Closed systems was the state of commercial Unixes. The things that Linux did effectively replace so I can see why/.ers would rejoice.
The way OpenSource won on the server was a progression:
1) Commercial OSes running commercial applications 2) Commercial OSes running primarily commercial applications with some open source 3) Commercial OSes running primarily open source applications with some commercial. 4) Open source OSes running primarily open source applications with some/no commercial 5) Open sources OSes running open source applications.
In the last decade the windows platform moved from (1) to (2). In this decade it may be moving from (2) to (3). The OpenSystem movement didn't come fast enough, and as a result the ubiquitous platform was Windows.
Well tieing together all the TV set top box platforms was the goal of Oak. A write once run anyway language that Sun developed and then saw a more general use for, rebranded and called Java. I don't see any reason that Java isn't fine for simple TV applications.
As for iPhone and Android they have enough market share / interest to demand custom applications.
I've seen the diagrams / propaganda from Microsoft. I don't believe it. That would mean essentially even deeper than ActiveX used to be, where web applications are making low level OS calls? That sounds just crazy.
I loved the features of Internet Explorer 4 with Active Desktop and Active X. But security is not that simple.
I agree it has those 2 but it is missing a 3rd flavor.
Easy scriptable applications, that allow end users to extend the functionality of their existing software and tie it together. Scriptability is the big limitation of the platform. I can get that security concerns may outweigh the benefits, but that's a big hole.
Oh I see what you mean. I didn't know there was a fee for the OSX store. I hope that one fails miserably.
The time invested in learning "the Apple way of thinking" is certainly "part of the additional cost of development" as I see it.
Yes and no. All systems have cultures. Unix/Linux users make better Linux developers than Windows users. Windows users get the whole Microsoft paradigm.
I don't see how that's relevant. A DVR made by TiVo runs Linux, but it doesn't allow applications from small application vendors at all. That's one reason why some people refer to a "GNU/Linux" environment, as a system with many GNU components can't be Tivoized.
It depends on culture. If Tivo were just a bit more open they would have tons of 3rd party applications. They keep the door barred, but that's their choice.
jbolden: XCode is free, and there is a much larger fraction of free apps on Linux than on OSX, especially excluding darwinports/fink apps.
Someone is much more likely to own a machine capable of running GNU/Linux than a machine capable of (lawfully) running Mac OS X. This means that in a sense, XCode is $649 and comes with a free computer. You can code Linux apps on the Linux machine (or the dual-booting Windows machine) that you are much more likely to already own.
I don't know if that's true. A pretty high percentage of developers use Macs. I'd argue that likely more IT guys and/or more programmers own Macs than know Linux. I'd say Apple is often about 20-30% more money. But then again iPhone is not cheapo technology either.
We also may have a different perspective. I don't consider $649 to be much for development tools. Pro software (less now than a few decades ago, due to open source) often costs many thousands.
I was a huge palm user in the day. The app store didn't exist when palm was popular. The basic model was a) buy application b) use palm sync to install
As for blackberry and applications stores... those were useful and heavily used. I'd agree that technologically that was pretty similar but Blackberry was a corporate product and applications were more expensive and more general purpose. The cheap / free applications which does little was a fundamental shift.
There are certain other factors at play here that I think are worth mentioning.
1) The dominant platform (COM windows apps) had a huge transitioning problem to.NET. Nothing else has really stepped forward as an application library. Cocoa / IOS is arguably the first major platform and the penetration is nowhere near what COM's was. So except in gaming there has been a tremendous lag in terms of desktop applications showing their advantages.
2) Corporate desktops, who purchase the vast majority of non entertainment software, are very reluctant to allow software installation. However they have been unable to control web applications. So end users are essentially bring in "rogue software" via. the web. This is very similar to how windows desktops replaced minis and mainframes late 80s to mid 90s. Corporate IT is unresponsive to customer wants and the web provides a means of bypassing their controls. This goes into "cheap deployment" but I thought it was worth focusing on.
3) There has been a major shift down in power in terms of desktops. Moving from mid priced desktops to cheap desktops to laptops. This has prevented desktop applications from upping system requirements as easily as servers have upped their requirements. This may pause.
4) The dominant desktop platform has been expensive while the web platforms have been cheap. This is starting to reverse with high monthly per user fees.
This argument has been going on for a long time. I tend to see both sides of it. Different platforms have different features and expectations. One size doesn't fit all.
Android is a Linux. XCode is free, and there is a much larger fraction of free apps on Linux than on OSX, especially excluding darwinports/fink apps. iPhone started the whole app store concept, Apple created a huge market for small application vendors.
I just don't see the $99 as much of a disincentive. If I'm willing to donate days / weeks / months of my time why would $100 even be a question?
Just mentioning here the time line line is all wrong. ActiveX controls came very early in the web, 1996 as a way of exporting COM. It was an alternative to Java applets not JavaScript. Microsoft was fine with JavaScript though their main alternative was VBScript.
Over the last 8 years there have been huge improvements in JavaScript engines. It is the faster engines that made JS as an app platform possible.
Yes you can run anything you want. Any anyone else who knows what they are doing can run anything they want. And if you want to distribute to the ignorant then and only then you need to get permission from the people protecting them. Seems very fair.
Your claim was that you couldn't run what you want. This other weaker claim that Apple controls the default distribution mechanism, which is easily bypassable is true. The claim you cannot run what you want is not.
As for needing to own a mac. In theory I imagine its possible to do this without a mac. The provisioning stuff is fully documented. Someone could write it all for Linux. In theory GNUStep might catch up enough with OSX's developer's environment to allow you not to own a Mac. They offer some weak control mechanism from PCs but they just migrated another whole class of their server class stuff down into OSX. OSX is the parent platform for iOS.
That's not the way drivers work in modern windows. The way windows works is there is a generic driver written by microsoft that runs as the driver and then there is essentially a user application and datafile which is the hardware specific driver. The hardware never interfaces directly with the virtual drivers.
So what will happen is:
a) There are Microsoft drivers b) There are user mode virtual drivers c) There are signed drivers
What won't exist are unsigned non-user mode drivers.
I don't think the film companies care about Linux, they are about DRM. And of course they would like a situation of more DRMed platforms. And Microsoft has been talking for almost a decade of trying to provide a secure home DRMed solution that still can run most Windows software.
There is nothing to stop a DRMed Linux from existing other than that the GPL would likely require the encryption code to be in hardware.
I don't agree. I don't think Microsoft views OEM's as customers at all. They view corporate IT departments and the home market as customers. OEM's they view as commodity ancillary device suppliers whom they treat badly.
Then you run Linux on a VM and learn 95% of the same thing. And that's assuming you can't load something in windows to hack the bios and sign the Linux.
This is how we ended up with the iPad, a general-purpose computing device which is locked down tight to make sure it's incapable of doing anything that Apple hasn't explicitly deemed permissible.
Will you stop. You buy a $99 developers license and you can create your own provisioning files for you and your friends and do whatever you want with an iPad. Apple is not exercising strict control. What they are doing is preventing end users from having to evaluate developers whom they don't know. And if you don't like Apple's policies, you can reset your authentication servers and have the iPad authenticate against different servers and Apple will sell you the complete software solution at below cost ($299).
Yes, the system is secure by default. But that's it.
Microsoft already virtualizes most drivers. The "unsigned" drivers won't be executing at level 1, they will be running in level 3 user space. A buffer overflow won't matter, the virtual driver will just crash.
Could be. Or it could be that the end user just needs to load a key for any kernel. A signed OS upgrade includes the keys for your hardware. An unsigned one requires the end user to know what they are doing. Fair for geeks, safe for non geeks.
I agree that Paladium originally used the signed all the way down approach however.
I believe this is why TV, Video Games, and Media (Music/Art) has been going downhill for quite some time now; the media companies don't want to risk "money" on the 20%, so will only cater to the 80%, leaving anyone that wants something "different" out!
There is vastly more diversity in TV and Music than there was a generation ago. The problem, particularly in TV has been that customers are demanding diverse programming thus increasing production costs relative to viewership.
Slashdot Mirror
Closed systems was the state of commercial Unixes. The things that Linux did effectively replace so I can see why /.ers would rejoice.
The way OpenSource won on the server was a progression:
1) Commercial OSes running commercial applications
2) Commercial OSes running primarily commercial applications with some open source
3) Commercial OSes running primarily open source applications with some commercial.
4) Open source OSes running primarily open source applications with some/no commercial
5) Open sources OSes running open source applications.
In the last decade the windows platform moved from (1) to (2). In this decade it may be moving from (2) to (3). The OpenSystem movement didn't come fast enough, and as a result the ubiquitous platform was Windows.
Well tieing together all the TV set top box platforms was the goal of Oak. A write once run anyway language that Sun developed and then saw a more general use for, rebranded and called Java. I don't see any reason that Java isn't fine for simple TV applications.
As for iPhone and Android they have enough market share / interest to demand custom applications.
I've seen the diagrams / propaganda from Microsoft. I don't believe it. That would mean essentially even deeper than ActiveX used to be, where web applications are making low level OS calls? That sounds just crazy.
I loved the features of Internet Explorer 4 with Active Desktop and Active X. But security is not that simple.
I agree it has those 2 but it is missing a 3rd flavor.
Easy scriptable applications, that allow end users to extend the functionality of their existing software and tie it together. Scriptability is the big limitation of the platform. I can get that security concerns may outweigh the benefits, but that's a big hole.
Sorry I don't know the incident you are referring to. Can you link to something so I can respond.
Oh I see what you mean. I didn't know there was a fee for the OSX store. I hope that one fails miserably.
The time invested in learning "the Apple way of thinking" is certainly "part of the additional cost of development" as I see it.
Yes and no. All systems have cultures. Unix/Linux users make better Linux developers than Windows users. Windows users get the whole Microsoft paradigm.
I don't see how that's relevant. A DVR made by TiVo runs Linux, but it doesn't allow applications from small application vendors at all. That's one reason why some people refer to a "GNU/Linux" environment, as a system with many GNU components can't be Tivoized.
It depends on culture. If Tivo were just a bit more open they would have tons of 3rd party applications. They keep the door barred, but that's their choice.
jbolden: XCode is free, and there is a much larger fraction of free apps on Linux than on OSX, especially excluding darwinports/fink apps.
Someone is much more likely to own a machine capable of running GNU/Linux than a machine capable of (lawfully) running Mac OS X. This means that in a sense, XCode is $649 and comes with a free computer. You can code Linux apps on the Linux machine (or the dual-booting Windows machine) that you are much more likely to already own.
I don't know if that's true. A pretty high percentage of developers use Macs. I'd argue that likely more IT guys and/or more programmers own Macs than know Linux. I'd say Apple is often about 20-30% more money. But then again iPhone is not cheapo technology either.
We also may have a different perspective. I don't consider $649 to be much for development tools. Pro software (less now than a few decades ago, due to open source) often costs many thousands.
How is it $99 / yr for Mac OSX? At this point there have been 2 released in 5 years at $29 each.
Besides that's not part of the additional cost of development. Apple wants people who use Macs and understand the apple way of thinking as developers.
I was a huge palm user in the day. The app store didn't exist when palm was popular. The basic model was
a) buy application
b) use palm sync to install
As for blackberry and applications stores... those were useful and heavily used. I'd agree that technologically that was pretty similar but Blackberry was a corporate product and applications were more expensive and more general purpose. The cheap / free applications which does little was a fundamental shift.
If you want to count blackberry I'm OK with that.
Excellent comment, absolutely correct. And shows the problem of just replacing flash.
Well written and argued points!
Remember that flash player replaced things like java and shockwave and people loathed those applications too. End users care about performance.
There are certain other factors at play here that I think are worth mentioning.
1) The dominant platform (COM windows apps) had a huge transitioning problem to .NET. Nothing else has really stepped forward as an application library. Cocoa / IOS is arguably the first major platform and the penetration is nowhere near what COM's was. So except in gaming there has been a tremendous lag in terms of desktop applications showing their advantages.
2) Corporate desktops, who purchase the vast majority of non entertainment software, are very reluctant to allow software installation. However they have been unable to control web applications. So end users are essentially bring in "rogue software" via. the web. This is very similar to how windows desktops replaced minis and mainframes late 80s to mid 90s. Corporate IT is unresponsive to customer wants and the web provides a means of bypassing their controls.
This goes into "cheap deployment" but I thought it was worth focusing on.
3) There has been a major shift down in power in terms of desktops. Moving from mid priced desktops to cheap desktops to laptops. This has prevented desktop applications from upping system requirements as easily as servers have upped their requirements. This may pause.
4) The dominant desktop platform has been expensive while the web platforms have been cheap. This is starting to reverse with high monthly per user fees.
This argument has been going on for a long time. I tend to see both sides of it. Different platforms have different features and expectations. One size doesn't fit all.
Android is a Linux. XCode is free, and there is a much larger fraction of free apps on Linux than on OSX, especially excluding darwinports/fink apps. iPhone started the whole app store concept, Apple created a huge market for small application vendors.
I just don't see the $99 as much of a disincentive. If I'm willing to donate days / weeks / months of my time why would $100 even be a question?
Just mentioning here the time line line is all wrong. ActiveX controls came very early in the web, 1996 as a way of exporting COM. It was an alternative to Java applets not JavaScript. Microsoft was fine with JavaScript though their main alternative was VBScript.
Over the last 8 years there have been huge improvements in JavaScript engines. It is the faster engines that made JS as an app platform possible.
$99 / yr. Not terribly expensive.
Yes you can run anything you want. Any anyone else who knows what they are doing can run anything they want. And if you want to distribute to the ignorant then and only then you need to get permission from the people protecting them. Seems very fair.
Your claim was that you couldn't run what you want. This other weaker claim that Apple controls the default distribution mechanism, which is easily bypassable is true. The claim you cannot run what you want is not.
As for needing to own a mac. In theory I imagine its possible to do this without a mac. The provisioning stuff is fully documented. Someone could write it all for Linux. In theory GNUStep might catch up enough with OSX's developer's environment to allow you not to own a Mac. They offer some weak control mechanism from PCs but they just migrated another whole class of their server class stuff down into OSX. OSX is the parent platform for iOS.
That's not the way drivers work in modern windows. The way windows works is there is a generic driver written by microsoft that runs as the driver and then there is essentially a user application and datafile which is the hardware specific driver. The hardware never interfaces directly with the virtual drivers.
So what will happen is:
a) There are Microsoft drivers
b) There are user mode virtual drivers
c) There are signed drivers
What won't exist are unsigned non-user mode drivers.
I don't think the film companies care about Linux, they are about DRM. And of course they would like a situation of more DRMed platforms. And Microsoft has been talking for almost a decade of trying to provide a secure home DRMed solution that still can run most Windows software.
There is nothing to stop a DRMed Linux from existing other than that the GPL would likely require the encryption code to be in hardware.
I don't agree. I don't think Microsoft views OEM's as customers at all. They view corporate IT departments and the home market as customers. OEM's they view as commodity ancillary device suppliers whom they treat badly.
Then you run Linux on a VM and learn 95% of the same thing. And that's assuming you can't load something in windows to hack the bios and sign the Linux.
This is how we ended up with the iPad, a general-purpose computing device which is locked down tight to make sure it's incapable of doing anything that Apple hasn't explicitly deemed permissible.
Will you stop. You buy a $99 developers license and you can create your own provisioning files for you and your friends and do whatever you want with an iPad. Apple is not exercising strict control. What they are doing is preventing end users from having to evaluate developers whom they don't know. And if you don't like Apple's policies, you can reset your authentication servers and have the iPad authenticate against different servers and Apple will sell you the complete software solution at below cost ($299).
Yes, the system is secure by default. But that's it.
Microsoft already virtualizes most drivers. The "unsigned" drivers won't be executing at level 1, they will be running in level 3 user space. A buffer overflow won't matter, the virtual driver will just crash.
Could be. Or it could be that the end user just needs to load a key for any kernel. A signed OS upgrade includes the keys for your hardware. An unsigned one requires the end user to know what they are doing. Fair for geeks, safe for non geeks.
I agree that Paladium originally used the signed all the way down approach however.
I believe this is why TV, Video Games, and Media (Music/Art) has been going downhill for quite some time now; the media companies don't want to risk "money" on the 20%, so will only cater to the 80%, leaving anyone that wants something "different" out!
There is vastly more diversity in TV and Music than there was a generation ago. The problem, particularly in TV has been that customers are demanding diverse programming thus increasing production costs relative to viewership.