No his problem is that QT has an execution loop which incompatible with the Chrome engine. What makes QT so cool for event driven programming is an event handler that can't be easily changed to match the event handler in Chrome.
Offer a series of easily configurable systems loaded to the gills with software. All-in-one boxes with "everything you need". Moreover the update system can go through the hardware vendor which can also include support for things like data backup which locks the customer in.
Application developers shouldn't be targeting desktops. They should be working with the distribution system. So in other words helping: RedHat, Mandriva, Debian... bring out their version of Chrome and let them distribute the packages.
That's the big problem, commercial app developers want to bypass the distributions without understanding that is the natural point of contact.
NPOV asserts that no view is "the truth", "If your viewpoint is held by an extremely small minority, then â" whether it's true or not, whether you can prove it or not â" it doesn't belong in Wikipedia,"
I guess we need to take down the page about the Titanic sinking.
Why? The vast majority of the media regarding the Titanic indicates it sunk.
Or about Hitler being a bad guy, since the media most certainly did not support that point prior to at least 1941.
I'm not sure that is true at all. There was a lot of opposition to Hitler in the 1930s. But even assuming it were, so what. This is 2009 and in 2009 the overwhelming majority of reliable sources don't have a pro-Nazi position.
uess we need to take down the "fake" label on the "protocols of the elders of zion" pages since there is scarcely any media source in the non-western world that acknowledges it's fake (Iran's papers are certainly not alone in still considering them accurate)
I'm a pretty regular reader of Al-Ahram, and I have to tell you that just ain't so. The vast majority of Arab media considers protocols to be a fake and believes that tying Arab "legitimate" issues with Israel to racism is a mistake.
You see, "the vast majority of well-respected media" is a moving, very ill-defined target.
It is moving but so far it seems well defined.
Yes there is much open racism against jews all over the muslim world, and against hindus and against christians, but why not let them tell their story ?
I think if you were able to reliable sources for what you listed above you would be able to include those things. For example your comment regarding the child has a whole section in an article: http://en.wikipedia.org/wiki/Criticism_of_Muhammad#Aisha
Absolutely it does. But all authors (editors) are required to support the idea of building a NPOV encyclopedia. Neutrality isn't supposed to be something that happens accidentally but something that editors are deliberately striving to achieve.
And wikipedia has a very robust definition of neutral. Sure there are areas of judgement but what the Scientology editors were doing was well outside those areas.
t would be better to give scientology itself a page about themselves that only they can edit, that is labeled as such.
Every user has them, their user page. Editorials are permitted there. But every page in the encyclopedia has to have a neutral point of view. There are other wikis which allow biased pages. This is not MySpace
1) Language to MSIL 2) MSIL to JIT native 3) JIT native to low level runtime instructions
Those 3 steps are called the ".NET compiler"
C# -> MSIL is an example of the first step, as is VB.NET -> MSIL.
As far as the rest I'm including Visual Basic.NET. As for GCC that would include: C, C++, Objective-C, Fortran, Java, and Ada using the GCC front ends and unified back end (i.e. compiled under GCC).
I think the rest was hinging on a one language assumption.
Honestly I think the big corps make the majority of good music too.:-) We don't what appears in the vacuum once it exists. It could be filled with nothing or it could get filled with all sort of mediums which don't sell products very well. Perhaps we get high culture government subsidized, like we had 50 years ago. Great opera, symphonies, community theater...
As for gaming.... I'm old enough to remember what computer gaming was like in the early 80s and it was rather bleak by today's standards. But things improved rapidly and late 80s early - 90s was a time of huge experimentation and diversity. Small budget gaming is having trouble living in a world of big budget gaming right now. Who knows?
Most estimates on gaming on piracy is that it has little impact. Pirates buy lots of games and non pirates by few if any.
I have mixed feelings. The ones you are talking about aren't really Web 2.0 there are a very small number of contributors. The only 2.0 aspect is the player interactions. I'd say something like Second Life is coming close to what I would think of as really having user created content on a large scale.
What he was saying as far as I can tell is given no concern for programmer time, C is always faster than higher level languages which is basically true at least for mainstream CPUs. Your three examples here are easy to do in LISP but I'm not sure they are faster.
You seem for some reason to be comfortable letting high level languages use libraries but not C. To keep hammering, C can call the mathematica kerne via. the APIs and get the same or very slightly better performance.
So for example
1) C doesn't use this terminology. But just about any function application in any interpreted language is going to be a unification. If it is going to be compiled, you will end up with void function invocations and those are blazing fast in C. That's going to compile down to instruction loader and a memory address (essentially a goto).
2) c does support self modification. But LISPs are for example written in C. As for a JIT that has a good C,.NET is a fair comparison where you have a bytecode system with a C. And the C is just going to be a bunch of CIL calls so again fast.
3) Schemes are written in C. That's not a small program anymore.
Qi did basically the benchmarks you propose. Fairly complex routines written in Qi vs. low level common Lisp. Low level common lisp for 30 line QI program was about 11% faster. If that was worth taking a 1 hour job and converting into a 10 hour job is a big question. C probably would have bought another 20% in exchange for making it a 5000 line program that took 3 weeks.
About 8 years ago the CEO of Fox Media gave a keynote at Comdex on this topic. What they wanted was something like the trusted computing initiative. Moreover what he wanted was a partnership with IT and broad support from the developer / hardware communities. He realizes that the IT people by and large support the free exchange of information and thus undermine this partnership.
His feeling was that there were potentially hundreds of thousands to millions of jobs in IT supporting a massive customized entertainment system, that could exist if and only if the medium was relatively safe.
I'm not sure that with Web 2.0 another alternative, the return of the amateur, isn't the direction we are heading in instead.
RPL came out with the 28c in 1987. I personally used a 28s starting 1988. We are talking around the same time since the 48s came out when I was a junior. If they were using something like a 15c compared to a Casio, yeah I can imagine there would have been a difference.
My point is if you want to test coding benchmarks don't ask for a test that does nothing more than invoke a library. Your Integrate example is a little better but you are still just using a library.
Something like cram game is what you would want to test.
That's odd. Given how good the RPL was as a language for numerical computation I would have thought that the RPL guys beat the BASIC guys. It was effectively a higher level language.
I think the Debian Stable/Testing/Unstable system is fantastic. Same thing with Fedora/RedHat split.
One of the solution that started to work well around Mandrake 7 was their package updater (usually for bugs). You take a 6 month old Mandrake, install, download/install the 200 mb of package updates and your system works well.
Mandrake had a tough situation. The kinds of people who were attracted to the also ran desktop apps wanted newer software especially stuff like new versions of KDE and the main apps. Quality packages cost a fortune. So IMHO they did the best they could given their resources. Great distribution.
Slashdot Mirror
No his problem is that QT has an execution loop which incompatible with the Chrome engine. What makes QT so cool for event driven programming is an event handler that can't be easily changed to match the event handler in Chrome.
The business case:
Offer a series of easily configurable systems loaded to the gills with software. All-in-one boxes with "everything you need". Moreover the update system can go through the hardware vendor which can also include support for things like data backup which locks the customer in.
Application developers shouldn't be targeting desktops. They should be working with the distribution system. So in other words helping: RedHat, Mandriva, Debian... bring out their version of Chrome and let them distribute the packages.
That's the big problem, commercial app developers want to bypass the distributions without understanding that is the natural point of contact.
Its been tried:
UserLinux, United Linux....
I think you are ranting here. There is a whole category on Dalit related articles:
http://en.wikipedia.org/wiki/Category:Dalit
There is a dedicated article on various interpretations of the crusades:
http://en.wikipedia.org/wiki/Historiography_of_the_Crusades
I addressed the pedophilia issue above
In terms of slavery in the modern islam world:
http://en.wikipedia.org/wiki/Slavery_in_Islam#Slavery_in_the_contemporary_Muslim_world
etc...
The very first line of the verifiability policy is "The threshold for inclusion in Wikipedia is verifiability, not truth"
http://en.wikipedia.org/wiki/Wikipedia:Verifiability
NPOV asserts that no view is "the truth", "If your viewpoint is held by an extremely small minority, then â" whether it's true or not, whether you can prove it or not â" it doesn't belong in Wikipedia,"
Why? The vast majority of the media regarding the Titanic indicates it sunk.
I'm not sure that is true at all. There was a lot of opposition to Hitler in the 1930s. But even assuming it were, so what. This is 2009 and in 2009 the overwhelming majority of reliable sources don't have a pro-Nazi position.
I'm a pretty regular reader of Al-Ahram, and I have to tell you that just ain't so. The vast majority of Arab media considers protocols to be a fake and believes that tying Arab "legitimate" issues with Israel to racism is a mistake.
It is moving but so far it seems well defined.
They can. Their are tons of articles on those topics like:
http://en.wikipedia.org/wiki/Anti-Brahminism
http://en.wikipedia.org/wiki/Neo-Nazism
http://en.wikipedia.org/wiki/Anti-Hinduism
etc...
I think if you were able to reliable sources for what you listed above you would be able to include those things. For example your comment regarding the child has a whole section in an article:
http://en.wikipedia.org/wiki/Criticism_of_Muhammad#Aisha
The racial stuff you mention:
http://en.wikipedia.org/wiki/Race_and_intelligence
http://en.wikipedia.org/wiki/Race_(classification_of_human_beings)#Race_as_subspecies
etc...
Absolutely it does. But all authors (editors) are required to support the idea of building a NPOV encyclopedia. Neutrality isn't supposed to be something that happens accidentally but something that editors are deliberately striving to achieve.
And wikipedia has a very robust definition of neutral. Sure there are areas of judgement but what the Scientology editors were doing was well outside those areas.
They way they do it. A popular viewpoint representing the vast majority of well respected media on that topic.
Every user has them, their user page. Editorials are permitted there. But every page in the encyclopedia has to have a neutral point of view. There are other wikis which allow biased pages. This is not MySpace
Pro scientology viewpoints are not banned. A group of editors is banned. They aren't banning a viewpoint but a subgroup of people.
Compilation in windows involves 3 steps
1) Language to MSIL
2) MSIL to JIT native
3) JIT native to low level runtime instructions
Those 3 steps are called the ".NET compiler"
C# -> MSIL is an example of the first step, as is VB.NET -> MSIL.
As far as the rest I'm including Visual Basic.NET. As for GCC that would include: C, C++, Objective-C, Fortran, Java, and Ada using the GCC front ends and unified back end (i.e. compiled under GCC).
I think the rest was hinging on a one language assumption.
GCC doesn't have anybody behind it either. It is now, I think most would admit the #2 compiler after the .NET compiler.
Apache had a group about the same size and it is arguably in first place.
Linux is clearly successful in the server and embedded markets.
Success means at the very least that desktop purchasers see it as an option and actively decide for or against it in a meaningful way.
Honestly I think the big corps make the majority of good music too. :-) We don't what appears in the vacuum once it exists. It could be filled with nothing or it could get filled with all sort of mediums which don't sell products very well. Perhaps we get high culture government subsidized, like we had 50 years ago. Great opera, symphonies, community theater...
As for gaming.... I'm old enough to remember what computer gaming was like in the early 80s and it was rather bleak by today's standards. But things improved rapidly and late 80s early - 90s was a time of huge experimentation and diversity. Small budget gaming is having trouble living in a world of big budget gaming right now. Who knows?
Most estimates on gaming on piracy is that it has little impact. Pirates buy lots of games and non pirates by few if any.
I have mixed feelings. The ones you are talking about aren't really Web 2.0 there are a very small number of contributors. The only 2.0 aspect is the player interactions. I'd say something like Second Life is coming close to what I would think of as really having user created content on a large scale.
His thinking was about $1k / household / year x 130m american households + foreign. The idea would be:
1) Unlimited commercial free or commercial television shows from every show every made ....
2) Essentially every movie
3) All the electronic music
Would be worth some extra money.
What he was saying as far as I can tell is given no concern for programmer time, C is always faster than higher level languages which is basically true at least for mainstream CPUs. Your three examples here are easy to do in LISP but I'm not sure they are faster.
You seem for some reason to be comfortable letting high level languages use libraries but not C. To keep hammering, C can call the mathematica kerne via. the APIs and get the same or very slightly better performance.
So for example
1) C doesn't use this terminology. But just about any function application in any interpreted language is going to be a unification. If it is going to be compiled, you will end up with void function invocations and those are blazing fast in C. That's going to compile down to instruction loader and a memory address (essentially a goto).
2) c does support self modification. But LISPs are for example written in C. As for a JIT that has a good C, .NET is a fair comparison where you have a bytecode system with a C. And the C is just going to be a bunch of CIL calls so again fast.
3) Schemes are written in C. That's not a small program anymore.
Qi did basically the benchmarks you propose. Fairly complex routines written in Qi vs. low level common Lisp. Low level common lisp for 30 line QI program was about 11% faster. If that was worth taking a 1 hour job and converting into a 10 hour job is a big question. C probably would have bought another 20% in exchange for making it a 5000 line program that took 3 weeks.
About 8 years ago the CEO of Fox Media gave a keynote at Comdex on this topic. What they wanted was something like the trusted computing initiative. Moreover what he wanted was a partnership with IT and broad support from the developer / hardware communities. He realizes that the IT people by and large support the free exchange of information and thus undermine this partnership.
His feeling was that there were potentially hundreds of thousands to millions of jobs in IT supporting a massive customized entertainment system, that could exist if and only if the medium was relatively safe.
I'm not sure that with Web 2.0 another alternative, the return of the amateur, isn't the direction we are heading in instead.
RPL came out with the 28c in 1987. I personally used a 28s starting 1988. We are talking around the same time since the 48s came out when I was a junior. If they were using something like a 15c compared to a Casio, yeah I can imagine there would have been a difference.
My point is if you want to test coding benchmarks don't ask for a test that does nothing more than invoke a library. Your Integrate example is a little better but you are still just using a library.
Something like cram game is what you would want to test.
That's odd. Given how good the RPL was as a language for numerical computation I would have thought that the RPL guys beat the BASIC guys. It was effectively a higher level language.
Any insight as to what caused the slowdown?
What is that a test of.
1) F[33622711352843762413542039873684935021017] converts instantly to
2)Length[FactorInteger[33622711352843762413542039873684935021017]
Which essentially takes the same time as
3) FactorInteger[33622711352843762413542039873684935021017]
Which depends on your FactorInteger routine which is coded in C in mathematica.
That's not really a test of anything other than a very mature mathematical piece of software has good math libraries.
I think the Debian Stable/Testing/Unstable system is fantastic. Same thing with Fedora/RedHat split.
One of the solution that started to work well around Mandrake 7 was their package updater (usually for bugs). You take a 6 month old Mandrake, install, download/install the 200 mb of package updates and your system works well.
Mandrake had a tough situation. The kinds of people who were attracted to the also ran desktop apps wanted newer software especially stuff like new versions of KDE and the main apps. Quality packages cost a fortune. So IMHO they did the best they could given their resources. Great distribution.