Not as developer friendly? You've got to be kidding, being developer friendly was NeXTStep's raison d'etre. If you think that you can't do what.Net does on a Mac, you haven't seen WebObjects, which has been doing the web based app server stuff for a decade or so.
No I think I meant something different from that. If you did lock down a system like that, it would be a lot less useful to the user. The flexibility of general purpose computers is itself a desirable trait. We have embedded systems for single purpose devices, and those are generally designed and built to much more rigid specs. Having such systems doesn't make me want to give up my general purpose computer, or stop writing software for it.
I get tired of hearing about how much better it would be if software developers acted like 'real' engineers. Try designing the Golden Gate bridge, but rather than connecting two fixed points of land, it instead connected to the roofs of two different skyscrapers (other apps). And instead of sinking footings into the bedrock, you put the footings into floating barges (the OS). And it just so happens that the barges support hundreds of other bridges besides yours, sometimes all at once. Oh, and the skyscrapers and barges can be swapped out for newer versions at any time after you've finished building your bridge. Make it stand up for 10 minutes and i'll be impressed, never mind 10 years.
The bridge metaphor is not a good metaphor for software development.
Yes. All those people raving about how lovely OS X is the past 5 years? Somehow they all missed the ugly font rendering. Then someone posted a screenshot on Slashdot of someone's 2 month effort to port some open source project to OS X and arose (644256) finally pulled the wool away from everyone's eyes.
You're a hawkeye arose (644256), nothing gets passed you!
Unlike on Windows (or Classic Mac OS) when you insert a CD or DVD on OS X you have a limited set of actions that can be performed automatically. It's completely under the user's control ("CDs & DVDs" control panel), and basically amounts to running a pre-existing application on the hard drive (not the CD).
So I would argue that market share is irrelevant, the Macrovision style copy protection (running malware off the disc) isn't feasible on a Mac.
Also the installed base is probably higher than 3%, and if you limit it to home users it's probably a good deal higher than that.
You're free to have whatever opinion you like, but keep in mind that the exact same 'logic' was used to justify anti-miscegenation laws, and the equal protection clause was the basis for ruling them unconstitutional as well.
Quoting the ruling in Loving v. Virginia:
"Marriage is one of the 'basic civil rights of man,' fundamental to our very existence and survival.... To deny this fundamental freedom on so unsupportable a basis as the racial classifications embodied in these statutes, classifications so directly subversive of the principle of equality at the heart of the Fourteenth Amendment, is surely to deprive all the State's citizens of liberty without due process of law."
Substitute 'sexual orientation' for 'racial classification' and it's basically the same thing.
I'd like to see the standardization move up the stack from the common set of command line tools & libraries into the GUI. OS X has Cocoa, Windows has Win32, Linux has ???.Settle on a standard package format, standard GUI APIs, standard desktop environment, etc.
It sucks not being able to write a GUI app and know that it will run on most/all linux installations with no downloading of extra dependencies.
The distros could still differentiate based on add-on apps, much like the various WinTel vendors differentiate based on what software is bundled.
Other comments have pointed out that these 'styled' controls break consistency.
They have other problems: - don't work with keyboard input - accept clicks anywhere in their region rather than just the control itself - a click-drag should leave the control in it's current state and perform no other action. On my browser (safari) a click-drag initiates an image drag - don't work properly with the accessibility features of the OS (for vision impaired: voice over, sticky mouse, etc.)
Application developers - whether web or desktop - should always use native controls or you *will* be disrupting the user experience of at least some of your users (the accessibility problem is a biggy).
The lone hitch: Wi-Fi is useless without a hotspot.
Um, have they never heard of ad hoc networks? Plug in the EV-DO card into your PowerBook, open System Preferences, select the Sharing control panel, click the internet tab, and select "Share you Connection from to computers using ". Same idea, different steps on Windows.
I don't think the big problem with Linux on the desktop is the amount of work or lack of interest in doing the work, I think it's just that Linux developers have no taste. 'Taste' may be the wrong word for what I'm thinking of, 'aesthetic', or 'artistry' might be better choices. Point is, the Linux desktops and distros that I've seen just look butt ugly to me. Too much mimicry of Windows and too many slapped together in the most expedient manner designs.
From what I've seen Xfce has the best aesthetic of the Linux desktops. And even that is very influenced by Mac OS X. Someone needs to start with the toolkit and work their way up with a well defined theme ('theme' in the design sense, not in the 'bitmaps-go-here' sense). This takes a single creative personality, which is kind of the antithesis of how open source is developed.
UI design, like all design, requires a certain panache that really seems to be missing from most open source. Pure opinion and not the least bit helpful, I know, but that's how I see it.
Many Windows applications depend on QuickTime (just to use an example). The product's installer checks to see if QuickTime is already installed, and if not it installs a copy (you can license the QuickTime installer for redistribution for free).
WMP could work the same way. After all, if WMP was not a Microsoft product, it would HAVE to work the same way.
If Linux was only painful for things like sound cards or other expansion cards, I could live with it. But every time I try to set up a desktop Linux environment I run into dozens of tiny snags. Getting X Windows working with a scroll wheel, getting the correct default screen resolution preference set for desktop AND login screen, getting half the crap out of the Gnome menu (or KDE menu) that I don't want there, getting a printer setup, getting decent (non fugly) fonts/font rendering, etc. etc. etc.
Linux is still aimed at two groups: enthusiasts that enjoy messing around and sysadmins that are willing to build a tweaked internal version for their company. People in other groups (the elusive Grandmothers, as well as the group that just wants a machine that imposes the least hassle possible so they can get to work) are kind of put off by Linux.
Linux makes a whole lot of sense on the server, but not on (my) desktop (or laptop).
Hm. I just tried it. I hit command-space to bring up the spotlight search, hit 's', then immediately switched back to Safari. There was no problem switching back and forth repeatedly, each time I hit command-space, the Spotlight window re-displayed itself instantly and the search for everything on my drive containing 's' continued to update.
It found about 10,000 items in less than 30 seconds. So it seems fairly speedy to me (this is on a PowerBook so certainly not the fastest cpu/disk out there).
So if the KDE people say that they want every file and directory in WebCore to start with the letter 'K', does Apple have to comply with that?
What if the KDE developers say that the preferred form is on a 160GB SCSI drive installed in a dual G5 with a 30" Cinema display attached to it, does Apple have to comply with THAT?
I think that people may be taking those words from the GPL a little too far. To me, what they mean is that if I get a binary, I need to get the source required to recreate that binary.
That seems illogical on the face of it. If it's difficult to merge from WebCore into KHTML, then it should be equally difficult to merge from KHTML into WebCore.
It might be slightly easier to do a manual inspection of KHTML because of finer grained history. I don't know, I haven't looked, but it should be easy enough to do an inspection of the changes from one release of WebCore to the next.
Is the WebCore code commented?
Also, I never got an answer to my question in the previous WebCore vs. KHTML flamewar.. Is anyone working on a backport of WebCore to Linux? Seems like an interesting project to put together a Konqueror+WebCore package.
I'm happy about this. My hope is that the winner of Browser Wars 2 won't actually be a browser, it will hopefully be standards. When there are a plethora of appealing standards compliant browsers to choose from, site designers will be forced to stick to standards.
I'm not going to get into the politics of Safari vs. KHTML. It matters (to me) less how standards compliance was achieved than that it was.
I'll switch my OS X machines to Firefox as the default browser when it gets a little less sluggish. Hopefully the 1.1 release will make headway in that department.
Slashdot Mirror
Not as developer friendly? You've got to be kidding, being developer friendly was NeXTStep's raison d'etre. If you think that you can't do what .Net does on a Mac, you haven't seen WebObjects, which has been doing the web based app server stuff for a decade or so.
No I think I meant something different from that. If you did lock down a system like that, it would be a lot less useful to the user. The flexibility of general purpose computers is itself a desirable trait. We have embedded systems for single purpose devices, and those are generally designed and built to much more rigid specs. Having such systems doesn't make me want to give up my general purpose computer, or stop writing software for it.
I get tired of hearing about how much better it would be if software developers acted like 'real' engineers. Try designing the Golden Gate bridge, but rather than connecting two fixed points of land, it instead connected to the roofs of two different skyscrapers (other apps). And instead of sinking footings into the bedrock, you put the footings into floating barges (the OS). And it just so happens that the barges support hundreds of other bridges besides yours, sometimes all at once. Oh, and the skyscrapers and barges can be swapped out for newer versions at any time after you've finished building your bridge. Make it stand up for 10 minutes and i'll be impressed, never mind 10 years.
The bridge metaphor is not a good metaphor for software development.
Yes. All those people raving about how lovely OS X is the past 5 years? Somehow they all missed the ugly font rendering. Then someone posted a screenshot on Slashdot of someone's 2 month effort to port some open source project to OS X and arose (644256) finally pulled the wool away from everyone's eyes.
You're a hawkeye arose (644256), nothing gets passed you!
Unlike on Windows (or Classic Mac OS) when you insert a CD or DVD on OS X you have a limited set of actions that can be performed automatically. It's completely under the user's control ("CDs & DVDs" control panel), and basically amounts to running a pre-existing application on the hard drive (not the CD).
So I would argue that market share is irrelevant, the Macrovision style copy protection (running malware off the disc) isn't feasible on a Mac.
Also the installed base is probably higher than 3%, and if you limit it to home users it's probably a good deal higher than that.
It's just like Apple when Jobs was ousted and Scully took over. The pinheads don't know how to inspire/lead/challenge the techies.
Same thing with HP when it was no longer a place for engineers, run by engineers.
You can probably find the same pattern repeating at lots of high tech companies.
You're free to have whatever opinion you like, but keep in mind that the exact same 'logic' was used to justify anti-miscegenation laws, and the equal protection clause was the basis for ruling them unconstitutional as well.
Quoting the ruling in Loving v. Virginia:
"Marriage is one of the 'basic civil rights of man,' fundamental to our very existence and survival.... To deny this fundamental freedom on so unsupportable a basis as the racial classifications embodied in these statutes, classifications so directly subversive of the principle of equality at the heart of the Fourteenth Amendment, is surely to deprive all the State's citizens of liberty without due process of law."
Substitute 'sexual orientation' for 'racial classification' and it's basically the same thing.
"So it's easy to say "thank god I'm using public transportation", cause others pay parts of your ticket's price."
This applies to driving automobiles as well. The roads aren't free.
There's plenty of land mass above the arctic circle upon which thousands of feet thick ice shelfs rest.
Pardon me if I avoid the urge to shove my head into the sand.
http://www.microsoft.com/presspass/press/1997/Aug9 7/MSMACpr.mspx
So even if MS's patents are valid it won't lead to a lawsuit.
I'd like to see the standardization move up the stack from the common set of command line tools & libraries into the GUI. OS X has Cocoa, Windows has Win32, Linux has ???.Settle on a standard package format, standard GUI APIs, standard desktop environment, etc.
It sucks not being able to write a GUI app and know that it will run on most/all linux installations with no downloading of extra dependencies.
The distros could still differentiate based on add-on apps, much like the various WinTel vendors differentiate based on what software is bundled.
If Apple had wanted this they could've just paid whatever licensing fees are involved and implemented it in their next PPC chipset.
So I really doubt this had anything to do with the switch to x86.
Other comments have pointed out that these 'styled' controls break consistency.
They have other problems:
- don't work with keyboard input
- accept clicks anywhere in their region rather than just the control itself
- a click-drag should leave the control in it's current state and perform no other action. On my browser (safari) a click-drag initiates an image drag
- don't work properly with the accessibility features of the OS (for vision impaired: voice over, sticky mouse, etc.)
Application developers - whether web or desktop - should always use native controls or you *will* be disrupting the user experience of at least some of your users (the accessibility problem is a biggy).
The lone hitch: Wi-Fi is useless without a hotspot.
Um, have they never heard of ad hoc networks? Plug in the EV-DO card into your PowerBook, open System Preferences, select the Sharing control panel, click the internet tab, and select "Share you Connection from to computers using ". Same idea, different steps on Windows.
Yes Macs are theoretically vulnerable to viruses or spyware.
In theory there is no difference between theory and practice, but in practice there is.
So I'll just keep enjoying using a computer that doesn't have viruses or spyware, and you keep warning me that theoretically I might get one some day.
Long answer: TVs are terrible output devices (low rez, interlaced), and couches are terrible ergonomic environments for keyboarding/mousing.
You'd be better off building very cheap laptops like the Indians are doing.
I don't think the big problem with Linux on the desktop is the amount of work or lack of interest in doing the work, I think it's just that Linux developers have no taste. 'Taste' may be the wrong word for what I'm thinking of, 'aesthetic', or 'artistry' might be better choices. Point is, the Linux desktops and distros that I've seen just look butt ugly to me. Too much mimicry of Windows and too many slapped together in the most expedient manner designs.
From what I've seen Xfce has the best aesthetic of the Linux desktops. And even that is very influenced by Mac OS X. Someone needs to start with the toolkit and work their way up with a well defined theme ('theme' in the design sense, not in the 'bitmaps-go-here' sense). This takes a single creative personality, which is kind of the antithesis of how open source is developed.
UI design, like all design, requires a certain panache that really seems to be missing from most open source. Pure opinion and not the least bit helpful, I know, but that's how I see it.
Many Windows applications depend on QuickTime (just to use an example). The product's installer checks to see if QuickTime is already installed, and if not it installs a copy (you can license the QuickTime installer for redistribution for free).
WMP could work the same way. After all, if WMP was not a Microsoft product, it would HAVE to work the same way.
If Linux was only painful for things like sound cards or other expansion cards, I could live with it. But every time I try to set up a desktop Linux environment I run into dozens of tiny snags. Getting X Windows working with a scroll wheel, getting the correct default screen resolution preference set for desktop AND login screen, getting half the crap out of the Gnome menu (or KDE menu) that I don't want there, getting a printer setup, getting decent (non fugly) fonts/font rendering, etc. etc. etc.
Linux is still aimed at two groups: enthusiasts that enjoy messing around and sysadmins that are willing to build a tweaked internal version for their company. People in other groups (the elusive Grandmothers, as well as the group that just wants a machine that imposes the least hassle possible so they can get to work) are kind of put off by Linux.
Linux makes a whole lot of sense on the server, but not on (my) desktop (or laptop).
Just my $.02
Hm. I just tried it. I hit command-space to bring up the spotlight search, hit 's', then immediately switched back to Safari. There was no problem switching back and forth repeatedly, each time I hit command-space, the Spotlight window re-displayed itself instantly and the search for everything on my drive containing 's' continued to update.
It found about 10,000 items in less than 30 seconds. So it seems fairly speedy to me (this is on a PowerBook so certainly not the fastest cpu/disk out there).
My wrists are killing me after 15 years in this business. I can't imagine how much worse they're going to get over the next 15.
:-P
I need a brain implant
So if the KDE people say that they want every file and directory in WebCore to start with the letter 'K', does Apple have to comply with that?
What if the KDE developers say that the preferred form is on a 160GB SCSI drive installed in a dual G5 with a 30" Cinema display attached to it, does Apple have to comply with THAT?
I think that people may be taking those words from the GPL a little too far. To me, what they mean is that if I get a binary, I need to get the source required to recreate that binary.
That seems illogical on the face of it. If it's difficult to merge from WebCore into KHTML, then it should be equally difficult to merge from KHTML into WebCore.
It might be slightly easier to do a manual inspection of KHTML because of finer grained history. I don't know, I haven't looked, but it should be easy enough to do an inspection of the changes from one release of WebCore to the next.
Is the WebCore code commented?
Also, I never got an answer to my question in the previous WebCore vs. KHTML flamewar.. Is anyone working on a backport of WebCore to Linux? Seems like an interesting project to put together a Konqueror+WebCore package.
I'm happy about this. My hope is that the winner of Browser Wars 2 won't actually be a browser, it will hopefully be standards. When there are a plethora of appealing standards compliant browsers to choose from, site designers will be forced to stick to standards.
I'm not going to get into the politics of Safari vs. KHTML. It matters (to me) less how standards compliance was achieved than that it was.
I'll switch my OS X machines to Firefox as the default browser when it gets a little less sluggish. Hopefully the 1.1 release will make headway in that department.