Access to full-text is by subscription or pay-per-view only, and registration is required. If you do not have an appropriate subscription, after you register, you may purchase one online or purchase a copy of the article. Single articles may be purchased at the following prices:
ACM Members:
Professionals, $5 per article
Students, $3 per article
> VisualStudio.NET (7) is buggy like hell.. languages aren`t even backward compatible any more. Sure, you can do all sorts of nice thigns with it, but you can also throw away every bit of old and perfectly healthy code..
Not only this, but you can encrypt JS code, MS has a tool that does it (for free), I'm not sure how well encrypted code can be moved around, its main use is in distribuing ASP pages whose source you want to keep.
IE, server's pages are encrypted, even if you do get the source, you can't get any info out of it.
And JS is by no mean cross-platform, it's not even cross-browser.
> If.NET handles the 30 odd languages they claim to support, with easy extendability for more.
Yes. Already it run over half a dozen languages, with more to come.
> If they make.NET a standard, allow others to freely innovate on it (with none of the licensing restrictions Sun likes to impose to keep companies like IBM in line).
They did.
I'm not sure about the licensing, though.
> If they can make sure.NET really is vendor neutral, so shipping the.NET foundation is not like shipping the JVM which is little more than a commercial Sun product.
Check Corel & MS deal.
Corel will port.NET to linux.
X86 assembly?
You are out of luck, X86 is dying.
It's dying for a year & a half already.
You can't just keep extending it on & on.
Welcome to IA-64 world, were even Intel isn't sure what assembly will look like.
I've met a Basic programmer some (long) time ago that hated C, but was forced to learn it.
He would be interested to know that C is so much alike Basic, after all, it can't be much different if there is a tool to convert basic to C.
And I've a web developer here that got frustrated by ASP on *nix, so he is learning PHP, he would also be amazed to learn that there can't be much different between ASP & PHP because there is a tool to covert ASP to PHP.
a) No, but if it had been C/C++/Pascal & had the time to train you, I would. You already has the right way of thinking, it's just to learn a different language, not a whole new way of thinking.
Perl is already there, thanks to ActiveStep.
I don't know about Python.
The main difference between Java &.NET is that the CLR was written to be used by multiply languages.
Yes, there are some restrictions on the way you write a program you want to compile to CLR, but try to compare it to writing a JVM program using non-Java language. You've to write just like you do in Java in order for it to work.
> I made a subtle tweak to the algorithm in the VB version, and Lo! it outran the C++ version 12 to 1.
Do the same tweak in c++, how fast is it then?
On identical algoritms, Java will be much slower than C/C++/any compiled language.
It's inherited from the nature of it.
Since all the registry calling are done via the APIs, you would need to create a set of APIs for Linux to handle this.
Since *nix seem to despite the very idea of the registry, (binary format, unified, obscure), another possible way is to do something like this:
create a directory which will be used as a storage room for text files.
The registry entries will be kept in a.reg format (to keep it compatible), with each application having it's unique file.
The main problems is speed, registry calling is extremely fast because of the nature of the registry, you'll need to phrase the text file on every call, which will be slow.
Some one else would've to calculate how much slower it will be.
Another problem is that you'll need to copy a lot of registry entires from windows, and make them avilable for programs.
I think that the best way to implement it would be to create one default file (just export a freshly installed registry to a.reg file), and when the program change a key, it will go to its unique file, and when it call it the next time, it will get its own key, not the one in the default registry.
But this has performance issues all over it.
Slashdot Mirror
>http://www.acm.org/pubs/citations/journals/cacm/1 9 99-42-10/p109-prechelt/
Respond in:
Access to full-text is by subscription or pay-per-view only, and registration is required. If you do not have an appropriate subscription, after you register, you may purchase one online or purchase a copy of the article. Single articles may be purchased at the following prices:
ACM Members:
Professionals, $5 per article
Students, $3 per article
Nonmembers:
$10 per article
Can't read it, need to pay.
Can you sumerize?
Windows 2K has 99.999% uptime [source www.microsoft.com]
> VisualStudio.NET (7) is buggy like hell.. languages aren`t even backward compatible any more. Sure, you can do all sorts of nice thigns with it, but you can also throw away every bit of old and perfectly healthy code..
Or UPward
Not only this, but you can encrypt JS code, MS has a tool that does it (for free), I'm not sure how well encrypted code can be moved around, its main use is in distribuing ASP pages whose source you want to keep.
IE, server's pages are encrypted, even if you do get the source, you can't get any info out of it.
And JS is by no mean cross-platform, it's not even cross-browser.
Have you ever tried porting a non trivial C (even ANSI) from one OS to another?
Thatn leaves out half of the Java developers, and the linux world is small enough to be statistically meaningless for MS strategy, you know.
What does the word "Beta" mean to you?
If .NET approach is write once, compile everywhere, I think that this is preferable to Java approach of write once, run everywhere.
> If .NET handles the 30 odd languages they claim to support, with easy extendability for more.
.NET a standard, allow others to freely innovate on it (with none of the licensing restrictions Sun likes to impose to keep companies like IBM in line).
.NET really is vendor neutral, so shipping the .NET foundation is not like shipping the JVM which is little more than a commercial Sun product.
.NET to linux.
Yes. Already it run over half a dozen languages, with more to come.
> If they make
They did.
I'm not sure about the licensing, though.
> If they can make sure
Check Corel & MS deal.
Corel will port
X86 assembly?
You are out of luck, X86 is dying.
It's dying for a year & a half already.
You can't just keep extending it on & on.
Welcome to IA-64 world, were even Intel isn't sure what assembly will look like.
It already did.
I've met a Basic programmer some (long) time ago that hated C, but was forced to learn it.
He would be interested to know that C is so much alike Basic, after all, it can't be much different if there is a tool to convert basic to C.
And I've a web developer here that got frustrated by ASP on *nix, so he is learning PHP, he would also be amazed to learn that there can't be much different between ASP & PHP because there is a tool to covert ASP to PHP.
a) No, but if it had been C/C++/Pascal & had the time to train you, I would. You already has the right way of thinking, it's just to learn a different language, not a whole new way of thinking.
Perl is already there, thanks to ActiveStep.
.NET is that the CLR was written to be used by multiply languages.
I don't know about Python.
The main difference between Java &
Yes, there are some restrictions on the way you write a program you want to compile to CLR, but try to compare it to writing a JVM program using non-Java language. You've to write just like you do in Java in order for it to work.
> I made a subtle tweak to the algorithm in the VB version, and Lo! it outran the C++ version 12 to 1.
Do the same tweak in c++, how fast is it then?
On identical algoritms, Java will be much slower than C/C++/any compiled language.
It's inherited from the nature of it.
Since all the registry calling are done via the APIs, you would need to create a set of APIs for Linux to handle this. Since *nix seem to despite the very idea of the registry, (binary format, unified, obscure), another possible way is to do something like this: create a directory which will be used as a storage room for text files. The registry entries will be kept in a .reg format (to keep it compatible), with each application having it's unique file.
The main problems is speed, registry calling is extremely fast because of the nature of the registry, you'll need to phrase the text file on every call, which will be slow.
Some one else would've to calculate how much slower it will be.
Another problem is that you'll need to copy a lot of registry entires from windows, and make them avilable for programs.
I think that the best way to implement it would be to create one default file (just export a freshly installed registry to a .reg file), and when the program change a key, it will go to its unique file, and when it call it the next time, it will get its own key, not the one in the default registry.
But this has performance issues all over it.