At some point, yes, someone looks at code in an IDE. Of course it is accompanied by unit and functional tests and the like but I fail to see how that factors in. Or are you telling me your reviewers read code in a text editor !?!
And seriously, tell me, how readable is this:
a = b + c;
Just looking at that line of code you have no clue whether I'm adding ints, doubles, floats, imaginary numbers, vectors, matrices, built-in types or another custom type for which I overloaded operators. You have to look up at the declaration of a, b, c. So what does it matter if its a built-in type or my type? The code looks the same, and a hell of a lot cleaner than
a = b.add( c);
(and, might I add, this blows up when you start doing things like reference frame transformations... your lines will get three times longer in a second. Code that looked clean and pretty now is nested five, six, seven levels deep in parenthesis) Sure it might take a few more seconds of work to find out how my type interacts if you were an idiot and actually used one-letter variable names instead of a good solid naming convention, but provided the API it really isn't any worse, in my opinion, which is all it is. Anyways I think I've exhausted all I have to say about that.
this poster provides a good starting point. and this article. Basically both companies are looking to leverage more than just an in-browser animation scheme, rather another layer between the web and the desktop. (and of course the Google wants in on this too, right?) It's anyones chance to dominante.
Support gets reduced to security hotfixes during the extended support phase - no additional features are added. End of life is merely the point in time when the security hotfixes stop altogether. So no, by definition, you won't get it. Do we need to call the Waaaaaaaahmbulance?
In fact, there are more W2K PCs connecting to the Internet than OSX.
No shit. There are more (insert any Windows OS, Linux, and major embedded internet-capable OS's running on major routers here) connected to the internet than OSX. See, OSX has what, about 6.38% as of February? yeaaaaaah...
I'm just saying... that's one little thing that could have brought in a lot of business. And they could have done something like C# did with the safe{} and unsafe{} code, if they wanted to.
And I'd have to say program readability is bunk. Any modern IDE will tell you in a quarter second (mouse hover) what object types are being added/subtracted/multiplied/divided, and by procession (right->left) it isn't hard to figure out who is operating on whom.
good for you. Nothings ever good enough unless it is under your terms, precisely, is it? I suppose you don't use the Linux kernel, it is GPLv2? Horror of horrors...
... and that's one reason C# is gaining so much traction, Managed DirectX integrates so nicely, and with all that 3D math and such, operator overloading, etc... it's a good fit. I still think not including operator overloading was a dumb move.
Yes, performance is good, but if clarity of code can be improved (I'm thinking specifically things like header/source files, etc... that's one thing Java and c# both got right, one source file) then sometimes a little performance hit is OK. Better to be right and concise, than wrong and speedy:)
But software development **is** considered a service... and your view of services is narrow. Wal-mart employees, gas station atendees, car wash owners, Jiffy lube workers, etc. are all services.
Did anyone notice how M$ doesn't mention a version will be available for W3.11 **at all**? and I smell a bait & switch coming... "We'll tell you it's coming sometime next year--but wouldn't you rather have Vista?"
Here's your hit of the cluestick buddy: W2k went into extended support in 2005, and will be EOL'd in 2010. During the Extended Support phase, Microsoft continues to provide security hot fixes and paid support but no longer provides complimentary support options, design change requests, and non-security hotfixes.
Should have thought a little harder, because when your code is 98% math, operator overloading is a godsend:) Modeling and simulation is huge, C++ is still firmly planted for a reason, despite its flaws...
Seriously man, why don't you go spend some time working on your own code rather than rag on other people about their choice of names?
I code during my day job... and at night working on my Ph.D., I need these few moments over lunch and break for my sanity:)
But regarding the name, I find it ironic. The distro basically takes a much more popular one and removes the binary blobs... and then when you say it, it sounds like "nuisance", which to me is exactly what it is. Makes you think, they didn't think, or ?
C# also remembered some things from C++ that Sun forgot about, like operator overloading. Had they thought about oo, they might have clinched more of the scientific computing market. But alas... most of us still program in C++. I do like to tinker in c# though.
Try asking the Mono team. I was surprised myself, I was able to port code over, in many cases, with no changes whatsoever! They hit the nail on the head.
You can write a program following MSDN documentation to the letter and have it crash horribly; even worse, you can ignore the MSDN documentation and have your program run great.
Maybe back in the MFC/COM days but I haven't had problems since.NET, and the DirectX docs are good too IMO.
gNewSense... nuisance... exactly what it is without the binary blobs:)
don't let Microsoft get away with this.
Funny thing is, Microsoft and Novell are already releasing source, but apparently you'd rather see Adobe's. It's funny how hate works. You hate Microsoft so much you go to what is considered one of the more annoying online presences (flash), (and do I need to even talk about them as a corporation?) and stroke their egos, begging them to release their source... when the 'other guys' are already out there with source in hand (check out the Mono site and download away!)
Why? All they need to do is make a point-and-click "silverlight director" and dictate that everybody must use that to create their silverlight files.
Because they are marketing it as a programmable language in Javascript/C#/Ruby/Python... they'd be killing their own market and give room for a competitor to arise.
The Linux version is being developed by the Mono team, with Microsoft's assistance and blessing. Mono came up with the name before hand. You are smelling the Open Source Creativity Rat, that comes up with names like gNewSense (nuisance? heh) and the like...
Note that Moonlight is being developed **outside** of Microsoft, although it has the support (not just verbally, but engineering support) of Microsoft. So it can't be killed quite that easy.
Slashdot Mirror
At some point, yes, someone looks at code in an IDE. Of course it is accompanied by unit and functional tests and the like but I fail to see how that factors in. Or are you telling me your reviewers read code in a text editor !?!
... your lines will get three times longer in a second. Code that looked clean and pretty now is nested five, six, seven levels deep in parenthesis) Sure it might take a few more seconds of work to find out how my type interacts if you were an idiot and actually used one-letter variable names instead of a good solid naming convention, but provided the API it really isn't any worse, in my opinion, which is all it is. Anyways I think I've exhausted all I have to say about that.
And seriously, tell me, how readable is this:
a = b + c;
Just looking at that line of code you have no clue whether I'm adding ints, doubles, floats, imaginary numbers, vectors, matrices, built-in types or another custom type for which I overloaded operators. You have to look up at the declaration of a, b, c. So what does it matter if its a built-in type or my type? The code looks the same, and a hell of a lot cleaner than
a = b.add( c);
(and, might I add, this blows up when you start doing things like reference frame transformations
this poster provides a good starting point. and this article. Basically both companies are looking to leverage more than just an in-browser animation scheme, rather another layer between the web and the desktop. (and of course the Google wants in on this too, right?) It's anyones chance to dominante.
Support gets reduced to security hotfixes during the extended support phase - no additional features are added. End of life is merely the point in time when the security hotfixes stop altogether. So no, by definition, you won't get it. Do we need to call the Waaaaaaaahmbulance?
...
In fact, there are more W2K PCs connecting to the Internet than OSX.
No shit. There are more (insert any Windows OS, Linux, and major embedded internet-capable OS's running on major routers here) connected to the internet than OSX. See, OSX has what, about 6.38% as of February? yeaaaaaah
I'm just saying ... that's one little thing that could have brought in a lot of business. And they could have done something like C# did with the safe{} and unsafe{} code, if they wanted to.
And I'd have to say program readability is bunk. Any modern IDE will tell you in a quarter second (mouse hover) what object types are being added/subtracted/multiplied/divided, and by procession (right->left) it isn't hard to figure out who is operating on whom.
Your firefox up to date? 2.0.0.6 works great ...
good for you. Nothings ever good enough unless it is under your terms, precisely, is it? I suppose you don't use the Linux kernel, it is GPLv2? Horror of horrors...
Check out my free, Flash-based MAME front end at mame.danzbb.com
...
hardly unbiased himself
... and that's one reason C# is gaining so much traction, Managed DirectX integrates so nicely, and with all that 3D math and such, operator overloading, etc... it's a good fit. I still think not including operator overloading was a dumb move.
Yes, performance is good, but if clarity of code can be improved (I'm thinking specifically things like header/source files, etc... that's one thing Java and c# both got right, one source file) then sometimes a little performance hit is OK. Better to be right and concise, than wrong and speedy :)
can't keep up the hype forever
so here is a bunny with a pancake on its head
They want to kill off Adobe, not Linux.
But software development **is** considered a service... and your view of services is narrow. Wal-mart employees, gas station atendees, car wash owners, Jiffy lube workers, etc. are all services.
Did anyone notice how M$ doesn't mention a version will be available for W3.11 **at all**? and I smell a bait & switch coming ... "We'll tell you it's coming sometime next year--but wouldn't you rather have Vista?"
Here's your hit of the cluestick buddy: W2k went into extended support in 2005, and will be EOL'd in 2010. During the Extended Support phase, Microsoft continues to provide security hot fixes and paid support but no longer provides complimentary support options, design change requests, and non-security hotfixes.
Should have thought a little harder, because when your code is 98% math, operator overloading is a godsend :) Modeling and simulation is huge, C++ is still firmly planted for a reason, despite its flaws ...
Seriously man, why don't you go spend some time working on your own code rather than rag on other people about their choice of names?
... and at night working on my Ph.D., I need these few moments over lunch and break for my sanity :)
I code during my day job
But regarding the name, I find it ironic. The distro basically takes a much more popular one and removes the binary blobs... and then when you say it, it sounds like "nuisance", which to me is exactly what it is. Makes you think, they didn't think, or ?
C# also remembered some things from C++ that Sun forgot about, like operator overloading. Had they thought about oo, they might have clinched more of the scientific computing market. But alas... most of us still program in C++. I do like to tinker in c# though.
Try asking the Mono team. I was surprised myself, I was able to port code over, in many cases, with no changes whatsoever! They hit the nail on the head.
.NET, and the DirectX docs are good too IMO.
You can write a program following MSDN documentation to the letter and have it crash horribly; even worse, you can ignore the MSDN documentation and have your program run great.
Maybe back in the MFC/COM days but I haven't had problems since
gNewSense ... nuisance ... exactly what it is without the binary blobs :)
don't let Microsoft get away with this.
Funny thing is, Microsoft and Novell are already releasing source, but apparently you'd rather see Adobe's. It's funny how hate works. You hate Microsoft so much you go to what is considered one of the more annoying online presences (flash), (and do I need to even talk about them as a corporation?) and stroke their egos, begging them to release their source... when the 'other guys' are already out there with source in hand (check out the Mono site and download away!)
Why? All they need to do is make a point-and-click "silverlight director" and dictate that everybody must use that to create their silverlight files.
... they'd be killing their own market and give room for a competitor to arise.
Because they are marketing it as a programmable language in Javascript/C#/Ruby/Python
It actually does work on Firefox. I have it working here, FF 2.0.0.6.
It is built on XAML and Javascript, hard to argue that it isn't open ...
Not that hard to figure out, I mean, Microsoft has to release documentation on how to program in the language, if they want adoption, correct? :)
(Oh, wait, I see... your website... FSFE... I now understand your bias and will proceed to take you with a grain of salt.)
If so, why the different name?
...
The Linux version is being developed by the Mono team, with Microsoft's assistance and blessing. Mono came up with the name before hand. You are smelling the Open Source Creativity Rat, that comes up with names like gNewSense (nuisance? heh) and the like
Note that Moonlight is being developed **outside** of Microsoft, although it has the support (not just verbally, but engineering support) of Microsoft. So it can't be killed quite that easy.