Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
I never said Prolog and Lisp are the same, just they're generally suited for the same problem domain. And they both have strong metaprogramming/symbolic computation features. (And I could have mentioned Oz, which combines the two paradigm, and is also homoiconic, but that's relatively new language)
When using pure Prolog (ISO or Edinborough) you rarely can enjoy the benefits of the logic style (i.e. just stating the problem and the computer solving it for you). When you're doing a theorem prover, you will write that procedural style, as the actual way of execution matters a great deal in performance, but the user of you theorem prover will use it declarative style. (I.e. inside procedural, outside API declarative). And that's very similar how you do it in Lisp. (See PDDL) People in US tend to use Lisp, people in EU use Prolog for the same type of problem, however established domains decide for a particular language and stick with it. So constraint programming went with Prolog, PDDL with LISP.
Please, what was Einstein's contribution to the Manhattan project besides the Einstein–Szilárd letter?
"In 1939, a group of Hungarian scientists that included Hungarian emigre physicist Leó Szilárd attempted to alert Washington of ongoing Nazi atomic bomb research. The group's warnings were discounted.[45]
Einstein and Szilárd, along with other refugees such as Edward Teller and Eugene Wigner, "regarded it as their responsibility to alert Americans to the possibility that German scientists might win the race to build an atomic bomb, and to warn that Hitler would be more than willing to resort to such a weapon."[40]:630[46] In the summer of 1939, a few months before the beginning of World War II in Europe, Einstein was persuaded to lend his prestige by writing a letter with Szilárd to President Franklin D. Roosevelt to alert him of the possibility. The letter also recommended that the U.S. government pay attention to and become directly involved in uranium research and associated chain reaction research."
Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
Because both are homoiconic AI languages. It takes about 100 lines to write a Prolog parser in Lisp, or to write a Lisp parser in Prolog. You can't do that for C-Lisp or C-Prolog combination. Also, it would help if you actually read my posts. They're really short, it doesn't take much effort.
Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
1, Prolog's main use is logical reasoners, especially the description logic (OWL/RDF/SPARQL etc.) 2, If you do something related to semantic web, it's useful to have a web server. As ISO Prolog defines high level networking interfaces, it it's not a really hard thing to do anyway.
But you would have known this, if you had any experience as a Prolog programmer.
Also see homoiconicity, I won't link it here again.
Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
You remind me the guy who told me this week that Glassfish sucks and nobody uses it, we need a stable proven technology like PHP/MySQL.
Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
"Prolog is a declarative programming language, like SQL. You don't use it to define an algorithm. The algorithm(s) are already in the runtime system."
And when it isn't there, you program it in procedural style. Prolog has web server, semantic reasoners, etc. These stuff had to be programmed procedural style. Any IO you do, you have to do it procedural style, unless you use it together with a different language to do the IO/GUI. Prolog never strived for the same level of declarative purity as Haskell, so in this respect it's more similar to Lisp.
I haven't seen any bugs in the Windows version, on the other hand it loads a bit slow, and the scripting API is just horribly convoluted, and underdocumented.
Owen recruited residents to his model community, but a number of factors led to an early breakup of the socialist community. [...]
The experiment was established in 1825 and dissolved in 1829 due to constant quarrels. The town banned money and other commodities. Individualist anarchist Josiah Warren, who was one of the original participants in the New Harmony Society, asserted that the community was doomed to failure due to a lack of individual sovereignty and private property. He wrote of the community: "It seemed that the difference of opinion, tastes and purposes increased just in proportion to the demand for conformity. Two years were worn out in this way; at the end of which, I believe that not more than three persons had the least hope of success. Most of the experimenters left in despair of all reforms, and conservatism felt itself confirmed. We had tried every conceivable form of organization and government. We had a world in miniature. --we had enacted the French revolution over again with despairing hearts instead of corpses as a result....It appeared that it was nature's own inherent law of diversity that had conquered us...our 'united interests' were directly at war with the individualities of persons and circumstances and the instinct of self-preservation... and it was evident that just in proportion to the contact of persons or interests, so are concessions and compromises indispensable." (Periodical Letter II 1856).
Yeah, I did the same with Firefox and KToon. And I think that's kind of ideal: you get old and stable in infrastructure packages, and new in applications.
"Overall this is a lock-in story by Microsoft. It is what it is and personally I think WinRT will fail overall because it means you are completely beholden to the Windows platform"
Well, you can say that about almost all Windows technologies: WPF, Silverlight, PowerShell, Windows Scripting Host, Active Directory, VBA etc.
You mean we can write C# applications using Metro that runs on tablets, but with a leaner C# library? Or if you want to target a tablet, you have to write C++? (I consider the latter less likely. )
Of course I know that, but the developer doesn't have to go hunt for zillions of different opensource/commercial libraries. And can expect decent documentation.
" Only after the low level programming has been done by somebody else there will be high level programming support. You're mistaken in where the power comes frome. "
And it's already done by the vendor (be it Oracle/Sun on PC, or phone manufacturer in Android or Windows Phone). The thing is, when you got the proper hardware abstraction libraries, the languages are on equal footing, and there's much easier to debug an application in Java/C# than in C/C++. Also, with Java/C# you can expect that the libraries will be somewhat standardized even if the hardware has different capabilities, so you don't have to learn new APIs all the time to do the same thing. This advantage might come from the fact that Java/C# are trademarked and are controlled by corporations, but it's still an advantage for the average developer.
However application developers usually aren't happy with "basic hardware support", they want to OpenGL, H264 hardware acceleration, want to access webcam, microphone etc. Pure ANSI C is fine and dandy, but the real power comes from the (cross-platform) libraries, and that's what Java/Python/C# got right. People want batteries included.
Slashdot Mirror
If everybody could comment on the internet, you would end up with a lot of boring shit. We need an editor to approve them before they appear.
Hi, Bill Gates.
I never said Prolog and Lisp are the same, just they're generally suited for the same problem domain. And they both have strong metaprogramming/symbolic computation features. (And I could have mentioned Oz, which combines the two paradigm, and is also homoiconic, but that's relatively new language)
When using pure Prolog (ISO or Edinborough) you rarely can enjoy the benefits of the logic style (i.e. just stating the problem and the computer solving it for you). When you're doing a theorem prover, you will write that procedural style, as the actual way of execution matters a great deal in performance, but the user of you theorem prover will use it declarative style. (I.e. inside procedural, outside API declarative). And that's very similar how you do it in Lisp. (See PDDL)
People in US tend to use Lisp, people in EU use Prolog for the same type of problem, however established domains decide for a particular language and stick with it. So constraint programming went with Prolog, PDDL with LISP.
Please, what was Einstein's contribution to the Manhattan project besides the Einstein–Szilárd letter?
"In 1939, a group of Hungarian scientists that included Hungarian emigre physicist Leó Szilárd attempted to alert Washington of ongoing Nazi atomic bomb research. The group's warnings were discounted.[45]
Einstein and Szilárd, along with other refugees such as Edward Teller and Eugene Wigner, "regarded it as their responsibility to alert Americans to the possibility that German scientists might win the race to build an atomic bomb, and to warn that Hitler would be more than willing to resort to such a weapon."[40]:630[46] In the summer of 1939, a few months before the beginning of World War II in Europe, Einstein was persuaded to lend his prestige by writing a letter with Szilárd to President Franklin D. Roosevelt to alert him of the possibility. The letter also recommended that the U.S. government pay attention to and become directly involved in uranium research and associated chain reaction research."
1933-1945: that's 12 years, not too much.
Because both are homoiconic AI languages.
It takes about 100 lines to write a Prolog parser in Lisp, or to write a Lisp parser in Prolog. You can't do that for C-Lisp or C-Prolog combination.
Also, it would help if you actually read my posts. They're really short, it doesn't take much effort.
http://web.student.tuwien.ac.at/~e0225855/lisprolog/lisprolog.html
http://treewalker.wordpress.com/p-lisp-a-lisp-interpreter-in-prolog/
1, Prolog's main use is logical reasoners, especially the description logic (OWL/RDF/SPARQL etc.)
2, If you do something related to semantic web, it's useful to have a web server. As ISO Prolog defines high level networking interfaces, it it's not a really hard thing to do anyway.
But you would have known this, if you had any experience as a Prolog programmer.
Also see homoiconicity, I won't link it here again.
You remind me the guy who told me this week that Glassfish sucks and nobody uses it, we need a stable proven technology like PHP/MySQL.
http://www.swi-prolog.org/pldoc/doc/swi/library/http/http_dispatch.pl?show=src
http://www.swi-prolog.org/pldoc/doc/swi/library/semweb/sparql_client.pl?show=src
Next time try some logical argument, ad hominems won't suffice.
Also see:
http://en.wikipedia.org/wiki/Homoiconicity
"Prolog is a declarative programming language, like SQL. You don't use it to define an algorithm. The algorithm(s) are already in the runtime system."
And when it isn't there, you program it in procedural style. Prolog has web server, semantic reasoners, etc. These stuff had to be programmed procedural style. Any IO you do, you have to do it procedural style, unless you use it together with a different language to do the IO/GUI.
Prolog never strived for the same level of declarative purity as Haskell, so in this respect it's more similar to Lisp.
I haven't seen any bugs in the Windows version, on the other hand it loads a bit slow, and the scripting API is just horribly convoluted, and underdocumented.
Yeah sightseeing in chinatown. Sounds reeeally fun ...
"Steve Jobs famously dropped out of college in order to study calligraphy. "
He studied calligraphy in a college. (While not being enrolled.)
Communism: this time it will work.
Owen recruited residents to his model community, but a number of factors led to an early breakup of the socialist community. [...]
The experiment was established in 1825 and dissolved in 1829 due to constant quarrels. The town banned money and other commodities. Individualist anarchist Josiah Warren, who was one of the original participants in the New Harmony Society, asserted that the community was doomed to failure due to a lack of individual sovereignty and private property. He wrote of the community: "It seemed that the difference of opinion, tastes and purposes increased just in proportion to the demand for conformity. Two years were worn out in this way; at the end of which, I believe that not more than three persons had the least hope of success. Most of the experimenters left in despair of all reforms, and conservatism felt itself confirmed. We had tried every conceivable form of organization and government. We had a world in miniature. --we had enacted the French revolution over again with despairing hearts instead of corpses as a result. ...It appeared that it was nature's own inherent law of diversity that had conquered us ...our 'united interests' were directly at war with the individualities of persons and circumstances and the instinct of self-preservation... and it was evident that just in proportion to the contact of persons or interests, so are concessions and compromises indispensable." (Periodical Letter II 1856).
http://en.wikipedia.org/wiki/New_Harmony,_Indiana#History
So what are you running?
Yeah, I did the same with Firefox and KToon. And I think that's kind of ideal: you get old and stable in infrastructure packages, and new in applications.
What about security patches? I guess you can't get them for your 2004 system as those versions aren't supported anymore.
I agree, but when I used Fedora it was even buggier. You can choose between up-to-date and stable. Writing this from 10.04 LTS.
"Overall this is a lock-in story by Microsoft. It is what it is and personally I think WinRT will fail overall because it means you are completely beholden to the Windows platform"
Well, you can say that about almost all Windows technologies: WPF, Silverlight, PowerShell, Windows Scripting Host, Active Directory, VBA etc.
You mean we can write C# applications using Metro that runs on tablets, but with a leaner C# library? Or if you want to target a tablet, you have to write C++? (I consider the latter less likely. )
Isn't Metro just a different GUI library on top of .Net?
Of course I know that, but the developer doesn't have to go hunt for zillions of different opensource/commercial libraries. And can expect decent documentation.
" Only after the low level programming has been done by somebody else there will be high level programming support. You're mistaken in where the power comes frome. "
And it's already done by the vendor (be it Oracle/Sun on PC, or phone manufacturer in Android or Windows Phone). The thing is, when you got the proper hardware abstraction libraries, the languages are on equal footing, and there's much easier to debug an application in Java/C# than in C/C++. Also, with Java/C# you can expect that the libraries will be somewhat standardized even if the hardware has different capabilities, so you don't have to learn new APIs all the time to do the same thing. This advantage might come from the fact that Java/C# are trademarked and are controlled by corporations, but it's still an advantage for the average developer.
I think in Hungary it works similar to the US. Also, evidence can't be reused in a different trial.
However application developers usually aren't happy with "basic hardware support", they want to OpenGL, H264 hardware acceleration, want to access webcam, microphone etc. Pure ANSI C is fine and dandy, but the real power comes from the (cross-platform) libraries, and that's what Java/Python/C# got right. People want batteries included.