I long for a better C++ but I don't think this is it.
Operator overloading is useful beyond the cases mentioned, e.g., for matrices.
Stack-allocated objects with destructor semantics are a must for convenient general resource deallocation. Garbage collection is nice but doesn't address deallocating resources other than memory.
Some kind of multiple inheritance (even if only of interfaces with delegation) is a must.
But maybe he will accept feedback and improve his proposal?
Java GUI apps _are_ slower, because Swing is basically all written in Java and it can't really take advantage of platform-specific tricks to run faster. The trade-off here is that Swing is very, very portable. For many applications this is a good trade-off, but it's not a 100% solution. I don't think Quake is going to be written in Java real soon.
On a 800 mHz machine, you will seldom notice the Swing slowdown in Windows. On Linux it's a little more noticeable. The next JDK has some performance enhancements for running on X, I think.
Re:1-button mice suck, and other thoughts on Mac
on
Mac Rants
·
· Score: 1
The problem I have with the studies is that I know so many folks who use the right mouse button routinely. But if Apple wants to market their machines to Grandma Moses, OK by me -- just don't expect anyone who actually uses the computer a lot to buy one. (And thanks for reminding me about scroll wheels. In some ways, that's more important than the right mouse button.)
Sure, I could add a multi-button mouse myself. But until they ship with one right-button support in applications will not be standard, so it would not be all that much use.
The games on the Mac (OS 9 or OS X) suck royally. Most of the games I've played in the last few years -- Everquest, Baldur's Gate (I & II), Fallout I, II & Tactics, Alpha Centauri, etc. -- were not available on the Mac. The Mac has a few of the most "general public" kind of games (a la Quake3), but no gamer is going to get near the platform until they support _most_ of the good gamer's games (for whatever genres you are interested in). Ideally, they need a few killer games that are _only_ available on the Mac.
If you don't like Dolby's behavior, organize a protest. Boycott their products. Show up in front of their offices with signs, whatever.
And don't forget to vote based on candidate's views on IP. So long as we go along with it, they will always support IP because of payola^H^H^H^H^H^Hcampaign contributions.
Bitching on slashdot, though -- that's worthless.
1-button mice suck, and other thoughts on Mac
on
Mac Rants
·
· Score: 1
I've been using Macs since 1987. I have both a Mac and a PC on my desks here at work. I owned a Mac until 1992 (at which time I bought a PC).
The one-button mouse has gotta go. It's an example of Apple stubbornly clinging to an inferior solution.
I have OS X on the desktop. It's OK (I like Unix). But unless they do something to catch up in games Macs are going to slowly become photoshop-only machines. And the market won't support that; there just aren't enough photoshop users to maintain a market.
No, we are all for making money, it pays the rent.
But money is not the only (or most) important thing. Some of us care _how_ we make our money.
If Caldera had written all the software in their distro themselves, there would be no argument. We might not respect them (a la Microsoft), but they would definitely has the right to do as they pleased.
Of course, they wrote 1% (to pick a number out of a hat), so it smells a little, but they still have the right to limit the distro as a whole (presuming they put some non-GPLed stuff into it).
But since they cannot hold GPLed software hostage, it isn't likely to help Caldera much in the long run. Discriminating customers will simply turn elsewhere.
They cannot make the free software in their distribution any less free. . . All they can do is add a little non-free software to the distribution and limit seats based on that.
And who cares? It isn't like they're adding anything so compelling that one is forced to use Caldera. The only practical effect is that folks who care about copyleft will use another distribution.
So long as we keep supporting GPL, we will be fine. (Caldera will probably go under, but that's their lookout.)
Despite Harlan's annoying use of uppercase and general volume level, I have a lot of sympathy for his case. Posting stories without authorization _is_ piracy. And most authors are poor -- it's one of most poorly paying professions in North America.
But. I am skeptical that it can be stopped. In the United States it surely can be, with enough lawyers (which most authors cannot afford). But the internet is global, and it only takes a few sites beyond the reach of local law...
The copyright laws were created in a time when copying was physical, labor-intensive, and therefore relatively easy to regulate. It's going to be much, much harder to enforce such laws in an era when copying is trivial and very non-local. I am not convinced that the current copyright laws can survive (or at least function).
But I have vacationed in Utah and could not see living there. The influence of the mormons is just too strong. An agnostic like me would be uncomfortable. So, in that sense their culture would keep me from taking a job there, absolutely.
There are sites I'd consider paying for, but I'm pretty picky. I might pay something ($5 a month? $10?) to get the New York Times. I'd pay to read lumthemad.net. Etc.
I really think that the only way pay would work well would be to pay for a network of affiliated sites. The whole micropayment thing doesn't work for me, because I end up worrying about the meter running. Consumers like macropayments.;-)
But I'm skeptical that it can work in the long run, because most users are cheapskates, and will just skate on by with free sites (or do without).
>Collections/containers/algorithms that aren't the abomination that is STL.
The STL is one of the more elegant and useful collection libraries I've seen in years. It makes the Java collections look amaturish.
One of the real weaknesses of Java currently is its lack of support for generic programming. (Another is the fact that while Swing is nice and portable, it's slow.)
What if my users need cross-platform apps? That depends. Alternatives to Java include Citrix (windows into an NT machine) and web applications. Might Java be the right solution for some hypothetical situation? Sure, but I don't think it is the common situation.
In practice, most corporations run pretty much exclusively in Windows, so the "cross platform" nature of Java is not a big selling point in that environment. (Which is the environment probably 80-90% of the software gets written for.) Shrinkwrap is another important market, but customers of shrinkwrap tend to be very particular about the UI matching that of their preferred OS.
Uh-huh. And that's why I'm not on the Java bandwagon: It leads to cross-platform but distinctly second-rate products.
Do you think the users of the software care how cross-platform it is? Or how "pure" it is? No, they just care about how well it works. And if the mouse wheels doesn't work, they won't be happy.
A cross-platform development tool makes engineering easier, but making engineering easier should not be the primary goal. Delivering software that meets the needs of users should be the primary goal.
Console games are a different bag, with a different (in my opinion, markedly inferior) interface. If you want to play PC-style games (and lots of us do), you need Windoze.
And hey, why would I bother to spend $4000 on a machine if not for games?;-) I mean, sure, it runs my compilers faster, but do I care? On a home machine?
I don't primarily program on Linux, but from what I've seen, the gui problems are 1) most of the programming has been by-geeks-for-geeks, which tends not to emphasize gui; 2) x and its window managers have traditionally been sluggish and offered poor functionality (for a desktop UI -- let's not get into the running programs in Moscow from NYC stuff), so *nix die-hardss don't expect much; 3) programming in traditional *nix toolkits wasn't particularly easy or pleasant, which cut down the incentive to do it; 4)Motif was not free.
There are probably other reasons. Better toolkits are beginning to emerge, but Windows and Macintosh are still lightyears ahead in terms of the GUI.
Slashdot Mirror
I long for a better C++ but I don't think this is it.
But maybe he will accept feedback and improve his proposal?
Java GUI apps _are_ slower, because Swing is basically all written in Java and it can't really take advantage of platform-specific tricks to run faster. The trade-off here is that Swing is very, very portable. For many applications this is a good trade-off, but it's not a 100% solution. I don't think Quake is going to be written in Java real soon. On a 800 mHz machine, you will seldom notice the Swing slowdown in Windows. On Linux it's a little more noticeable. The next JDK has some performance enhancements for running on X, I think.
The problem I have with the studies is that I know so many folks who use the right mouse button routinely. But if Apple wants to market their machines to Grandma Moses, OK by me -- just don't expect anyone who actually uses the computer a lot to buy one. (And thanks for reminding me about scroll wheels. In some ways, that's more important than the right mouse button.)
Sure, I could add a multi-button mouse myself. But until they ship with one right-button support in applications will not be standard, so it would not be all that much use.
The games on the Mac (OS 9 or OS X) suck royally. Most of the games I've played in the last few years -- Everquest, Baldur's Gate (I & II), Fallout I, II & Tactics, Alpha Centauri, etc. -- were not available on the Mac. The Mac has a few of the most "general public" kind of games (a la Quake3), but no gamer is going to get near the platform until they support _most_ of the good gamer's games (for whatever genres you are interested in). Ideally, they need a few killer games that are _only_ available on the Mac.
If you don't like Dolby's behavior, organize a protest. Boycott their products. Show up in front of their offices with signs, whatever.
And don't forget to vote based on candidate's views on IP. So long as we go along with it, they will always support IP because of payola^H^H^H^H^H^Hcampaign contributions.
Bitching on slashdot, though -- that's worthless.
I've been using Macs since 1987. I have both a Mac and a PC on my desks here at work. I owned a Mac until 1992 (at which time I bought a PC). The one-button mouse has gotta go. It's an example of Apple stubbornly clinging to an inferior solution. I have OS X on the desktop. It's OK (I like Unix). But unless they do something to catch up in games Macs are going to slowly become photoshop-only machines. And the market won't support that; there just aren't enough photoshop users to maintain a market.
No, we are all for making money, it pays the rent.
But money is not the only (or most) important thing. Some of us care _how_ we make our money.
If Caldera had written all the software in their distro themselves, there would be no argument. We might not respect them (a la Microsoft), but they would definitely has the right to do as they pleased.
Of course, they wrote 1% (to pick a number out of a hat), so it smells a little, but they still have the right to limit the distro as a whole (presuming they put some non-GPLed stuff into it).
But since they cannot hold GPLed software hostage, it isn't likely to help Caldera much in the long run. Discriminating customers will simply turn elsewhere.
They cannot make the free software in their distribution any less free. . . All they can do is add a little non-free software to the distribution and limit seats based on that.
And who cares? It isn't like they're adding anything so compelling that one is forced to use Caldera. The only practical effect is that folks who care about copyleft will use another distribution.
So long as we keep supporting GPL, we will be fine. (Caldera will probably go under, but that's their lookout.)
Static linking has a lot to recommend it.
Yes, it uses more disk space. Disk space is cheap nowadays. But it reduces the hassle factor enormously.
Smalltalk or Eiffel would be my choices; they are both better for teaching OO than either C++ or Java.
Eiffel is the cleanest and most readable, imo.
I program in Java for a living. It gets the job done, but it isn't very elegant.
Why buy it if I can't avoid the commericals? Even a VCR can do that...
Without the ability to follow threads of conversation, it has very limited value to me. The old deja was much better. :(
The thermal throttling is going to generate a huge (and well-deserved) backlash.
Can't companies just move to Mexico and ship from there? Or is the additional shipping cost too large?
In general, I think states are coming at this from the wrong direction. They should get rid of their local sales taxes, and use income taxes instead.
I hate anti-aliased fonts, they always look blurry to me. Mucho eyestrain.
Despite Harlan's annoying use of uppercase and general volume level, I have a lot of sympathy for his case. Posting stories without authorization _is_ piracy. And most authors are poor -- it's one of most poorly paying professions in North America.
But. I am skeptical that it can be stopped. In the United States it surely can be, with enough lawyers (which most authors cannot afford). But the internet is global, and it only takes a few sites beyond the reach of local law...
The copyright laws were created in a time when copying was physical, labor-intensive, and therefore relatively easy to regulate. It's going to be much, much harder to enforce such laws in an era when copying is trivial and very non-local. I am not convinced that the current copyright laws can survive (or at least function).
I don't drink so I don't care about that.
But I have vacationed in Utah and could not see living there. The influence of the mormons is just too strong. An agnostic like me would be uncomfortable. So, in that sense their culture would keep me from taking a job there, absolutely.
There are sites I'd consider paying for, but I'm pretty picky. I might pay something ($5 a month? $10?) to get the New York Times. I'd pay to read lumthemad.net. Etc.
;-)
I really think that the only way pay would work well would be to pay for a network of affiliated sites. The whole micropayment thing doesn't work for me, because I end up worrying about the meter running. Consumers like macropayments.
But I'm skeptical that it can work in the long run, because most users are cheapskates, and will just skate on by with free sites (or do without).
>Collections/containers/algorithms that aren't the abomination that is STL.
The STL is one of the more elegant and useful collection libraries I've seen in years. It makes the Java collections look amaturish.
One of the real weaknesses of Java currently is its lack of support for generic programming. (Another is the fact that while Swing is nice and portable, it's slow.)
Aspect-oriented programming:
http://www.parc.xerox.com/csl/projects/aop/
Also see AspectJ:
http://aspectj.org/
Last I heard Berlin wanted to make everything a CORBA service, and didn't sound to lickety-split fast itself.
For games to make it on the linux desktop, they need something FAST.
What if my users need cross-platform apps? That depends. Alternatives to Java include Citrix (windows into an NT machine) and web applications. Might Java be the right solution for some hypothetical situation? Sure, but I don't think it is the common situation.
In practice, most corporations run pretty much exclusively in Windows, so the "cross platform" nature of Java is not a big selling point in that environment. (Which is the environment probably 80-90% of the software gets written for.) Shrinkwrap is another important market, but customers of shrinkwrap tend to be very particular about the UI matching that of their preferred OS.
Uh-huh. And that's why I'm not on the Java bandwagon: It leads to cross-platform but distinctly second-rate products.
Do you think the users of the software care how cross-platform it is? Or how "pure" it is? No, they just care about how well it works. And if the mouse wheels doesn't work, they won't be happy.
A cross-platform development tool makes engineering easier, but making engineering easier should not be the primary goal. Delivering software that meets the needs of users should be the primary goal.
Console games are a different bag, with a different (in my opinion, markedly inferior) interface. If you want to play PC-style games (and lots of us do), you need Windoze.
;-) I mean, sure, it runs my compilers faster, but do I care? On a home machine?
And hey, why would I bother to spend $4000 on a machine if not for games?
The article linked to didn't really do much to explain CDSA... I'm curious. What does it do?
I don't primarily program on Linux, but from what I've seen, the gui problems are
1) most of the programming has been by-geeks-for-geeks, which tends not to emphasize gui;
2) x and its window managers have traditionally been sluggish and offered poor functionality (for a desktop UI -- let's not get into the running programs in Moscow from NYC stuff), so *nix die-hardss don't expect much;
3) programming in traditional *nix toolkits wasn't particularly easy or pleasant, which cut down the incentive to do it;
4)Motif was not free.
There are probably other reasons. Better toolkits are beginning to emerge, but Windows and Macintosh are still lightyears ahead in terms of the GUI.