I don't see how this kind of replication of effort best serves the adoption of Linux on the desktop in the long run.
Sometimes things needs to be rewritten because the existing code base just cannot be extended in any sensible way because of its architecture. Actually, the Krita folks probably learned from the design mistake of the gimp code base to avoid to fall in the same pitfalls, so in that sense they would have reused knowledge gathered by the gimp project.
I firmly believe that iterative development is the only way to get the architecture of complicated projects right. Rewriting code from scratch is not necessarily evil, as long as you are careful to really make the most out of the knowledge acquired in writing/working with the old code.
UI does not have to be the prettiest to be effective.
No, but it has to have a good usability. They are horribly lagging behind in that area. Open the project properties or the configuration manager in visual studio 2005, for instance.
As an investor I applaud those sound business practices.
And as a (potential) customer, I despise these practices. Especially when their marketing rethoric paints them as an innovative company that is out to improve the life of its customers.
But five years from now the convergence (yea, that's an overused term) of these brands will surprise many people and MSFT will be viewed as an innovator
Convergence? It's overused indeed. How many times did we hear those "convergence, digital convergence, etc. predictions?"
I think convergence is like perpetual motion. Everyone knows what it is, but it just cannot work. What would the convergence between zune and xbox be about, exactly? And most importantly, what innovative, useful thing would it bring the customer?
Microsoft spends $6B annual on R&D. They own patents that currently aren't reasonable to move to production because of hardware and network limitations.
Microsoft is not able to innovate. They can't get an OS or a GUI framework right even though most of the concepts involved were invented decades ago, so I don't think they are able to implement the stuff their R&D department comes up with in a practical way.
Besides, Microsoft has an history of being followers, not leaders. They don't introduce new things until some competitors does it. IE was a response to Netscape. IE7 is a response to Firefox. Vista is a response to osx, Linux, kde, gnome etc. (most of the big new features like the 3d hardware accelerated rendering framework are just response to similar things done before by competitors).net is a response to Java.
I don't know if they introduced refactoring in visual 2005 for c#, but in any case it has been done before by eclipse and netbeans.
Funnily enough, last time I argued with Microsoft shills^H^H^H^H^H^Hapologists here, it ended up on a defensive tone: "me: visual c++ doesn't do refactoring. them: is there any ide that does c++ refactoring?", which is not the way of thinking I thought an innovator would have. But I can see it being Microsoft's position.
And indeed, it seems that Microsoft never takes the initiative of introducing something new that no competitor has, which is how I would define "innovating". They are reactionary.
All those patents they have is not for them to implement new stuff, it's to go after people who do. It's a way to stifle innovation from the competitor, rather than innovating themselves. I have yet to hear of a software patent that wasn't complete bullshit anyway, like "patent on how to apply [widely known solution to solve problem of type X] to [obvious problem of type X]" (like, for instance, microsoft patent application on "using xml to store word processing documents", aka "using popular and widespread file storage specification to store a certain kind of files" )
My arrogance? I dunno. Care to get me to speed about it? I'm not sure how what I said is arrogant or what my supposed arrogance might have to do with the subject at hand.
1. This bug actually apparently fucks up the vista installation. They had a lot of bugs in the past that were incredibly annoying but didn't force you to reinstall. My point is that this doesn't prove that they would have stopped the presses for something not forcing you to reinstall but still critical.
2. The fact that they actually discovered one huge bug in time to fix it before release doesn't mean that there won't be major bugs discovered after release. With their track record, their arrogance, and the way they have to force the IT industry to leave the OS business to it, they should be held to the highest standards. We shouldn't cut them any slack just because they happened to discover a critical bug just before release for once.
I also always wondered if they take advantage of people sending stuff to another gmail account by not actually duplicating attached files (with some reference counting scheme to detect files that no one use anymore)
So they have been the first to implement an obvious idea (storing numeric values using floats instead of integers to get the kind of advantages one usually get by using floating point numbers instead of integers? OMG THAT'S BRILLIANT RUSH TO TEH PATENT OFFICE NOW!!1!). Why should they be entitled them to reap benefit from other people managing to build a successful product using the same, painfully obvious idea?
There is not only the issue of being able to complain, but also the issue of being properly informed. You can't exercise your democratic rights properly if you're not properly (or at all) informed of what your government does wrong.
Not trying to ignite a political flamewar, just speaking in general.
+5 Informative on a post explaining how much HP a children would have in D&D, which is a reply to a mildly joke vaguely related to the subject at hand (which is Jack Thompson getting his ass handed to him by the justice system)?
The version I'm talking about, the one I (and most of my co-workers) dislike is the professional edition, not the free one. You know, the one that the company I work for had to spend some big bucks on (given that we are around 20 C++ coders only on the project I'm part of, with two other projects in development in house).
This is getting hilarious. First, "B... b... b... but free ides don't do better", then "B... b... b... but there is a free version of visual studio for hobbyists, so it's free and therefore shouldn't be expected to be better than free IDEs, even the non-free version".
> Are there any IDEs that offer nontrivial c++ refactorings?
Does visual even offer trivial refactorings?
And what's with using the excuse of what the concurrence does or doesn't do to justify that this expensive product doesn't do it anyway? That's what I was on about regarding mediocrity. And I should remind you that most other IDEs are free. Visual is not by any stretch, so why would they get off the hook because some free competitor doesn't do something either?
This "no one else does XXX, so why bother" mentality is not one that spawn good and innovative products. This is a mentality that spawns mediocre shit.
Well, yeah, 7 needs more CPU than 6. But they're separated by... what? 5 years? And it does far more than 6 ever did. This is to be expected.
What does the number of years separating two versions have with the cpu consumption? And what did 7 do that 6 didn't do that could justify using more CPU?
By the time you have projects large enough to start choking VS, most of your other IDEs will have choked as well, unless they don't support the inspection necessary to support things like intellisense or refactorings.
Where are the C++ refactorings? I must have missed them. And intellisense doesn't work most of the time.
Oh, and I only actually work on a subset of these projects most of the time. I guess they haven't heard of the concept of lazy updates. They seem to need some reading on the concept of multithreading, too. Eclipse does its indexing in a background task. Visual decide to do it at any random time and becomes unavailable for a few minutes when it does it (sometime after I start using it again after leaving it idle for a while for a reson or another).
I didn't say "it been going downhill since 5", I said "it's going downhill".
7 was good for a lot of things, but you could already feel the performance starting going down the drain. Nothing critical yet, but there's clearly an architectural change that wasn't that good happening from 6 to 7.
Then comes 2005. As far as C++ is concerned, it brought basically two things: improving the data breakpoint stuff so that you don't have to bypass their context thingy that never seemes to work and just write "*(long*)0xADDR", and displaying the contents of stl containers.
Scalability-wise, it sucks. Have you ever used it with a solution containing several dozen of projects and thousand of files? It might very well be the cause of most of the performance problems I experience. But then again, I'd expect an expensive, so called "top of the line" ide to be able to scale.
Don't bash Microsoft for the sake of bashing Microsoft.
Oh, sorry. I'll just keep that in mind next time it crap itself on me. I guess that all that frustration I get dealing with their crap all day long is just something I invented, too.
The VS IDE is one of the greatest.
You're clearly not picky. Seriously, one day I'll have to capture a video of one of my working session with that thing. But then again, people like you will explain me that whatever happens is somehow my fault.
As for which IDEs are better, I think about any one of them is. Even KDevelop, which has the advantage of not being gratuitously slow (I use a goddamn 3.3.something SVN version of kdevelop at home, and despite the crashes it's snappy and so enjoyable to use)
Heck, even Eclipse is faster than that turd.
As for your other remark, I'm not disliking Microsoft out of immaturity or zealousness. I've been using their products professionally (translation: most of the time I am awake is spent working with these) for more than 6 years. I used every version of their ide since visual 5, and it's going downhill.
Which IDEs have you used on a day to day basis except visual? If the answer is none, you have no basis of comparison to quantify the quality of visual.
Right at the top? Heh. Just goes to show the culture of mediocrity that symbolize the Microsoft era of computer science.
Right after this post, when I click into the visual studio window and it takes 30 seconds of doing god knows what before I can do anything (and perhaps it will throw a complete, recursive update of the solution explorer, redrawing the damn thing after reinsterting each file), I'll keep in mind that it's "right at the top as far as IDEs go".
I guess you get what you pay for. Oh wait no, you don't. This thing cost an arm and a leg.
And this is different from visual.net 2005 how? Of course, perhaps the lack of any competition in the c# ide market makes it actually the best there is. Another good reason to avoid c# altogether.
(I must admit that I only experienced the c++ part of that awful ide. On the other hand, I'd have a hard time to believe that the general sluggishness, random hang ups, gratuitous corruption of project files, abysmal useability and other awesome features are c++ only)
Ok, I understood what you wrote backwards. But still, as you point out, we know how life works on ONE planet. How can we assume that it always works like this? I think that the idea is that some rudimentary life form could exist on Mars and unable to evolve and thrive into somethign very complex because of the environment.
I think they might aswell get rid of the parametric prims in second life and just store regular meshes. People seem to make so complicated things in that game anyway that I think it would be much more efficient (with client side caching of the meshes like they do for textures, and along with some form of progressive lod)
Or perhaps a mix of primitive based stuff for simple objects like walls etc., and switch to a mesh representation for anything complex.
Well, on one hand the procedure installation is needlessly complicated. By the way, you can forego the awful gentoo live cd and use knoppix, kubuntu or whatever live cd, which allows to skip the tedious first steps of the installation manual and start directly with the partitioning. And you can use a good browser to read the doc as you go, too.
On the other hand, the installation procedure teaches you the basics of administrating your system: a lots of things aren't entirely automated or wrapped into nice GUI because you don't necessarily do them often, and you don't want to have any choice forced on you. That's the point of Gentoo, really. You forego some ease of installation to be able to decide exactly what goes in your system and what doesn't, and for the benefit to be quickly able to use bleeding edge packages.
I would care for the death of such a person if I knew about it. You don't have to be a friend or from the family of someone to care about his death.
If the human race were to be obliterated at once however, there would be no one (friend, family, stranger, or even foe) at all to worry about it as everyone would be a casualty. Of course, perhaps some aliens somewhere else in the universe could know about us, but then again, we don't know about them, so it's a moot point:)
Slashdot Mirror
I don't see how this kind of replication of effort best serves the adoption of Linux on the desktop in the long run.
Sometimes things needs to be rewritten because the existing code base just cannot be extended in any sensible way because of its architecture.
Actually, the Krita folks probably learned from the design mistake of the gimp code base to avoid to fall in the same pitfalls, so in that sense they would have reused knowledge gathered by the gimp project.
I firmly believe that iterative development is the only way to get the architecture of complicated projects right. Rewriting code from scratch is not necessarily evil, as long as you are careful to really make the most out of the knowledge acquired in writing/working with the old code.
UI does not have to be the prettiest to be effective.
No, but it has to have a good usability. They are horribly lagging behind in that area. Open the project properties or the configuration manager in visual studio 2005, for instance.
As an investor I applaud those sound business practices.
And as a (potential) customer, I despise these practices. Especially when their marketing rethoric paints them as an innovative company that is out to improve the life of its customers.
But five years from now the convergence (yea, that's an overused term) of these brands will surprise many people and MSFT will be viewed as an innovator
Convergence? It's overused indeed. How many times did we hear those "convergence, digital convergence, etc. predictions?"
I think convergence is like perpetual motion. Everyone knows what it is, but it just cannot work.
What would the convergence between zune and xbox be about, exactly? And most importantly, what innovative, useful thing would it bring the customer?
Microsoft spends $6B annual on R&D. They own patents that currently aren't reasonable to move to production because of hardware and network limitations.
.net is a response to Java.
Microsoft is not able to innovate. They can't get an OS or a GUI framework right even though most of the concepts involved were invented decades ago, so I don't think they are able to implement the stuff their R&D department comes up with in a practical way.
Besides, Microsoft has an history of being followers, not leaders. They don't introduce new things until some competitors does it.
IE was a response to Netscape.
IE7 is a response to Firefox.
Vista is a response to osx, Linux, kde, gnome etc. (most of the big new features like the 3d hardware accelerated rendering framework are just response to similar things done before by competitors)
I don't know if they introduced refactoring in visual 2005 for c#, but in any case it has been done before by eclipse and netbeans.
Funnily enough, last time I argued with Microsoft shills^H^H^H^H^H^Hapologists here, it ended up on a defensive tone: "me: visual c++ doesn't do refactoring. them: is there any ide that does c++ refactoring?", which is not the way of thinking I thought an innovator would have. But I can see it being Microsoft's position.
And indeed, it seems that Microsoft never takes the initiative of introducing something new that no competitor has, which is how I would define "innovating". They are reactionary.
All those patents they have is not for them to implement new stuff, it's to go after people who do. It's a way to stifle innovation from the competitor, rather than innovating themselves.
I have yet to hear of a software patent that wasn't complete bullshit anyway, like "patent on how to apply [widely known solution to solve problem of type X] to [obvious problem of type X]" (like, for instance, microsoft patent application on "using xml to store word processing documents", aka "using popular and widespread file storage specification to store a certain kind of files" )
My arrogance? I dunno. Care to get me to speed about it?
I'm not sure how what I said is arrogant or what my supposed arrogance might have to do with the subject at hand.
1. This bug actually apparently fucks up the vista installation.
They had a lot of bugs in the past that were incredibly annoying but didn't force you to reinstall. My point is that this doesn't prove that they would have stopped the presses for something not forcing you to reinstall but still critical.
2. The fact that they actually discovered one huge bug in time to fix it before release doesn't mean that there won't be major bugs discovered after release.
With their track record, their arrogance, and the way they have to force the IT industry to leave the OS business to it, they should be held to the highest standards. We shouldn't cut them any slack just because they happened to discover a critical bug just before release for once.
Most probably.
I also always wondered if they take advantage of people sending stuff to another gmail account by not actually duplicating attached files (with some reference counting scheme to detect files that no one use anymore)
So they have been the first to implement an obvious idea (storing numeric values using floats instead of integers to get the kind of advantages one usually get by using floating point numbers instead of integers? OMG THAT'S BRILLIANT RUSH TO TEH PATENT OFFICE NOW!!1!).
Why should they be entitled them to reap benefit from other people managing to build a successful product using the same, painfully obvious idea?
I just hate the patent system for this.
There is not only the issue of being able to complain, but also the issue of being properly informed.
You can't exercise your democratic rights properly if you're not properly (or at all) informed of what your government does wrong.
Not trying to ignite a political flamewar, just speaking in general.
+5 Informative on a post explaining how much HP a children would have in D&D, which is a reply to a mildly joke vaguely related to the subject at hand (which is Jack Thompson getting his ass handed to him by the justice system)?
Only on Slashdot.
The version I'm talking about, the one I (and most of my co-workers) dislike is the professional edition, not the free one. You know, the one that the company I work for had to spend some big bucks on (given that we are around 20 C++ coders only on the project I'm part of, with two other projects in development in house).
This is getting hilarious. First, "B... b... b... but free ides don't do better", then "B... b... b... but there is a free version of visual studio for hobbyists, so it's free and therefore shouldn't be expected to be better than free IDEs, even the non-free version".
> Are there any IDEs that offer nontrivial c++ refactorings?
Does visual even offer trivial refactorings?
And what's with using the excuse of what the concurrence does or doesn't do to justify that this expensive product doesn't do it anyway? That's what I was on about regarding mediocrity. And I should remind you that most other IDEs are free. Visual is not by any stretch, so why would they get off the hook because some free competitor doesn't do something either?
This "no one else does XXX, so why bother" mentality is not one that spawn good and innovative products. This is a mentality that spawns mediocre shit.
Well, yeah, 7 needs more CPU than 6. But they're separated by ... what? 5 years? And it does far more than 6 ever did. This is to be expected.
What does the number of years separating two versions have with the cpu consumption?
And what did 7 do that 6 didn't do that could justify using more CPU?
By the time you have projects large enough to start choking VS, most of your other IDEs will have choked as well, unless they don't support the inspection necessary to support things like intellisense or refactorings.
Where are the C++ refactorings? I must have missed them.
And intellisense doesn't work most of the time.
Oh, and I only actually work on a subset of these projects most of the time. I guess they haven't heard of the concept of lazy updates.
They seem to need some reading on the concept of multithreading, too. Eclipse does its indexing in a background task.
Visual decide to do it at any random time and becomes unavailable for a few minutes when it does it (sometime after I start using it again after leaving it idle for a while for a reson or another).
I didn't say "it been going downhill since 5", I said "it's going downhill".
7 was good for a lot of things, but you could already feel the performance starting going down the drain. Nothing critical yet, but there's clearly an architectural change that wasn't that good happening from 6 to 7.
Then comes 2005. As far as C++ is concerned, it brought basically two things: improving the data breakpoint stuff so that you don't have to bypass their context thingy that never seemes to work and just write "*(long*)0xADDR", and displaying the contents of stl containers.
Scalability-wise, it sucks. Have you ever used it with a solution containing several dozen of projects and thousand of files?
It might very well be the cause of most of the performance problems I experience. But then again, I'd expect an expensive, so called "top of the line" ide to be able to scale.
Don't bash Microsoft for the sake of bashing Microsoft.
Oh, sorry. I'll just keep that in mind next time it crap itself on me. I guess that all that frustration I get dealing with their crap all day long is just something I invented, too.
The VS IDE is one of the greatest.
You're clearly not picky.
Seriously, one day I'll have to capture a video of one of my working session with that thing. But then again, people like you will explain me that whatever happens is somehow my fault.
I don't hate IDEs.
As for which IDEs are better, I think about any one of them is. Even KDevelop, which has the advantage of not being gratuitously slow (I use a goddamn 3.3.something SVN version of kdevelop at home, and despite the crashes it's snappy and so enjoyable to use)
Heck, even Eclipse is faster than that turd.
As for your other remark, I'm not disliking Microsoft out of immaturity or zealousness. I've been using their products professionally (translation: most of the time I am awake is spent working with these) for more than 6 years. I used every version of their ide since visual 5, and it's going downhill.
Which IDEs have you used on a day to day basis except visual? If the answer is none, you have no basis of comparison to quantify the quality of visual.
Right at the top?
Heh. Just goes to show the culture of mediocrity that symbolize the Microsoft era of computer science.
Right after this post, when I click into the visual studio window and it takes 30 seconds of doing god knows what before I can do anything (and perhaps it will throw a complete, recursive update of the solution explorer, redrawing the damn thing after reinsterting each file), I'll keep in mind that it's "right at the top as far as IDEs go".
I guess you get what you pay for.
Oh wait no, you don't. This thing cost an arm and a leg.
1) Monodevelop is a pile of shit right now.
.net 2005 how?
And this is different from visual
Of course, perhaps the lack of any competition in the c# ide market makes it actually the best there is. Another good reason to avoid c# altogether.
(I must admit that I only experienced the c++ part of that awful ide. On the other hand, I'd have a hard time to believe that the general sluggishness, random hang ups, gratuitous corruption of project files, abysmal useability and other awesome features are c++ only)
Ok, I understood what you wrote backwards.
But still, as you point out, we know how life works on ONE planet. How can we assume that it always works like this?
I think that the idea is that some rudimentary life form could exist on Mars and unable to evolve and thrive into somethign very complex because of the environment.
Life doesn't imply sentience.
You forgot:
- fud
- notfud
I never noticed that this guy had the same name as that gnu kernel thing, now that's a funny and novel joke.
I can barely contain my laughter. Well done, sire.
I think they might aswell get rid of the parametric prims in second life and just store regular meshes. People seem to make so complicated things in that game anyway that I think it would be much more efficient (with client side caching of the meshes like they do for textures, and along with some form of progressive lod)
Or perhaps a mix of primitive based stuff for simple objects like walls etc., and switch to a mesh representation for anything complex.
Well, on one hand the procedure installation is needlessly complicated. By the way, you can forego the awful gentoo live cd and use knoppix, kubuntu or whatever live cd, which allows to skip the tedious first steps of the installation manual and start directly with the partitioning. And you can use a good browser to read the doc as you go, too.
On the other hand, the installation procedure teaches you the basics of administrating your system: a lots of things aren't entirely automated or wrapped into nice GUI because you don't necessarily do them often, and you don't want to have any choice forced on you. That's the point of Gentoo, really.
You forego some ease of installation to be able to decide exactly what goes in your system and what doesn't, and for the benefit to be quickly able to use bleeding edge packages.
How can you care about someone if you don't know they exist?
I would care for the death of such a person if I knew about it.
:)
You don't have to be a friend or from the family of someone to care about his death.
If the human race were to be obliterated at once however, there would be no one (friend, family, stranger, or even foe) at all to worry about it as everyone would be a casualty. Of course, perhaps some aliens somewhere else in the universe could know about us, but then again, we don't know about them, so it's a moot point