[If] you, for a single moment, believe that ending was made up on the spot....
I'll do you one better.
Romo Lampkin: Gosh, ex-president Lee Adama! I never would have thought that everyone would spontaneously and unanimously vote to throw away all technology, including all of the lampshades in the fleet, but golly gee they sure did, and after I finish this sentence it'll be canon, so let's get on with the finale, why don't we, because we're a few maudlin and confusing scenes away from cute dancing robots made sinister by a pointless voiceover?
Participants in a FOSS project are largely self selected, with few obstacles or inducements other than personal interest and skill level.
Of course.
If you talk to multiple women who seem likely to contribute to free software projects, how many of them will tell you that they do not feel comfortable participating in communities which demonstrate certain tendencies? I've heard that from quite a few potential contributors -- not solely women, but people with many characteristics which differentiate them from the average 20-30 something white upper middle class male with a taste for flamewar.
Nominations take place from September until February. President Obama's nomination had nothing to do with anything that's happened since February 2009.
As for Moose, it's not part of the core language today.
So what?
Plenty of widely-used Ruby libraries aren't part of the core language. Is Rake in 1.9 yet? The Enterprise Ruby patches make Ruby much more usable, and they're not in the core language either.
To say, however, that Perl 5 is better than Ruby 1.8/1.9 in every respect....
That would be stupid, which is why I never wrote that.
As I wrote earlier, the 23rd monthly release in a row of Rakudo Perl 6 will occur in two weeks. It supports more features than Ruby 1.9 does and you can use it today. It may not support the features your code needs -- but that's okay. Ruby 1.9 doesn't support the features my code needs.
Have you spent at least 6 months trying to become proficient in Ruby?
Indeed I have. It's a decent language. It has nothing on Perl 6 (and, frankly, it's missing a lot of nice features of Perl 5).
Where do you get proof for that....
Larry's exact quote is "I don't really know much about Python. I only stole its object system." You should have no trouble finding confirmation on the Internet.
Both Ruby 1.8 and 1.9 have severely more powerful OOP constructs than Perl 5.
Nonsense. Compare Perl roles to Ruby monkeypunching.
Perl 6 is "any decade now."
The twenty-second monthly release of Rakudo Perl 6 in a row will take place in two and a half weeks. You're almost two years out of date with respect to your Perl 6 information.
Perl 5 has bolt-on class-based objects. These superficially resemble C++/Java/C#'s classes.
Nonsense.
Ruby has objects resembling the king of OOP: Smalltalk.
Perl 5 objects (using Moose) have more features from Smalltalk than Ruby objects do then. Of course, that's because Moose backported them from Perl 6, which borrowed some of the formalisms from Smalltalk for the design of the particular feature I have in mind.
Ruby has open classes, true exception handling, method_missing(), duck typing, and reflection. Lambdas are truly useful (check out iterators).
Perl 5 has all of those as well. Perl 6 adds built-in support for roles, class finalization, junction types, optional typing, serialization, multi dispatch, and more. You can try them today.
In Perl 5, this is not an error....
... because of Perl 5's type system and Perl's nature as an operator-oriented language. I can understand how this might be confusing if you never learned the language, but it's a deliberate design principle. Don't pretend it's an accident perpetuated by clueless buffoons who never saw the grape-flavored wonder that is Ruby.
Speed, by the way, is *not* one of Ruby's strengths, although its performance and scalability are greatly improving and have more "tweakability."
Over Perl 5? Unlikely. The last I looked (I haven't looked at Ruby 1.9.x), Ruby closures had to close over all lexicals in a program. There's a great way to reduce scalability, memory wise. (I can also mention that the last I knew of Ruby's GC, it unshared way too many pages during a naive mark and sweep.) As well, I seem to recall that a Perl 5 program won Tim Bray's Wide Finder benchmark.
However, the opinions that PHP is cleaner and easier to learn are shared by many....
Let's talk numbers. How many default functions and operators does PHP have? How many does Perl 5 have? How many of PHP's default functions and operators are compile-time settings or settings in php.ini?
Why not learn from other languages, like Ruby, Lisp, or Smalltalk, to become a better Perl programmer?
I first programmed Ruby in 2001. I've written my own Scheme. I have the Smalltalk-80 books on my desk, and I've referred to them in the past few days.
The only thing I consider worth learning from PHP is the necessity of ease of deployment.
Perl is notorious for unreadable code! Just check the opinions of *seasoned* Perl programmers on the net. There is an attitude that "if it was hard to write, it should be hard to read."
I would be happy to introduce you to dozens of seasoned Perl programmers who object strenuously to that mischaracterization.
[Perl 5's] "blessed" object model, for instance is an ugly, bolt-on hack.
Lifted straight from Python.
Perl 6, of course, isn't "finished" yet.
Neither is Ruby 2 or Python 3 or ECMAScript 4 or....
Unlike Perl 6, [Ruby is] here and now.
If Perl 6 wanted to support only those language features that Ruby 1.9 supports, Perl 6 would have been "here and now" years ago. That's a modest goal, and my goals are much, much more ambitious. Ruby's still playing catch up with Perl 5 (let's talk object systems, platform compatibility, library support, compiler warnings and errors, a built-in test suite...).
Ruby is a far more powerful language, and I can read other's code (and even my code, 6 months later).
I know plenty of Perl hackers who have no trouble reading their own code six months later. I suspect this reflects more on your code than the language.
Gems is cleaner, without all manner of compilation and portability issues (and less code rot).
Ruby extensions written in C require compilation. Ruby Gems would also have portability issues if they were actually tested on all of the platforms that CPAN supports.
In short, Ruby is a better Perl 5/6 than Perl 5/6....
In commit logs and occasional journal posts I've made. You should be able to find them by searching for "Parrot optimization" or "Rakudo optimization" or "Parrot benchmark" or "Rakudo benchmark". You're also welcome to ask any other developer of Parrot or Rakudo.
Great job defending FOSS from reality guys, congratulations.
You could read it that way. I read it more that moderators are tired of blowhard know-it-alls who love to complain but rarely back up their apparently strongly-held opinions.
I've written books in XML and I've written books in PseudoPOD. I don't write books in XML anymore. My company's entire production system uses PseudoPOD.
[The] Python community has shown with Python 3 that they're capable of working together to create a major release with many new features in a relatively short amount of time (especially compared to the Perl 6 effort).
Guido switched jobs in early 2000 in part to work on Python 3000. Eight and a half years is not a "relatively short amount of time" for the delta between Python 2 and Python 3.
Jordan took the time to create an entire universe and populate it entirely with three-dimensional characters and landscapes.
Characters don't count as three-dimensional if all female characters wear the same high-necked, well-embroidered dresses, have the same braid-tugging, nose-sniffing mannerisms, and consider the male characters wooly-headed fools, repeatedly.
A higher percentage of smaller sales is still less money last time I looked.
Suppose I write a book for you. It has a $30 cover price. That's roughly $15 wholesale. I get $1.50 per copy. Suppose it sells 5000 copies. That's $7500 for me.
Suppose I publish a book on my own. It has a $30 cover price. That's roughly $15 wholesale and no more than $5 to print and distribute (probably less than $4.25). I get at least $10 per copy. Suppose it sells 750 copies. That's
$7500 for me.
If I can sell more than 15% of the copies you can, I come out ahead.
I can't imagine that you'd publish a book you believe would sell only 5000 copies, however. Even if you did, I doubt it would receive sufficient editorial or production attention to justify the difference in author remuneration.
I've had several technical books published. The process has left me so dissatisfied I'm unlikely to work with a so-called traditional publisher ever again -- certainly not with a standard contract. See The Value of a Publisher for some thoughts representing what you give up to work with such a contract.
I believe that any publisher that refuses to pay you 25% royalties on the wholesale price is not worth your time.
Slashdot Mirror
I'll do you one better.
Romo Lampkin: Gosh, ex-president Lee Adama! I never would have thought that everyone would spontaneously and unanimously vote to throw away all technology, including all of the lampshades in the fleet, but golly gee they sure did, and after I finish this sentence it'll be canon, so let's get on with the finale, why don't we, because we're a few maudlin and confusing scenes away from cute dancing robots made sinister by a pointless voiceover?
Of course.
If you talk to multiple women who seem likely to contribute to free software projects, how many of them will tell you that they do not feel comfortable participating in communities which demonstrate certain tendencies? I've heard that from quite a few potential contributors -- not solely women, but people with many characteristics which differentiate them from the average 20-30 something white upper middle class male with a taste for flamewar.
Nominations take place from September until February. President Obama's nomination had nothing to do with anything that's happened since February 2009.
Try PAR.
So what?
Plenty of widely-used Ruby libraries aren't part of the core language. Is Rake in 1.9 yet? The Enterprise Ruby patches make Ruby much more usable, and they're not in the core language either.
That would be stupid, which is why I never wrote that.
As I wrote earlier, the 23rd monthly release in a row of Rakudo Perl 6 will occur in two weeks. It supports more features than Ruby 1.9 does and you can use it today. It may not support the features your code needs -- but that's okay. Ruby 1.9 doesn't support the features my code needs.
Indeed I have. It's a decent language. It has nothing on Perl 6 (and, frankly, it's missing a lot of nice features of Perl 5).
Larry's exact quote is "I don't really know much about Python. I only stole its object system." You should have no trouble finding confirmation on the Internet.
Nonsense. Compare Perl roles to Ruby monkeypunching.
The twenty-second monthly release of Rakudo Perl 6 in a row will take place in two and a half weeks. You're almost two years out of date with respect to your Perl 6 information.
Nonsense.
Perl 5 objects (using Moose) have more features from Smalltalk than Ruby objects do then. Of course, that's because Moose backported them from Perl 6, which borrowed some of the formalisms from Smalltalk for the design of the particular feature I have in mind.
Perl 5 has all of those as well. Perl 6 adds built-in support for roles, class finalization, junction types, optional typing, serialization, multi dispatch, and more. You can try them today.
... because of Perl 5's type system and Perl's nature as an operator-oriented language. I can understand how this might be confusing if you never learned the language, but it's a deliberate design principle. Don't pretend it's an accident perpetuated by clueless buffoons who never saw the grape-flavored wonder that is Ruby.
Over Perl 5? Unlikely. The last I looked (I haven't looked at Ruby 1.9.x), Ruby closures had to close over all lexicals in a program. There's a great way to reduce scalability, memory wise. (I can also mention that the last I knew of Ruby's GC, it unshared way too many pages during a naive mark and sweep.) As well, I seem to recall that a Perl 5 program won Tim Bray's Wide Finder benchmark.
Let's talk numbers. How many default functions and operators does PHP have? How many does Perl 5 have? How many of PHP's default functions and operators are compile-time settings or settings in php.ini?
I first programmed Ruby in 2001. I've written my own Scheme. I have the Smalltalk-80 books on my desk, and I've referred to them in the past few days.
The only thing I consider worth learning from PHP is the necessity of ease of deployment.
I'd really like to see you support most of this sentence.
Regarding the influence of Rails, note that the strongest influences for Catalyst (likely the strongest influence on Mojo) predate Rails measurably.
I would be happy to introduce you to dozens of seasoned Perl programmers who object strenuously to that mischaracterization.
Lifted straight from Python.
Neither is Ruby 2 or Python 3 or ECMAScript 4 or ....
If Perl 6 wanted to support only those language features that Ruby 1.9 supports, Perl 6 would have been "here and now" years ago. That's a modest goal, and my goals are much, much more ambitious. Ruby's still playing catch up with Perl 5 (let's talk object systems, platform compatibility, library support, compiler warnings and errors, a built-in test suite...).
I trade most milestones in the Perl renaissance directly to Perl 6.
(Sclerotic? If that's the word you intended, the metaphor is... difficult.)
Far more and far better than any other dynamic language of which I am aware, even.
I know plenty of Perl hackers who have no trouble reading their own code six months later. I suspect this reflects more on your code than the language.
Ruby extensions written in C require compilation. Ruby Gems would also have portability issues if they were actually tested on all of the platforms that CPAN supports.
I suspect you know very little about Perl 6 then.
In commit logs and occasional journal posts I've made. You should be able to find them by searching for "Parrot optimization" or "Rakudo optimization" or "Parrot benchmark" or "Rakudo benchmark". You're also welcome to ask any other developer of Parrot or Rakudo.
You could read it that way. I read it more that moderators are tired of blowhard know-it-alls who love to complain but rarely back up their apparently strongly-held opinions.
I've written books in XML and I've written books in PseudoPOD. I don't write books in XML anymore. My company's entire production system uses PseudoPOD.
No, I'm actually programming. How about you? What have you produced lately?
I have a year's worth of benchmarks and profiles that disagree with you.
For your first two points, try Padre. For your fourth point, try MooseX::Declare.
I disagree on your third point. Ouch.
Guido switched jobs in early 2000 in part to work on Python 3000. Eight and a half years is not a "relatively short amount of time" for the delta between Python 2 and Python 3.
If you fix the issues, Coverity moves the project to a new rung and performs stricter analysis to find more types of errors.
Their constraint is very likely QA. If QA is perpetually busy, it doesn't matter how many more programmers they add.
Compare the OpenSolaris kernel to the Linux kernel.
Characters don't count as three-dimensional if all female characters wear the same high-necked, well-embroidered dresses, have the same braid-tugging, nose-sniffing mannerisms, and consider the male characters wooly-headed fools, repeatedly.
I expect that there may be a few people in the world better at telling a consistent and coherent story than George Lucas.
Suppose I write a book for you. It has a $30 cover price. That's roughly $15 wholesale. I get $1.50 per copy. Suppose it sells 5000 copies. That's $7500 for me.
Suppose I publish a book on my own. It has a $30 cover price. That's roughly $15 wholesale and no more than $5 to print and distribute (probably less than $4.25). I get at least $10 per copy. Suppose it sells 750 copies. That's $7500 for me.
If I can sell more than 15% of the copies you can, I come out ahead.
I can't imagine that you'd publish a book you believe would sell only 5000 copies, however. Even if you did, I doubt it would receive sufficient editorial or production attention to justify the difference in author remuneration.
I've had several technical books published. The process has left me so dissatisfied I'm unlikely to work with a so-called traditional publisher ever again -- certainly not with a standard contract. See The Value of a Publisher for some thoughts representing what you give up to work with such a contract.
I believe that any publisher that refuses to pay you 25% royalties on the wholesale price is not worth your time.