First I must concede defeat on the etymology of Cocoa. It seems that my memory of the NeXTStep days are a bit hazy... maybe I should have held onto that cube longer. When Apple coined the term Cocoa, much of the open/nextstep documentation was rereleased virtually unchanged except in title. I suppose it would have confused people to keep developing under the same name which was used interchangeably for the OS and high level API. I'll go eat my crow now, sorry folks.
Cocoa was an interesting choice in names. Its ironic (or perhaps deliberate) that the name is a play off of Java's name. Both provide hardware abstraction and platform agnosticism via high-level APIs. As programmer pointed out, framework is a more precise classification for Cocoa.
Java's virtual machine is an interesting and prominent anomaly in the world of 'virtual machines'. While Java native hardware was initially envisioned, the world only adopted Java's API and the binary libraries needed to make applications run. The role of byte code has also evolved as Java's role is nudge by Sun in new directions. From a programmer's and user's perspective, there is little functional differences in what must be done to get code running on multiple platforms. Java is the primary reason why the scope of the term 'virtual machine' has become ambiguous. Anyone care to categorize .net?
Doesn't it sucks when improper or laymen use of a term can obfuscate a word's original technical meaning? In this case, the most prominent use of the the term happens to be a slight misuse. I'll even admit that hardware-abstracting APIs (frameworks) might be an acceptable peg on which to hang the 'virtual machine' name tag. It is quite reasonable to dispute this last point. Yet, its important to note that the term's connotation is changing, as evidenced by conversations with recent CS BS recipients.
The question is, how long do you fight to preserve a word against an improper connotation becoming definition number-one in the dictionary?
I guess it depends on whether you are British or not?
.NET's "common runtime environment" is a virtual machine as well, and it is designed to support multiple languages (C#, VB, MC++, etc). I don't know much about it, although I've heard that it is a better designed VM than Java's. Although I'm loathe to admit it, C# is actually quite a nice language from what I'm told, and MC++ (Managed C++) is a nice modification of C++ which interacts nicely with natively compiled C++. From what I know it looks like a superiour solution to Objective-C, although Cocoa vs .NET is roughly equal. Again, I'm basing this on 2nd hand opinion not my own personal experience. I hope somebody is looking at whether it makes sense for Apple or Microsoft to provide a .NET runtime for MacOS X. If they could support the Aqua look-and-feel on top of CRE .NET apps this could be a real boon for Apple.
Cocoa was an interesting choice in names. Its ironic (or perhaps deliberate) that the name is a play off of Java's name. Both provide hardware abstraction and platform agnosticism via high-level APIs. As programmer pointed out, framework is a more precise classification for Cocoa.
Apple basically added Java support to Yellow Box, which up until that point was Obj.C-only, so they renamed it Cocoa to reflex the Java "flavour".
.NET's "common runtime environment" is a virtual machine as well, and it is designed to support multiple languages (C#, VB, MC++, etc). I don't know much about it, although I've heard that it is a better designed VM than Java's. Although I'm loathe to admit it, C# is actually quite a nice language from what I'm told, and MC++ (Managed C++) is a nice modification of C++ which interacts nicely with natively compiled C++. From what I know it looks like a superiour solution to Objective-C, although Cocoa vs .NET is roughly equal. Again, I'm basing this on 2nd hand opinion not my own personal experience. I hope somebody is looking at whether it makes sense for Apple or Microsoft to provide a .NET runtime for MacOS X. If they could support the Aqua look-and-feel on top of CRE .NET apps this could be a real boon for Apple.
It looks like Microsoft has also adopted the 'Framework' moniker. Although, they?ve opted to use the name ?common runtime environment? rather than ?virtual machine?. How about I just let this topic alone with the final note that I dislike both terms but am willing to use which ever one prevails. (Virtual Machine, Framework, API, runtime environment, Common Language Infrastructure, Platform Abstraction Layer ?)
An implementation of the runtime for the Common Language Infrastructure (ECMA-335) that builds and runs on Windows XP, the FreeBSD operating system, and Mac OS X 10.2.
The Platform Adaptation Layer (PAL) used to port the Shared Source CLI from Windows XP to FreeBSD and Mac OS X.
I too have yet to use .NET although I?d be willing to bet that just about any implementation would be have a cleaner API than java. As much as I love to hate MS, an industry wide trend towards cross-platform compatibility can only be good for our platform.
The Shared Source CLI archive contains the following technologies in source code form: An implementation of the runtime for the Common Language Infrastructure (ECMA-335) that builds and runs on Windows XP, the FreeBSD operating system, and Mac OS X 10.2. The Platform Adaptation Layer (PAL) used to port the Shared Source CLI from Windows XP to FreeBSD and Mac OS X. I too have yet to use .NET although I?d be willing to bet that just about any implementation would be have a cleaner API than java. As much as I love to hate MS, an industry wide trend towards cross-platform compatibility can only be good for our platform.
Thanks for the link -- the looks like good news if it operates as claimed. The main complaint I've heard about CLI is that it uses a ridiculous amount of memory just to get running (~256 MB). And why did they call it CLI when the traditional meaning of that acronym is Command Line Interface? Ugh...
It runs on the Common Language Infrastructure, restricts some language features, adds reflection / introspection, garbage collection, access to their new class frameworks, and a few other of CLI's services.
Wouldn't Motorola be slightly peeved? Without knowing the agreements between IBM, Motorola and Apple concerning PowerPC development this is almost impossible to determine the veracity of this rumor.
Wouldn't Motorola be slightly peeved? Without knowing the agreements between IBM, Motorola and Apple concerning PowerPC development this is almost impossible to determine the veracity of this rumor.
Considering all the hints from folks we know to be under NDA's on this board, it probably makes sense ...
To re-cap.
(this has already been said many times elsewhere)
1 - Moto for years had a great design team but dirty fabs.
2 - Staying competitive in the hi-end fab market is getting to be more and more impossible for Moto, it's just too expensive:
THUS:
3 - Work out a deal with another chip maker that has good fabs, [AMD, Apple's potential "wonderful Ally" as Moki once said] so your kick butt designs can get made on a half decent fab and they can then see the light of day.
Who knows if this is still going on however, maybe the G5 from Moto is a complete waste of time since the 970 is getting so much press (IBM certainly doesn't need AMD's fabs) ...
Anywho ... that's the recap.
Hello 'sckimo, hope all is going well at the chip company
The first point in no way correctly represents AMD. AMD's yields and process technology rival anyone in the industry for advanced CMOS fabrication.
It's true they don't have dirty fabs (I was being lazy and trying not to type) but it's also true they have been overly variable with yields over the past year and lets not mention SOI.
Point is AMD was a frontrunner before they slipped. AMD is better than most (throwing them in the same bin as Motorola's fabs was unfair) but they aren't a front runner currently. Maybe they are on the verge of a jump back into the running but they are still yet to actually do it.
Comments
Cocoa was an interesting choice in names. Its ironic (or perhaps deliberate) that the name is a play off of Java's name. Both provide hardware abstraction and platform agnosticism via high-level APIs. As programmer pointed out, framework is a more precise classification for Cocoa.
Java's virtual machine is an interesting and prominent anomaly in the world of 'virtual machines'. While Java native hardware was initially envisioned, the world only adopted Java's API and the binary libraries needed to make applications run. The role of byte code has also evolved as Java's role is nudge by Sun in new directions. From a programmer's and user's perspective, there is little functional differences in what must be done to get code running on multiple platforms. Java is the primary reason why the scope of the term 'virtual machine' has become ambiguous. Anyone care to categorize .net?
Doesn't it sucks when improper or laymen use of a term can obfuscate a word's original technical meaning? In this case, the most prominent use of the the term happens to be a slight misuse. I'll even admit that hardware-abstracting APIs (frameworks) might be an acceptable peg on which to hang the 'virtual machine' name tag. It is quite reasonable to dispute this last point. Yet, its important to note that the term's connotation is changing, as evidenced by conversations with recent CS BS recipients.
The question is, how long do you fight to preserve a word against an improper connotation becoming definition number-one in the dictionary?
I guess it depends on whether you are British or not?
Originally posted by dfiler
Anyone care to categorize .net?
.NET's "common runtime environment" is a virtual machine as well, and it is designed to support multiple languages (C#, VB, MC++, etc). I don't know much about it, although I've heard that it is a better designed VM than Java's. Although I'm loathe to admit it, C# is actually quite a nice language from what I'm told, and MC++ (Managed C++) is a nice modification of C++ which interacts nicely with natively compiled C++. From what I know it looks like a superiour solution to Objective-C, although Cocoa vs .NET is roughly equal. Again, I'm basing this on 2nd hand opinion not my own personal experience. I hope somebody is looking at whether it makes sense for Apple or Microsoft to provide a .NET runtime for MacOS X. If they could support the Aqua look-and-feel on top of CRE .NET apps this could be a real boon for Apple.
Originally posted by dfiler
Cocoa was an interesting choice in names. Its ironic (or perhaps deliberate) that the name is a play off of Java's name. Both provide hardware abstraction and platform agnosticism via high-level APIs. As programmer pointed out, framework is a more precise classification for Cocoa.
Apple basically added Java support to Yellow Box, which up until that point was Obj.C-only, so they renamed it Cocoa to reflex the Java "flavour".
Originally posted by Programmer
.NET's "common runtime environment" is a virtual machine as well, and it is designed to support multiple languages (C#, VB, MC++, etc). I don't know much about it, although I've heard that it is a better designed VM than Java's. Although I'm loathe to admit it, C# is actually quite a nice language from what I'm told, and MC++ (Managed C++) is a nice modification of C++ which interacts nicely with natively compiled C++. From what I know it looks like a superiour solution to Objective-C, although Cocoa vs .NET is roughly equal. Again, I'm basing this on 2nd hand opinion not my own personal experience. I hope somebody is looking at whether it makes sense for Apple or Microsoft to provide a .NET runtime for MacOS X. If they could support the Aqua look-and-feel on top of CRE .NET apps this could be a real boon for Apple.
It looks like Microsoft has also adopted the 'Framework' moniker. Although, they?ve opted to use the name ?common runtime environment? rather than ?virtual machine?. How about I just let this topic alone with the final note that I dislike both terms but am willing to use which ever one prevails.
The Shared Source CLI archive contains the following technologies in source code form:
- An implementation of the runtime for the Common Language Infrastructure (ECMA-335) that builds and runs on Windows XP, the FreeBSD operating system, and Mac OS X 10.2.
- The Platform Adaptation Layer (PAL) used to port the Shared Source CLI from Windows XP to FreeBSD and Mac OS X.
I too have yet to use .NET although I?d be willing to bet that just about any implementation would be have a cleaner API than java. As much as I love to hate MS, an industry wide trend towards cross-platform compatibility can only be good for our platform.Originally posted by dfiler
The Shared Source CLI archive contains the following technologies in source code form: An implementation of the runtime for the Common Language Infrastructure (ECMA-335) that builds and runs on Windows XP, the FreeBSD operating system, and Mac OS X 10.2.
The Platform Adaptation Layer (PAL) used to port the Shared Source CLI from Windows XP to FreeBSD and Mac OS X.
I too have yet to use .NET although I?d be willing to bet that just about any implementation would be have a cleaner API than java. As much as I love to hate MS, an industry wide trend towards cross-platform compatibility can only be good for our platform.
Thanks for the link -- the looks like good news if it operates as claimed. The main complaint I've heard about CLI is that it uses a ridiculous amount of memory just to get running (~256 MB). And why did they call it CLI when the traditional meaning of that acronym is Command Line Interface? Ugh...
Originally posted by Programmer
and MC++ (Managed C++) is a nice modification of C++ which interacts nicely with natively compiled C++.
"Managed C++"?
Sounds like an oxymoron.
What's the difference?
Originally posted by OverToasty
"Managed C++"?
Sounds like an oxymoron.
What's the difference?
It runs on the Common Language Infrastructure, restricts some language features, adds reflection / introspection, garbage collection, access to their new class frameworks, and a few other of CLI's services.
Topic: AMD to make PPCs?!
Hmmm, that would be interesting...
Originally posted by Eskimo
Hmmm, that would be interesting...
hmmm... wouldn't it though...
Originally posted by Eskimo
Hmmm, that would be interesting...
Wouldn't Motorola be slightly peeved? Without knowing the agreements between IBM, Motorola and Apple concerning PowerPC development this is almost impossible to determine the veracity of this rumor.
Originally posted by rickag
Wouldn't Motorola be slightly peeved? Without knowing the agreements between IBM, Motorola and Apple concerning PowerPC development this is almost impossible to determine the veracity of this rumor.
Considering all the hints from folks we know to be under NDA's on this board, it probably makes sense ...
To re-cap.
(this has already been said many times elsewhere)
1 - Moto for years had a great design team but dirty fabs.
2 - Staying competitive in the hi-end fab market is getting to be more and more impossible for Moto, it's just too expensive:
THUS:
3 - Work out a deal with another chip maker that has good fabs, [AMD, Apple's potential "wonderful Ally" as Moki once said] so your kick butt designs can get made on a half decent fab and they can then see the light of day.
Who knows if this is still going on however, maybe the G5 from Moto is a complete waste of time since the 970 is getting so much press (IBM certainly doesn't need AMD's fabs) ...
Anywho ... that's the recap.
Hello 'sckimo, hope all is going well at the chip company
If true, I would hope it would not involve the MPC7457 with its' MPX 167MHz FSB.
Originally posted by OverToasty
1 - Moto for years had a great design team but dirty fabs.
2 - Staying competitive in the hi-end fab market is getting to be more and more impossible for Moto, it's just too expensive:
Those same two points are true for AMD as well as Moto. AMD is progressively doing less and less of its own fabrication let alone anybody elses.
Originally posted by Telomar
Those same two points are true for AMD as well as Moto. AMD is progressively doing less and less of its own fabrication let alone anybody elses.
The first point in no way correctly represents AMD. AMD's yields and process technology rival anyone in the industry for advanced CMOS fabrication.
Originally posted by Eskimo
Hmmm, that would be interesting...
Is that a moki imitation I hear? And wasn't Eskimo an AMD insider?
Originally posted by Eskimo
The first point in no way correctly represents AMD. AMD's yields and process technology rival anyone in the industry for advanced CMOS fabrication.
Interesting article today suggests that IBM will be making chips for AMD
http://www.timesunion.com/AspStories/story.asp?storyID=118236&category=BUSINESS&BCCode= HOME&newsdate=3/23/2003
Why not skip the middleman and have IBM make the chips directly for Apple?8)
Originally posted by Eskimo
The first point in no way correctly represents AMD. AMD's yields and process technology rival anyone in the industry for advanced CMOS fabrication.
It's true they don't have dirty fabs (I was being lazy and trying not to type) but it's also true they have been overly variable with yields over the past year and lets not mention SOI.
Point is AMD was a frontrunner before they slipped. AMD is better than most (throwing them in the same bin as Motorola's fabs was unfair) but they aren't a front runner currently. Maybe they are on the verge of a jump back into the running but they are still yet to actually do it.