I thought this might be addressed by the concept of "Podsafe" music which allows for free play over web (under Creative Commons license). However, it's not clear to me that Podsafe is always a guarantee for all types of web transmission.
If you're a sole programmer writing everything yourself, I'd be inclined to agree with you. Most work I see, however, involves significantly larger chunks than 40,000 LOC and coordination with other teams or contributors. C/C++ has no inherent unified memory management scheme, and so the patterns and responsibilities for memory management can vary widely between teams, libraries, or chunks of code. That is something that often gets overlooked when discussing Java memory management. Java GC has literally has made the coordination and use of library code very simple.
And add the fact that US auto manufactures lose out on markets like China because they balk at emission standards, or fuel efficiency requirements. Ford "focus" is definitely in the wrong direction.
I understand the step as a progression to overturning out-dated or corrupt systems. However, the problem is that there are an increasing number of laws by which government and quasi-government organizations can cause you uncomfort. Think of how the BSA is assisted by armed agents used to seize computers. And civil disobedience takes on a whole different meaning when the arrested have no rights as in the Padilla case. It's much easier to stop silly laws from being written than encourage or survive a systemic collapse (in this case the patent system).
Excellent point. Our household is one of those statistics. And, to boot, I find that VCR quality often is more reliable that rental/public DVDs. That is, we find far more digital glitches in DVDs than VCR tapes! Oh, I wish it weren't so!
Reminds me of Microbrews. All the major labels put out fuller bodied suds to take advantage of the beer trend. That's how capitalism works. Caveat Emptor.
I'm not so certain the market imploded as it never materialized. We kept thinking our G2 was wrong compared to NeXT, but a few years into the company we had to restructure and layed off half of Pages employees. That cut hurt efforts to bring the design editor to market, and essentially put Pages into maintenance mode. We did pursue the custom/vertical app space with the PRO Kit (Pages Remote Objects), which was used on Wall St.
The VC story is similar to many startups... losing control of the company when forced to get more cash. The details get murky and are hard to prove, but it appears that VC's killed the Pages effort to take the technology and fund another. That left founders very, very bitter.
There was no software to create document templates, which is why it was never released. Design models were rolled by hand. Yes, a design model editor would have been an improvement, and was certainly part of the intended path.
Pages was an ISV, not owned by NeXT or Apple. However, the codebase was purchased (Lighthouse?) and it was also being used for production templates at some Wall Street firms.
Uhm... well, I guess everyone has an opinion. Actually being there when the doors closed on the company I have a slightly different perspective. Like, maybe it had a bit more to do VC interaction on the board... and the fact that the NeXT market never materialized (Steve's reality distortion field at work in marketing). The engineering team actually had a skunkworks Windows effort underway; it had to be because it wasn't condoned.
Still, lots of the technology went on via Digital Style Inc., purchased by Netscape, then purchased by AOL. You'll find a whole lot of DOM and XML contributions came from Pages. I think that's where it's had it's biggest effect.
Maybe you hadn't noticed that IBM completely remade itself. IBM is about services, not hardware or software (both of which they sell). IBM fully backs the open source movement on several fronts, and strongly sells Linux into corporate accounts where it makes sense. IBM is willing to go to the mat for open source (ie. spend money in the courtroom).
I, for one, want options other than GPL. The Berkeley license works just fine, thanks.
There are plenty of developers that don't want to touch and taint commercial code with the GPL. Java has plenty of commercial success. Not everyone has to chant the GPL mantra, nor join a jihad, including Sun. There are room for multiple licensing models.
China will balance the counter-trend of the US forcing other nations to adopt technologies, standards, and policies that are for US benefit. Look at how the US is forcing their IP policy on the rest of the world, and how Europe and Australia have recently capitulated. We rant at the problems with the US patent system, and the DMCA, but the US uses strong-arm tactics so that countries adopt our problematic approaches. Next will be a slog of worldwide litigation. Free as in speech and bear will disappear.
In this sense, China is a breath of fresh air.
It is a very smart move. Recall that Indemnification is one of those boogey words SCO and Microsoft bantered about to strike FUD into the hearts of corporate executives. Novell just took the air out of their FUD balloons.
Agreed. Nothing like SQL variants, which you might have a hope and a prayer of converting between two DBs. Try it sometime and you'll experience the difference first hand.
Yes, Smalltalk was used on a lot of platforms. But there are also very specific Smalltalk VMs that only run in single environments. Dolphin Smalltalk is one of those... cool environment to code in, your code will only run on Windows.
The same was ALMOST true for Digitalk, but they did produce Smalltalks for different enviroments. However, they didn't completely resolve cross-platform differences by the time the went out of business. There was even an effort to meld Visual Smalltalk with ParcPlace, but it too just kinda dried up.
Smalltalk was not known for portability across vendors. ParcPlace and IBM were the most consistent across platforms. They also had the largest footprint.
I did my own productivity benchmarks between C++ and Smalltalk in 1996. I consider myself very adept at both languages. At the time I was coding C++ CORBA internals that had to function across 10 platforms (gawd, what a pain). I was also involved in a Smalltalk ORB project.
My productivy benchmark was completing foundation frameworks for a financial trading package. The time required to complete equal functionality in C++ was 10x the amount it took in Smalltalk. I found it agreed with similar claims in language productivity.
It's important to consider the context in which the program will operate. This drives the requirements a solution will need to fulfill, and in turn, may influence the choice of environment, frameworks, libraries, and language.
There are cases where the speed of delivering accurate, new functionality is paramount. In these cases, I wouldn't want to be using C++.
I challenge your logic: How many times has OPEN been important for a language? Do/did you actively lobby the ANSI committee for ANSI C? Have you participated in any other standards bodies? Do you participate in the JCP for Java?
My assertion is that most often this is a moot point. If the language is sufficiently available on multiple platforms and from multiple vendors you prevent lock-in (e.g. choose Java flavors from Sun, IBM, or BEA). More-over, some type of standards will assure 3rd party library consistency.
Recently I saw comments where someone had separated out "performance" from "responsiveness". IMHO: More of this thinking needs to pervade software consciousness.
I think it would be accurate to say that native applications are more responsive than Swing apps. On the other hand, I don't believe it's impossible to write responsive Swing apps; I've certainly seen and used some that rival native equivalents.
My experience is that it takes a very talented Swing GUI engineer to create a responsive Swing application. Unfortunately, most IT departments don't have the type of resources required to pull it off. This is why VisualBasic and PowerBuilder win for custom departmental apps. I think with a suitable GUI builder, SWT stands a chance of competing in this arena.
It would be interesting to see a closer "Apples to Apples" comparison, as you suggest. Personally, I would like to see how this stacks up against Smalltalk arbitrary precision, in varying flavors of Smalltalk (you could start with Squeak). There are Smalltalk benchmarks where functions like factorial easily beat C.
Slashdot Mirror
I thought this might be addressed by the concept of "Podsafe" music which allows for free play over web (under Creative Commons license). However, it's not clear to me that Podsafe is always a guarantee for all types of web transmission.
http://en.wikipedia.org/wiki/Podsafe
If you're a sole programmer writing everything yourself, I'd be inclined to agree with you. Most work I see, however, involves significantly larger chunks than 40,000 LOC and coordination with other teams or contributors. C/C++ has no inherent unified memory management scheme, and so the patterns and responsibilities for memory management can vary widely between teams, libraries, or chunks of code. That is something that often gets overlooked when discussing Java memory management. Java GC has literally has made the coordination and use of library code very simple.
And add the fact that US auto manufactures lose out on markets like China because they balk at emission standards, or fuel efficiency requirements. Ford "focus" is definitely in the wrong direction.
I understand the step as a progression to overturning out-dated or corrupt systems. However, the problem is that there are an increasing number of laws by which government and quasi-government organizations can cause you uncomfort. Think of how the BSA is assisted by armed agents used to seize computers. And civil disobedience takes on a whole different meaning when the arrested have no rights as in the Padilla case. It's much easier to stop silly laws from being written than encourage or survive a systemic collapse (in this case the patent system).
Excellent point. Our household is one of those statistics. And, to boot, I find that VCR quality often is more reliable that rental/public DVDs. That is, we find far more digital glitches in DVDs than VCR tapes! Oh, I wish it weren't so!
Reminds me of Microbrews. All the major labels put out fuller bodied suds to take advantage of the beer trend. That's how capitalism works. Caveat Emptor.
Not to mention that in many cases where you'd want to make that emergency call ... you can't get reception! Doh!
Well, it's history anyway.
... losing control of the company when forced to get more cash. The details get murky and are hard to prove, but it appears that VC's killed the Pages effort to take the technology and fund another. That left founders very, very bitter.
I'm not so certain the market imploded as it never materialized. We kept thinking our G2 was wrong compared to NeXT, but a few years into the company we had to restructure and layed off half of Pages employees. That cut hurt efforts to bring the design editor to market, and essentially put Pages into maintenance mode. We did pursue the custom/vertical app space with the PRO Kit (Pages Remote Objects), which was used on Wall St.
The VC story is similar to many startups
Sven
There was no software to create document templates, which is why it was never released. Design models were rolled by hand. Yes, a design model editor would have been an improvement, and was certainly part of the intended path.
Pages was an ISV, not owned by NeXT or Apple. However, the codebase was purchased (Lighthouse?) and it was also being used for production templates at some Wall Street firms.
Uhm ... well, I guess everyone has an opinion. Actually being there when the doors closed on the company I have a slightly different perspective. Like, maybe it had a bit more to do VC interaction on the board ... and the fact that the NeXT market never materialized (Steve's reality distortion field at work in marketing). The engineering team actually had a skunkworks Windows effort underway; it had to be because it wasn't condoned.
Still, lots of the technology went on via Digital Style Inc., purchased by Netscape, then purchased by AOL. You'll find a whole lot of DOM and XML contributions came from Pages. I think that's where it's had it's biggest effect.
Sven
Maybe you hadn't noticed that IBM completely remade itself. IBM is about services, not hardware or software (both of which they sell). IBM fully backs the open source movement on several fronts, and strongly sells Linux into corporate accounts where it makes sense. IBM is willing to go to the mat for open source (ie. spend money in the courtroom).
I don't see Microsoft behaving like IBM at all.
I, for one, want options other than GPL. The Berkeley license works just fine, thanks.
There are plenty of developers that don't want to touch and taint commercial code with the GPL. Java has plenty of commercial success. Not everyone has to chant the GPL mantra, nor join a jihad, including Sun. There are room for multiple licensing models.
China will balance the counter-trend of the US forcing other nations to adopt technologies, standards, and policies that are for US benefit. Look at how the US is forcing their IP policy on the rest of the world, and how Europe and Australia have recently capitulated. We rant at the problems with the US patent system, and the DMCA, but the US uses strong-arm tactics so that countries adopt our problematic approaches. Next will be a slog of worldwide litigation. Free as in speech and bear will disappear. In this sense, China is a breath of fresh air.
Hmmmm. Sounds like "if it weren't for applications, the OS would run just fine"! Now that must be one fine OS.
Gee, and I always heard that re-use is a good thing!
Also F# ... only on MS Windows.
It is a very smart move. Recall that Indemnification is one of those boogey words SCO and Microsoft bantered about to strike FUD into the hearts of corporate executives. Novell just took the air out of their FUD balloons.
No no no. Topic at hand is Dolphin Smalltalk. Try converting that to a different platform and see if you're willing to make the same statement.
Agreed. Nothing like SQL variants, which you might have a hope and a prayer of converting between two DBs. Try it sometime and you'll experience the difference first hand.
Bzzzt. I said Dolphin Smalltalk.
... cool environment to code in, your code will only run on Windows.
Yes, Smalltalk was used on a lot of platforms. But there are also very specific Smalltalk VMs that only run in single environments. Dolphin Smalltalk is one of those
The same was ALMOST true for Digitalk, but they did produce Smalltalks for different enviroments. However, they didn't completely resolve cross-platform differences by the time the went out of business. There was even an effort to meld Visual Smalltalk with ParcPlace, but it too just kinda dried up.
Smalltalk was not known for portability across vendors. ParcPlace and IBM were the most consistent across platforms. They also had the largest footprint.
ROFLMAO ... and laughing all the way to the bank!
... and Smalltalk zealots not too far behind.
I did my own productivity benchmarks between C++ and Smalltalk in 1996. I consider myself very adept at both languages. At the time I was coding C++ CORBA internals that had to function across 10 platforms (gawd, what a pain). I was also involved in a Smalltalk ORB project.
My productivy benchmark was completing foundation frameworks for a financial trading package. The time required to complete equal functionality in C++ was 10x the amount it took in Smalltalk. I found it agreed with similar claims in language productivity.
It's important to consider the context in which the program will operate. This drives the requirements a solution will need to fulfill, and in turn, may influence the choice of environment, frameworks, libraries, and language.
There are cases where the speed of delivering accurate, new functionality is paramount. In these cases, I wouldn't want to be using C++.
I challenge your logic: How many times has OPEN been important for a language? Do/did you actively lobby the ANSI committee for ANSI C? Have you participated in any other standards bodies? Do you participate in the JCP for Java?
My assertion is that most often this is a moot point. If the language is sufficiently available on multiple platforms and from multiple vendors you prevent lock-in (e.g. choose Java flavors from Sun, IBM, or BEA). More-over, some type of standards will assure 3rd party library consistency.
Recently I saw comments where someone had separated out "performance" from "responsiveness". IMHO: More of this thinking needs to pervade software consciousness.
I think it would be accurate to say that native applications are more responsive than Swing apps. On the other hand, I don't believe it's impossible to write responsive Swing apps; I've certainly seen and used some that rival native equivalents.
My experience is that it takes a very talented Swing GUI engineer to create a responsive Swing application. Unfortunately, most IT departments don't have the type of resources required to pull it off. This is why VisualBasic and PowerBuilder win for custom departmental apps. I think with a suitable GUI builder, SWT stands a chance of competing in this arena.
It would be interesting to see a closer "Apples to Apples" comparison, as you suggest. Personally, I would like to see how this stacks up against Smalltalk arbitrary precision, in varying flavors of Smalltalk (you could start with Squeak). There are Smalltalk benchmarks where functions like factorial easily beat C.