in #linuxhelp on FreeNode and can honestly say that I have never seen this kind of unjustified newbie bashing. Is this when people use DALnet or EFnet or something? On the rare occasions when I do try to connect to those networks they are indeed full of idiots, but freenode and gimpnet are an absolute treasuretrove. I know I would still be learning the CLI the hard way if it weren't for those guys.
Not true. In fact, there is a COM implemented for UNIX called XPCOM. It's the foundation of Mozilla. In addition, COM itself is a copy of CORBA/IDL (the tools for Corba/IDL are fairly poor tho). Gnome uses CORBA, and KDE also uses a distributed object protocol. In additional, almost every language supports SOAP (including.NET) and Perl's.NET support is awesome.
The power of Windows COM comes from the fact that it's a) universal and b) capable of both inproc and outproc coding.
If I load up Delphi and write a ActiveX (com with extensions) object, it can then be used in any language easily. A bit more work and now it's available for automation by external processes that may or may not be local. The tools are there. It's universal.
An example. For the longest time, almost every Jabber client implemented on Windows used JabberCOM. Whether they were written in Delphi (the language jabbercom was written in), C++, Visual Basic or even JavaScript (check out RhymBox) they all used the same code. That stuff just doesn't happen on Linux. There are several competing component models, all of which suck.
CORBA sucks, it's great for outproc communications, but for inproc code it's awful. Miguel himself has admitted as such, and he was a prime CORBA advocate for many years. KDE doesn't really even have an equivalent, KParts is limited to inproc C++ and DCOP is just a simple RPC protocol. UNIX still lacks a decent object model. Windows has always had COM, and even though it's far from perfect, it was there.
Now MS are at it again, they are addressing the problems that made all these hacks necessary in the first place by standardising the in memory object layouts and ABIs. Once more, they are taking object technology to the next level. None of these ideas are innovative (COM was bought in), but they are setting standards and people use them, and all Windows developers benefit as a result. We need something like that for Linux, but so far all our attempts have failed. I'm still not sure whether to trust MS with.NET, but it's a good solution to the problem, and we need a solution. Bring it on I say.
Thank Anders Hejlsberg, chief architect of the C# language, instead.
It's interesting to compare his previous work (Delphi) work.NET
Delphi was and still is rightly praised for being incredibly kickass. Outside of indie Windows development, Pascal is a dead language, but there is a truly astonishing amount of code written in Object Pascal for Windows - even Microsoft have used it for some products.
So I found it interesting to do a comparison. He clearly took some ideas from it to integrate into C#. For instance language integrated properties are something I constantly wish that Java had. The "uses" clause became "using".
Anders pats himself on the back for using XML comment tags though. Big mistake IMHO, xml comments are one of the most irritating things about C#. The start and end tags serve no purpose except to make the file larger and harder to read. JavaDocs are more elegant and easier to write. Using XML in many cases is a good decision, but here it is not.
But then there are some curious omissions. Delphi doesn't use an IDE that generates gobs of code for user interfaces, it uses binary resources that are reconstructed from data embedded in the EXE/DLL files at runtime by the class library. It's clean and elegant. But what does VS.NET use? IDE generated code of course.
I guess it's only realistic that most of the innovation in.NET comes from perhaps one or two people, while the rest merely reinvents what came before. After all, the reason that innovation (whatever that is) is so highly prized is that it's rare.
This isn't a case of poor misunderstood Microsoft, either. Why else would they codename the original Millennium JVM "Borg" (http://research.microsoft.com/research/sn/)?
[sigh] It's called Borg because they were researching self tuning distributed systems, which is how the fictional Borg were supposed to operate. It is NOT called Borg because they wish to take over the world.
Marketshare is the percentage of sales, either by unit count or revenue of a given market. Typically this is done in a quarterly or annual basis.
Well, you are quite correct, I went and looked it up. I was confusing the two, though I normally see marketshare and installed base treated as the same by others, hence my confusion.
So right now, Apple has roughly 3-4% marketshare depending on whose numbers you trust. That means, out of the 100% of machines sold in the past quarter, Apple accounts for roughly 3-4%
So does that mean you can't count things that are given away for free as marketshare? If I download Redhat 8, does that download count as part of their marketshare or not?
He didn't want to pay for Partition Magic. He was curious like most of them, but was happy with Windows and didn't want to pay for another OS, including utilities
Marketshare is not the same thing as installed base, and you interchange the two. Macintoshes currently ship with Mac OS X as the default operating system, so out of the box 100% of these machines are running Mac OS X
Huh? I thought market share was how much of the market used a particular product. So if 4% of the computer desktop market uses Macs, but only 2% of those machines use OS X, the the market share for the Mac is 4%, and the marketshare for OS X is 2%. Am I missing something here?
The next two things cannot be easily measured. The number of people that do not run Mac OS X and always switch back to Mac OS 9 on their new Macs. Plus, some people switch between the two and spend more time in one or the other. The other difficult to measure item is not only the number of desktop Linux seats, but the number of them that spend a majority of their time in Linux. I would bet that a majority of the Linux desktop seats out there spend a significant amount of time, if not a majority of time booted into Windows.
When those Linux stats were measured, it was what you use the majority of the time iirc. So for instance I use Linux perhaps 90% of the time, but still use Windows occasionally. I'd be measured as a Linux user. If you count Linux desktop seat as any machine with a desktop distro installed, then you might well be correct (though i'd guess not, after all if you hardly ever use it why keep it installed?).
1) No, not all Linux UIs look like Windows. Check out this for an ironic screenshot - that's RedHat 8 in GNOME2. GNOME doesn't really look much like anything else, yet is still usable. If you want something a bit more far out, try Enlightenment or Ion (no windows for them). You might think that is ugly, but whatever floats your boat you know? Clearly they liked it that way.
2) Macs are not easy to use for newbies. I wish more people would realise this. Pretty much all newbies on Macs I've observed have had big problems, because they are used to Windows. Note: these are newbies using other peoples Macs, not people who have bought them, so they are under less psychological pressure to adapt. They have problems with the mouse, the keyboard, the "closing window doesn't quit app" issue and more. Macs have high internal consistency, which means once you've got used to its quirks and habits, those quirks tend to stay the same.
3) There are Linux HIG guidelines, the GNOME project created them but they are available for everyone to use. A lot of development software (non gnome software) is starting to use them, for instance xchat and gaim CVS are both partially compliant. Expect to see more in future.
4) Apple themselves blatantly ignore their own HIG. The whole brushed-metal justification was inserted into the HIG after 10.2 was shipped, with pathetically weak guidance - "something that resembles a real world object". The number of contradictions in Apples own software is phenomenal with respect to this.
Ah, yes: we must maintain our choices. No consistent interface for us. Long live the Revolution.
That's like claiming all cars should have manual gearboxes, or they should all be automatics, because that way they are consistant. Oh yeah, make sure they're all identically sized as well. Multiple user interfaces increase usability, because instead of people adapting their own habits to fit the machine, the machine can be adapted to fit theirs.
I can't recall exactly when, about 4/5 months ago perhaps, Apple were under pressure from MS to sell more copies of OS X because Office X wasn't recouping costs. Remember that?
Anyway, at the time, Mac marketshare was about 4% as measured by various independant statistics bodies and (of course;) Google. At the time, Apple claimed 1 in 10 Mac users had upgraded to OS X. Linux market share was a fraction under 2%. If 1/10ths of the Mac market at 4% had upgraded, that gives a market share for OS X as being 0.4% right? That means that desktop Linux had about 4x the market share of MacOS. Now, I've been generous in the last post and said double, because undoubtably since then as more apps have been ported to OS X and so on more Macs will be running os x than were previously. I'm also assuming Linux hasn't grown in that time at all (unlikely, but we can let it stand for now).
Hence my original assertion. I hope that satisfies you, I could dig out the actual URLs to Apples figures and such but Slashdot covered it at the time and I'm going out soon.
Everybody now is talking about Apple. Sure, It's pretty cool that Macs run Unix, but well, it's not Linux is it?
Geeks are now starting to get Macs, instead of developing for Linux they're playing with all the cool stuff on their mac.
I know it might seem like that from reading Slashdot, buy my experiences in the Real World don't bear it out.
Back home, my best mate was a massive Mac advocate. When I went round to his house, he'd always show me a Mac advert. He loved the Switch campaign of course. He was always telling people, even strangers, how great Apple hardware was. I used his Mac quite a bit, all my circle of friends used it at least once, many more times for me. Out of my friends who were geeks (about 4 or 5 of us:) they have all installed Linux at some point. Did they all stick with it? No. Andrew tried a live CD but his disk was all NTFS so he was a bit stuck. Ken set it up as a small server to allow net connection sharing (he's now using it also at university). Hugh used it because his brother used it. Even Paul, the huge Mac advocate, has installed it 3 times (he got pissed off at the poor PPC support each time though;).
Then I moved away from home, I got a job at a research company. It used to be Ministry of Defence research before it was privatised. Inside the department, Linux is slowly taking over. There are about 25 of us who joined together, they are my new friends now. I use Linux on the desktop all the time, several others use it part time (rebooting for games and such). Now Dave is getting interested, he was enthusing to me today about Knoppix (he didn't want to disturb his current setup). The company I work in is full of geeks, some of them not badly off at all.
How many Macs have I seen since I left Paul behind? None. Zilch. Zero. How many Linux boxes? Loads.
It's easy to forget that it appears that MacOS is losing market share, and by Apples own admission a few months ago, Linux has nearly double the desktop market share of OS X. Hard to believe isn't it, but outside of Slashdot, there are millions of people with PCs, who don't like Windows and want to try something different. It's easy to try Linux, it costs nothing, and isn't a huge decision. Install it in an evening, try it, if you don't like it, remove it and go back to Windows. Try again in a year or two.
That last sentance is the crucial one, I've seen lots of people try Linux and go back to Windows. But they always try again. And again. I tried Linux 3 times before dropping Windows.
Bah. Mac running unix might sound like a good thing, but all we're going to end up with is Apple as the new M$ instead.
I just think it's a really bad thing for the whole Linux community.
Don't worry:) It's alright, Linux is chugging along just nicely. The Mac isn't, and never was, a realistic proposition for most people. That's why it has 4% of the market (far less if you only count os x machines), when it originally had more like 40%. In much the same way that the PC won out over the Mac in the early/mid 90s because the PC was economically if not technically superior (competition drove down margins, powered massive speed increases and so on), Linux gains from the same effect. The presence of competition is usually beneficial to a market as a whole.
For the small number of geeks that decide to stop kernel hacking in favour of writing a new Cocoa IRC client or whatever, they will be replaced 2 times over by newcomers and business. There aren't any flash adverts about it, but the statistics stand.
Config files in *nix are often inconsistent and obscure. Not that hairy, undocumented registry keys are any better.
Hmm, about about clean, documented registry keys? That's the approach GConf takes, it has a lot of advantages over raw text. Too bad all the keys aren't documented at the moment, but that's bugs with individual apps rather than a design flaw.
"Eat your own dogfood" is a popular saying both at Microsoft and other tech companies such as Netscape (search bugzilla.mozilla.org for nsDogfood and nsCatfood)
does quicktime work in mozilla on a *nix system (other than osx obviously)? i don't think so.
Yes, just install CrossOver if you must have QuickTime. I believe Xine has integration with Mozilla to some extent. But who actually uses QuickTime apart from Apple and people who have signed "content deals" with Apple anyway? I can't remember the last time I saw a QT movie that wasn't on Apples site.
do webpages look prettier in mozilla (on any system)? nope.
Yes. Mozilla antialiases text far better than IE on Windows. I pigged about for ages when I was using XP trying to make it antialias text nicely. I'm sure there's a way, but I never found it.
is there a better spreadsheet program than excel? no way.
I'm not a business type, so Gnumeric solves all of my needs. If you are in the minority of people who needs an ultra powerful spreadsheet and you refuse to use OpenOffice, you can run MS Excel under Linux.
in general, is windows easier to use for a huge gigantic percentage of the population? yes.
Worthless statement, as ease of use is entirely subjective. People find what they're used to easy regardless of how insanely it may be designed.
do companies want to pay someone to fix problems that might arise in a *nix system (again other than osx)? never.
Confusing statement. You can't pay people to fix stuff in OS X, it's proprietary. And yes, at my company we find issues in Windows or some other commercial product we'd like to fix but can't all the time.
Not quite accurate. I think the ReiserFS bugs in the earlier 2.4 kernels was due to the vastly different algorithms they used exposing obscure bugs in the kernel itself, as well as buggy chipsets. Those kernel bugs seem to have largely been shaken out now, and RFS is pretty solid.
Sure, but I'd raise a few issues with that. The first one is that if BeOS gets larger, you will start seeing the kind of splits that so characterise Linux at the moment. In fact, we're already seeing this, with different groups attacking the problem of resurrecting BeOS in different ways, and none of them are even at version 1 yet!
It's taken over a decade of effort for Linux to get to where it is today, and over that time as more and more people got involved, surprise surprise not all of them shared exactly the same vision. It hasn't all gone smoothly - look at the KDE/GNOME split for what can happen. Luckily Linux was designed for modularity right from the word go, and can just about cope with it. In some cases the interfaces need to be better defined, especially on the desktop, but that's happening (see freedesktop.org)
I wouldn't rate BeOSs chances at that. It was designed on the fundamental assumption that it'd have one overriding organisation in control. Without that, it's going to be hard to really move things forward once they reach parity with the old BeOS, in much the same way that once Linux blew away UNIX state of the art things began fracturing until standards bodies started forming to get stuff back under control.
Anyway, why bother? BeOS had some nice ideas, but they're being reimplemented in Linux right now. For BeFS done properly, see ReiserFS. No, it can't do live queries today, but that's because Reiser and the team have spent years getting performance to a level where they can actually do justice to the original vision of the BeFS team (who iirc were severely limited by performance problems). If you simply must have the BeOS gui that can be implemented as an X desktop environment, but with good hardware support.
The stuff you mentioned about consistency and clean design only applies when there are very few people involved, the moment you get lots of people working on it, they have different ideas or want to experiment and suddenly things aren't so consistent anymore. And of course everything gets cruft, what happens when a few years down the line the OpenBeOS people come up with a great new widget toolkit design. They have to maintain backwards compatability though.... oops, suddenly you've got cruft and inconsistency. That's not a good reason for starting from scratch.
Actually that's just in the US. In Europe SuSE is probably most used in the corporate world. I'm using version 8 of the 'hat here at work right now to type this, but all the other Linux servers and desktops in this department (and there are several) are SuSE.
This is precisely why I should support the platform which does without silly DRM garbage
Loose thinking I'm afraid. Actually, the Mac does indeed have some DRM in it, see QuickTime for instance. More to the point, Apple can reverse their stance at any time they like, there are no guarantees that in the future the iPod will make do with only a sticker. If you buy a Mac because it doesn't have DRM in it today, you're inherantly taking a chance by assuming there won't be any tomorrow, or the day after, or the day after that.
Partly it depends on the apps. Running WinCVS and ICQ at the same time as Delphi on Windows 98 kills it within minutes, because one of them (i suspect wincvs) leaks resources like a sieve. Is this my fault for running known buggy apps, is it the apps fault for leaking resources, or is the fault of the OS for having braindead design in the first place?
Slashdot Mirror
For instance, fresco has subpixel addressing, a scene graph based windowing model etc. These aren't buzzwords, but I can't really explain them all in depth, i'm too tired right now.
Fresco is based on corba which is implicitly multilanguage, so that's not just picogui.
NeWS was an early attempt but it had design issues and never really went anywhere.
Yeah, I'll second the last comment. It's good to see people trying out new ideas. I suspect that soon X will have most of the features people want in the short term, and we'll all be happy, but that doesn't mean X is perfect and cannot be improved upon. It's good to know people are thinking about a future beyond "we want transparency".
RENDER is basically done. What you're thinking of is, we don't have true transparency yet. Well what do you know, that's because
a) Transparency is useless for virtually anything except screenshots, making a list of places where it enhances usability for instance is very hard.... and
b) X has bigger issues which need resolving first, like on the fly resolution switching (R&R, done), decent cursors (XCursor, done), reducing the amount of configuration work needed (ongoing, the aim is to eliminate the XF86Config file eventually).
And FYI the Matrox drivers already have hardware accelerated alpha blending.
People make such a big deal of having "true" transparency, but you know what? I'd probably turn most of it off. Truly transparent terminal windows I find are harder to use than the desktop wallpaper transparency that KDE and GNOME use. One shows you all the text, lines etc, the latter just gives you a nice working background. Maybe if they were blended as well to make the background less distracting, that'd be cool, but at the end of the day it's just FUN, not USEFUL.
I don't like IDE generated code because it means you have to use that IDE to edit it - kind of tool lockin if you like.
That's how it is with my family now. A few years ago now when Code Red and Melissa were going around, my father came to me with a concerned look on his face and asked if I could recommend a virus checker. "Why do you want one?" I asked, as he is by no means stupid and knows not to run random programs from strangers etc.
"I've heard there are these new email viruses that you only have to open to get infected with", he said. "Sure, but you don't need a virus checker, just don't use Outlook Express". At the time I was using MozMail, which worked great for me, and I get way more mail than he does, so I thought that would be fine. But no, he didn't even want to hear that idea. "No, I like Outlook Express, I know you don't like Microsoft but I'm more reasonable".
So what happens? Of course he gets a virus and wants me to help remove it. So I do, he's my father and he's helped me god knows enough times. I look up the virus specs on the net, remove it and send him on his way, along with an admonition - "See, I told you using OE was a bad idea. Are you sure you don't want to try the competition". No, he didn't. He didn't even want to experiment, look at the competition even. He was sure Outlook Express was fine, and this was just a one off.
About 6 months later it happens again. By this time I've been using Linux long enough that I don't really understand the virus specs anymore, and can't really be bothered leaving the comfy seat in front of the TV to remove a Windows virus *again* (i'd done it several times in the last few months for brothers/friends etc). Of course he gets pissed off, tells me I should help him because he's my father, tells me to drop the stupid hate of Microsoft just because they are successful etc.
Sorry, but nobody is obliged to give out free tech support time and time again when your advice on how to save money and hassle is ignored. He knows now that I'm happy to help with other stuff, but if he has virus problems he's on his own. And he still uses Outlook Express. He just got an expensive virus scanner, that must be used about once a week, if that often.
Moral of the story: only give out tech support as long as it's appreciated and they are willing to learn from the experience. If they aren't, you're just wasting your time.
Would somebody please explain where people get insulted on IRC? I regularly help in #linuxhelp on FreeNode and can honestly say that I have never seen this kind of unjustified newbie bashing. Is this when people use DALnet or EFnet or something? On the rare occasions when I do try to connect to those networks they are indeed full of idiots, but freenode and gimpnet are an absolute treasuretrove. I know I would still be learning the CLI the hard way if it weren't for those guys.
The power of Windows COM comes from the fact that it's a) universal and b) capable of both inproc and outproc coding.
If I load up Delphi and write a ActiveX (com with extensions) object, it can then be used in any language easily. A bit more work and now it's available for automation by external processes that may or may not be local. The tools are there. It's universal.
An example. For the longest time, almost every Jabber client implemented on Windows used JabberCOM. Whether they were written in Delphi (the language jabbercom was written in), C++, Visual Basic or even JavaScript (check out RhymBox) they all used the same code. That stuff just doesn't happen on Linux. There are several competing component models, all of which suck.
CORBA sucks, it's great for outproc communications, but for inproc code it's awful. Miguel himself has admitted as such, and he was a prime CORBA advocate for many years. KDE doesn't really even have an equivalent, KParts is limited to inproc C++ and DCOP is just a simple RPC protocol. UNIX still lacks a decent object model. Windows has always had COM, and even though it's far from perfect, it was there.
Now MS are at it again, they are addressing the problems that made all these hacks necessary in the first place by standardising the in memory object layouts and ABIs. Once more, they are taking object technology to the next level. None of these ideas are innovative (COM was bought in), but they are setting standards and people use them, and all Windows developers benefit as a result. We need something like that for Linux, but so far all our attempts have failed. I'm still not sure whether to trust MS with .NET, but it's a good solution to the problem, and we need a solution. Bring it on I say.
It's interesting to compare his previous work (Delphi) work .NET
Delphi was and still is rightly praised for being incredibly kickass. Outside of indie Windows development, Pascal is a dead language, but there is a truly astonishing amount of code written in Object Pascal for Windows - even Microsoft have used it for some products.
So I found it interesting to do a comparison. He clearly took some ideas from it to integrate into C#. For instance language integrated properties are something I constantly wish that Java had. The "uses" clause became "using".
Anders pats himself on the back for using XML comment tags though. Big mistake IMHO, xml comments are one of the most irritating things about C#. The start and end tags serve no purpose except to make the file larger and harder to read. JavaDocs are more elegant and easier to write. Using XML in many cases is a good decision, but here it is not. But then there are some curious omissions. Delphi doesn't use an IDE that generates gobs of code for user interfaces, it uses binary resources that are reconstructed from data embedded in the EXE/DLL files at runtime by the class library. It's clean and elegant. But what does VS.NET use? IDE generated code of course.
I guess it's only realistic that most of the innovation in .NET comes from perhaps one or two people, while the rest merely reinvents what came before. After all, the reason that innovation (whatever that is) is so highly prized is that it's rare.
[sigh] It's called Borg because they were researching self tuning distributed systems, which is how the fictional Borg were supposed to operate. It is NOT called Borg because they wish to take over the world.
Get a grip!
Well, you are quite correct, I went and looked it up. I was confusing the two, though I normally see marketshare and installed base treated as the same by others, hence my confusion.
So right now, Apple has roughly 3-4% marketshare depending on whose numbers you trust. That means, out of the 100% of machines sold in the past quarter, Apple accounts for roughly 3-4%
So does that mean you can't count things that are given away for free as marketshare? If I download Redhat 8, does that download count as part of their marketshare or not?
thanks -mike
He didn't want to pay for Partition Magic. He was curious like most of them, but was happy with Windows and didn't want to pay for another OS, including utilities
Huh? I thought market share was how much of the market used a particular product. So if 4% of the computer desktop market uses Macs, but only 2% of those machines use OS X, the the market share for the Mac is 4%, and the marketshare for OS X is 2%. Am I missing something here?
The next two things cannot be easily measured. The number of people that do not run Mac OS X and always switch back to Mac OS 9 on their new Macs. Plus, some people switch between the two and spend more time in one or the other. The other difficult to measure item is not only the number of desktop Linux seats, but the number of them that spend a majority of their time in Linux. I would bet that a majority of the Linux desktop seats out there spend a significant amount of time, if not a majority of time booted into Windows.
When those Linux stats were measured, it was what you use the majority of the time iirc. So for instance I use Linux perhaps 90% of the time, but still use Windows occasionally. I'd be measured as a Linux user. If you count Linux desktop seat as any machine with a desktop distro installed, then you might well be correct (though i'd guess not, after all if you hardly ever use it why keep it installed?).
1) No, not all Linux UIs look like Windows. Check out this for an ironic screenshot - that's RedHat 8 in GNOME2. GNOME doesn't really look much like anything else, yet is still usable. If you want something a bit more far out, try Enlightenment or Ion (no windows for them). You might think that is ugly, but whatever floats your boat you know? Clearly they liked it that way.
2) Macs are not easy to use for newbies. I wish more people would realise this. Pretty much all newbies on Macs I've observed have had big problems, because they are used to Windows. Note: these are newbies using other peoples Macs, not people who have bought them, so they are under less psychological pressure to adapt. They have problems with the mouse, the keyboard, the "closing window doesn't quit app" issue and more. Macs have high internal consistency, which means once you've got used to its quirks and habits, those quirks tend to stay the same.
3) There are Linux HIG guidelines, the GNOME project created them but they are available for everyone to use. A lot of development software (non gnome software) is starting to use them, for instance xchat and gaim CVS are both partially compliant. Expect to see more in future.
4) Apple themselves blatantly ignore their own HIG. The whole brushed-metal justification was inserted into the HIG after 10.2 was shipped, with pathetically weak guidance - "something that resembles a real world object". The number of contradictions in Apples own software is phenomenal with respect to this.
Ah, yes: we must maintain our choices. No consistent interface for us. Long live the Revolution.
That's like claiming all cars should have manual gearboxes, or they should all be automatics, because that way they are consistant. Oh yeah, make sure they're all identically sized as well. Multiple user interfaces increase usability, because instead of people adapting their own habits to fit the machine, the machine can be adapted to fit theirs.
Anyway, at the time, Mac marketshare was about 4% as measured by various independant statistics bodies and (of course ;) Google. At the time, Apple claimed 1 in 10 Mac users had upgraded to OS X. Linux market share was a fraction under 2%. If 1/10ths of the Mac market at 4% had upgraded, that gives a market share for OS X as being 0.4% right? That means that desktop Linux had about 4x the market share of MacOS. Now, I've been generous in the last post and said double, because undoubtably since then as more apps have been ported to OS X and so on more Macs will be running os x than were previously. I'm also assuming Linux hasn't grown in that time at all (unlikely, but we can let it stand for now).
Hence my original assertion. I hope that satisfies you, I could dig out the actual URLs to Apples figures and such but Slashdot covered it at the time and I'm going out soon.
But do you like it? I think it's like, most likely, you and her seem like very much alike to me :)
Back home, my best mate was a massive Mac advocate. When I went round to his house, he'd always show me a Mac advert. He loved the Switch campaign of course. He was always telling people, even strangers, how great Apple hardware was. I used his Mac quite a bit, all my circle of friends used it at least once, many more times for me. Out of my friends who were geeks (about 4 or 5 of us :) they have all installed Linux at some point. Did they all stick with it? No. Andrew tried a live CD but his disk was all NTFS so he was a bit stuck. Ken set it up as a small server to allow net connection sharing (he's now using it also at university). Hugh used it because his brother used it. Even Paul, the huge Mac advocate, has installed it 3 times (he got pissed off at the poor PPC support each time though ;).
Then I moved away from home, I got a job at a research company. It used to be Ministry of Defence research before it was privatised. Inside the department, Linux is slowly taking over. There are about 25 of us who joined together, they are my new friends now. I use Linux on the desktop all the time, several others use it part time (rebooting for games and such). Now Dave is getting interested, he was enthusing to me today about Knoppix (he didn't want to disturb his current setup). The company I work in is full of geeks, some of them not badly off at all.
How many Macs have I seen since I left Paul behind? None. Zilch. Zero. How many Linux boxes? Loads.
It's easy to forget that it appears that MacOS is losing market share, and by Apples own admission a few months ago, Linux has nearly double the desktop market share of OS X. Hard to believe isn't it, but outside of Slashdot, there are millions of people with PCs, who don't like Windows and want to try something different. It's easy to try Linux, it costs nothing, and isn't a huge decision. Install it in an evening, try it, if you don't like it, remove it and go back to Windows. Try again in a year or two.
That last sentance is the crucial one, I've seen lots of people try Linux and go back to Windows. But they always try again. And again. I tried Linux 3 times before dropping Windows.
Bah. Mac running unix might sound like a good thing, but all we're going to end up with is Apple as the new M$ instead. I just think it's a really bad thing for the whole Linux community.
Don't worry :) It's alright, Linux is chugging along just nicely. The Mac isn't, and never was, a realistic proposition for most people. That's why it has 4% of the market (far less if you only count os x machines), when it originally had more like 40%. In much the same way that the PC won out over the Mac in the early/mid 90s because the PC was economically if not technically superior (competition drove down margins, powered massive speed increases and so on), Linux gains from the same effect. The presence of competition is usually beneficial to a market as a whole.
For the small number of geeks that decide to stop kernel hacking in favour of writing a new Cocoa IRC client or whatever, they will be replaced 2 times over by newcomers and business. There aren't any flash adverts about it, but the statistics stand.
Hmm, about about clean, documented registry keys? That's the approach GConf takes, it has a lot of advantages over raw text. Too bad all the keys aren't documented at the moment, but that's bugs with individual apps rather than a design flaw.
Why does Java not qualify for that?
"Eat your own dogfood" is a popular saying both at Microsoft and other tech companies such as Netscape (search bugzilla.mozilla.org for nsDogfood and nsCatfood)
Wow, thanks! I have a new favourite putdown!
Yes, just install CrossOver if you must have QuickTime. I believe Xine has integration with Mozilla to some extent. But who actually uses QuickTime apart from Apple and people who have signed "content deals" with Apple anyway? I can't remember the last time I saw a QT movie that wasn't on Apples site.
do webpages look prettier in mozilla (on any system)? nope.
Yes. Mozilla antialiases text far better than IE on Windows. I pigged about for ages when I was using XP trying to make it antialias text nicely. I'm sure there's a way, but I never found it.
is there a better spreadsheet program than excel? no way.
I'm not a business type, so Gnumeric solves all of my needs. If you are in the minority of people who needs an ultra powerful spreadsheet and you refuse to use OpenOffice, you can run MS Excel under Linux.
in general, is windows easier to use for a huge gigantic percentage of the population? yes.
Worthless statement, as ease of use is entirely subjective. People find what they're used to easy regardless of how insanely it may be designed.
do companies want to pay someone to fix problems that might arise in a *nix system (again other than osx)? never.
Confusing statement. You can't pay people to fix stuff in OS X, it's proprietary. And yes, at my company we find issues in Windows or some other commercial product we'd like to fix but can't all the time.
Not quite accurate. I think the ReiserFS bugs in the earlier 2.4 kernels was due to the vastly different algorithms they used exposing obscure bugs in the kernel itself, as well as buggy chipsets. Those kernel bugs seem to have largely been shaken out now, and RFS is pretty solid.
It's taken over a decade of effort for Linux to get to where it is today, and over that time as more and more people got involved, surprise surprise not all of them shared exactly the same vision. It hasn't all gone smoothly - look at the KDE/GNOME split for what can happen. Luckily Linux was designed for modularity right from the word go, and can just about cope with it. In some cases the interfaces need to be better defined, especially on the desktop, but that's happening (see freedesktop.org)
I wouldn't rate BeOSs chances at that. It was designed on the fundamental assumption that it'd have one overriding organisation in control. Without that, it's going to be hard to really move things forward once they reach parity with the old BeOS, in much the same way that once Linux blew away UNIX state of the art things began fracturing until standards bodies started forming to get stuff back under control.
Anyway, why bother? BeOS had some nice ideas, but they're being reimplemented in Linux right now. For BeFS done properly, see ReiserFS. No, it can't do live queries today, but that's because Reiser and the team have spent years getting performance to a level where they can actually do justice to the original vision of the BeFS team (who iirc were severely limited by performance problems). If you simply must have the BeOS gui that can be implemented as an X desktop environment, but with good hardware support.
The stuff you mentioned about consistency and clean design only applies when there are very few people involved, the moment you get lots of people working on it, they have different ideas or want to experiment and suddenly things aren't so consistent anymore. And of course everything gets cruft, what happens when a few years down the line the OpenBeOS people come up with a great new widget toolkit design. They have to maintain backwards compatability though.... oops, suddenly you've got cruft and inconsistency. That's not a good reason for starting from scratch.
Actually that's just in the US. In Europe SuSE is probably most used in the corporate world. I'm using version 8 of the 'hat here at work right now to type this, but all the other Linux servers and desktops in this department (and there are several) are SuSE.
Loose thinking I'm afraid. Actually, the Mac does indeed have some DRM in it, see QuickTime for instance. More to the point, Apple can reverse their stance at any time they like, there are no guarantees that in the future the iPod will make do with only a sticker. If you buy a Mac because it doesn't have DRM in it today, you're inherantly taking a chance by assuming there won't be any tomorrow, or the day after, or the day after that.
Partly it depends on the apps. Running WinCVS and ICQ at the same time as Delphi on Windows 98 kills it within minutes, because one of them (i suspect wincvs) leaks resources like a sieve. Is this my fault for running known buggy apps, is it the apps fault for leaking resources, or is the fault of the OS for having braindead design in the first place?