Re: Production Release - was Re: Questions for Survey about Perl
On 01/05/2011 02:51 PM, Gabor Szabo wrote: Let me just give a probably totally irrelevant comment here. I think most of the open source projects have been in use by many people in production environment before the project had a production release. I guess there are still places that think Linux is not good for their production environment. Probably it is true for all the projects Pm mentioned but a lot of others as well. I remember I was using svn from v0.32 or so. In most technologies I am a very late early adopter. I believe Rakudo and Perl 6 will see a gradual increase in use as they improve, get faster, have more modules etc. It will probably happen a long time before any official 1.0 release will be seen. (if ever) It is very frustrating that the progress is so slow and I can't yet use it for my daily work. It would make both my programming life and my marketing life a lot easier if I could use Rakudo at my clients. But can I seriously complain about the slow progress? Have I made a lot (or any) effort to help Rakudo? I wish I had some time contributing to the effort. Gabor http://szabgab.com/ Maybe we should focus on porting Perl 5 modules on hackathons around the events and blog about the process. Not that I did any serious shot at Perl 6 :-! Regards Racke -- LinuXia Systems = http://www.linuxia.de/ Expert Interchange Consulting and System Administration ICDEVGROUP = http://www.icdevgroup.org/ Interchange Development Team
Re: Production Release - was Re: Questions for Survey about Perl
I would be very interested to see something that allowed Rakudo to talk to Fortran 95. I am going to use Fortran 95 for my thesis work, and maybe I could write a module to give Rakudo a basic array language. Nothing fancy like MATLAB, NumPy or PDL, but enough to try out algorithms and prototype ideas. As it is, I'll probably use PDL or NumPy for that purpose. Daniel. On Thu, Jan 6, 2011 at 2:39 PM, Stefan Hornburg (Racke) ra...@linuxia.de wrote: On 01/05/2011 02:51 PM, Gabor Szabo wrote: Let me just give a probably totally irrelevant comment here. I think most of the open source projects have been in use by many people in production environment before the project had a production release. I guess there are still places that think Linux is not good for their production environment. Probably it is true for all the projects Pm mentioned but a lot of others as well. I remember I was using svn from v0.32 or so. In most technologies I am a very late early adopter. I believe Rakudo and Perl 6 will see a gradual increase in use as they improve, get faster, have more modules etc. It will probably happen a long time before any official 1.0 release will be seen. (if ever) It is very frustrating that the progress is so slow and I can't yet use it for my daily work. It would make both my programming life and my marketing life a lot easier if I could use Rakudo at my clients. But can I seriously complain about the slow progress? Have I made a lot (or any) effort to help Rakudo? I wish I had some time contributing to the effort. Gabor http://szabgab.com/ Maybe we should focus on porting Perl 5 modules on hackathons around the events and blog about the process. Not that I did any serious shot at Perl 6 :-! Regards Racke -- LinuXia Systems = http://www.linuxia.de/ Expert Interchange Consulting and System Administration ICDEVGROUP = http://www.icdevgroup.org/ Interchange Development Team -- No trees were destroyed in the generation of this email. However, a large number of electrons were severely inconvenienced.
Re: Production Release - was Re: Questions for Survey about Perl
On Thu, 2011-06-01 at 14:53 +0100, Daniel Carrera wrote: I would be very interested to see something that allowed Rakudo to talk to Fortran 95. I am going to use Fortran 95 for my thesis work, and maybe I could write a module to give Rakudo a basic array language. Nothing fancy Is there anything like this for perl5 ? In 2001/2 or so someone asked me to convert their perl implementation of a published algorithm to C. Took two hours to do the prototype from the journal article and the run-time went from 24 hours to 5 minutes. The algorithm was the ruelle-takens algorithm (ca 1979, iirc) to compute the fractal dimension of a series. Application was bioinformatics and the journal was a political science one. Very weird mix. Never had a chance to get back to it but I was thinking that an array module for perl5 would be useful. I probably still have the code stashed somewhere. like MATLAB, NumPy or PDL, but enough to try out algorithms and prototype ideas. As it is, I'll probably use PDL or NumPy for that purpose. -- --gh
Re: Production Release - was Re: Questions for Survey about Perl
On Thu, Jan 6, 2011 at 3:32 PM, Guy Hulbert gwhulb...@eol.ca wrote: On Thu, 2011-06-01 at 14:53 +0100, Daniel Carrera wrote: I would be very interested to see something that allowed Rakudo to talk to Fortran 95. I am going to use Fortran 95 for my thesis work, and maybe I could write a module to give Rakudo a basic array language. Nothing fancy Is there anything like this for perl5 ? Yes, PDL. That's the Perl Data Language. And NumPy is the same thing for Python. The algorithm was the ruelle-takens algorithm (ca 1979, iirc) to compute the fractal dimension of a series. Application was bioinformatics and the journal was a political science one. Very weird mix. Never had a chance to get back to it but I was thinking that an array module for perl5 would be useful. I probably still have the code stashed somewhere. If the algorithm can be expressed largely as array operations, then PDL should give a speed more in the ballpark of the C version. Daniel. -- No trees were destroyed in the generation of this email. However, a large number of electrons were severely inconvenienced.
Re: Production Release - was Re: Questions for Survey about Perl
Wendell Hatcher wendell_hatc...@comcast.net writes: My point is make it a production release so peeps can push it to the powers that be in the corporate world. Valid point. Will http://packages.debian.org/experimental/rakudo be continued? This has been the longest production build in test in the history of mankind. If this was a real world project it would have been dead sometime ago. Don't worry too much. Python 3000 took about 8 years. (Though not sure about betas for testing.) Kind regards, Steffen -- Steffen Schwigon s...@renormalist.net Dresden Perl Mongers http://dresden-pm.org/
Re: Production Release - was Re: Questions for Survey about Perl
Daniel Carrera dcarr...@gmail.com writes: If they are critics to begin with, the size of the test suite will not impress them. They could just as well conclude that Perl 6 must have a million corner cases and gotchas that have to be tested. I have never seen a language review that I thought was worth reading that made a point out of the number of tests. The relevance of testing has changed over the last decade, so it is by itself quite a new trend, relative to programming in general. Today “test coverage” actually became a very strong argument for software. It become trendy on “normal” software, spilled over to formerly “untestable” software like Web applications and is even getting momentum on other difficult areas like Operating Systems. So I think the test suite is a strong Pro matching the zeitgeist. Kind regards, Steffen -- Steffen Schwigon s...@renormalist.net Dresden Perl Mongers http://dresden-pm.org/
Re: Production Release - was Re: Questions for Survey about Perl
Stefan Hornburg (Racke) ra...@linuxia.de writes: Maybe we should focus on porting Perl 5 modules With the current size of CPAN this is IMHO not the way to go. A Perl5 embedding interface is more promising. Pugs had that in a not perfect but usable state. Not sure about Rakudo. An embedded Perl5 in Rakudo would even legitimate a special handling that does not need to be generic in the usual “all foreign-language vs. all Perl6-compilers” standard, because it's about Perl-on-Perl. Once I could easily access CPAN modules I would immediately start using Perl 6 for the daily routine work. The language has everything I need, I just can't hack all the things I regularly use from CPAN. Kind regards, Steffen -- Steffen Schwigon s...@renormalist.net Dresden Perl Mongers http://dresden-pm.org/
Re: Production Release - was Re: Questions for Survey about Perl
Although everything you said is technically true, I must point out that without a definitive release, potential users will tend to avoid the software. For people not involved in the process (i.e. 99.995% of Perl users) it is impossible to know when the software is good enough for use. You may talk about strange attractors and orbits, but I haven't the faintest clue how big the orbit of either Perl 6 or Rakudo is. Therefore, I cannot recommend it to other people, and I will hesitate to use it on anything that is very important. Daniel. On Wed, Jan 5, 2011 at 12:38 PM, Richard Hainsworth rich...@rusrating.ru wrote: So I'd change that to after a production release of a Perl 6 compiler Out of curiosity (because I think it will illuminate some of the difficulty Rakudo devs have in declaring something to be a production release): - What constitues a production release? - What was the first production release of Perl 4? - What was the first production release of Perl 5? - What was the first production release of Linux? - At what point was each of the above declared a production release; was it concurrent with the release, or some time afterwards? Pm Larry responded to a post of mine asking about when Perl6 would be finished - the post was about the time that Pugs was still being actively developed. He pointed to the difference between the waterfall model and the strange attractor model for software development, perl6 progress being measured using the strange attractor model. Many of the questions and answers about a 'production release' imply the waterfall model. The concept here is that some one 'in authority' sets criteria which define 'finished'. Once the software / language / project fulfils the criteria - the edge of the waterfall - it is 'finished'. This has the advantage that everyone knows when to break out the champaign and have a party. It has the disadvantage that criteria of 'finished' can rarely be written in advance because to do so requires precognition, or knowledge of the future. Is there any sophisticated piece of software that is 'perfect', has no bugs, is easy to use? Was MS Vista 'production' quality? Perl 5.0 was quickly replaced by Perl 5.004 (I think), which include references. The strange attractor model implies a process that is never ending, in that there will always be deviations from the solution 'orbit' or 'path'. However, there comes a time when for most normal purposes, the solution orbit will be so 'narrow' that the blips will be not be noticed for most situations. In this respect, qualitative statements such as 'when developers accept it' or 'providers such as ActiveState etc' bundle it are recognition of the strange attractor measure of progress of Perl6. Personally, I think that we are in sight of acceptance for Rakudo Star. This is an implementation of a subset of Perl6. I also believe that when Rakudo begins to implement Sets, Macros and deals with the problems posed by GUI, we will see further changes in the Perl6 specification. It is unlikely that such changes will 'break' Rakudo *. A question that would be useful to ask is: When will Rakudo Star be useful for some of your purposes? a) It is already useful; b) When running precompiled Rakudo * versions for a test suite of example programs is as fast as running Perl5 versions, on average. c) When running (from human readable text to final result) Rakudo * versions for a test suite of example programs is as fast as Perl5 versions, on average. d) When Rakudo * implements a larger subset of Perl6 and/or access well-written C/C++ libraries efficiently, presupposing (c). Another question would be what should be in the test suite of example programs? The example programs are not the test suite, which verifies consistency with the specification. The example programs should be designed - I suggest - to test speed and memory footprint. Ultimately, programmers are interested in solutions that are quick and use least hardware resources (the human resource of writing a simple and understandable program being the strongest part of Perl6, at least I think so). -- No trees were destroyed in the generation of this email. However, a large number of electrons were severely inconvenienced.
RE: Production Release - was Re: Questions for Survey about Perl
Hear! Hear! -Original Message- From: Daniel Carrera [mailto:dcarr...@gmail.com] Sent: Wednesday, January 05, 2011 7:15 AM To: Richard Hainsworth Cc: perl6-us...@perl.org Subject: Re: Production Release - was Re: Questions for Survey about Perl Although everything you said is technically true, I must point out that without a definitive release, potential users will tend to avoid the software. For people not involved in the process (i.e. 99.995% of Perl users) it is impossible to know when the software is good enough for use. You may talk about strange attractors and orbits, but I haven't the faintest clue how big the orbit of either Perl 6 or Rakudo is. Therefore, I cannot recommend it to other people, and I will hesitate to use it on anything that is very important. Daniel. On Wed, Jan 5, 2011 at 12:38 PM, Richard Hainsworth rich...@rusrating.ru wrote: So I'd change that to after a production release of a Perl 6 compiler Out of curiosity (because I think it will illuminate some of the difficulty Rakudo devs have in declaring something to be a production release): - What constitues a production release? - What was the first production release of Perl 4? - What was the first production release of Perl 5? - What was the first production release of Linux? - At what point was each of the above declared a production release; was it concurrent with the release, or some time afterwards? Pm Larry responded to a post of mine asking about when Perl6 would be finished - the post was about the time that Pugs was still being actively developed. He pointed to the difference between the waterfall model and the strange attractor model for software development, perl6 progress being measured using the strange attractor model. Many of the questions and answers about a 'production release' imply the waterfall model. The concept here is that some one 'in authority' sets criteria which define 'finished'. Once the software / language / project fulfils the criteria - the edge of the waterfall - it is 'finished'. This has the advantage that everyone knows when to break out the champaign and have a party. It has the disadvantage that criteria of 'finished' can rarely be written in advance because to do so requires precognition, or knowledge of the future. Is there any sophisticated piece of software that is 'perfect', has no bugs, is easy to use? Was MS Vista 'production' quality? Perl 5.0 was quickly replaced by Perl 5.004 (I think), which include references. The strange attractor model implies a process that is never ending, in that there will always be deviations from the solution 'orbit' or 'path'. However, there comes a time when for most normal purposes, the solution orbit will be so 'narrow' that the blips will be not be noticed for most situations. In this respect, qualitative statements such as 'when developers accept it' or 'providers such as ActiveState etc' bundle it are recognition of the strange attractor measure of progress of Perl6. Personally, I think that we are in sight of acceptance for Rakudo Star. This is an implementation of a subset of Perl6. I also believe that when Rakudo begins to implement Sets, Macros and deals with the problems posed by GUI, we will see further changes in the Perl6 specification. It is unlikely that such changes will 'break' Rakudo *. A question that would be useful to ask is: When will Rakudo Star be useful for some of your purposes? a) It is already useful; b) When running precompiled Rakudo * versions for a test suite of example programs is as fast as running Perl5 versions, on average. c) When running (from human readable text to final result) Rakudo * versions for a test suite of example programs is as fast as Perl5 versions, on average. d) When Rakudo * implements a larger subset of Perl6 and/or access well-written C/C++ libraries efficiently, presupposing (c). Another question would be what should be in the test suite of example programs? The example programs are not the test suite, which verifies consistency with the specification. The example programs should be designed - I suggest - to test speed and memory footprint. Ultimately, programmers are interested in solutions that are quick and use least hardware resources (the human resource of writing a simple and understandable program being the strongest part of Perl6, at least I think so). -- No trees were destroyed in the generation of this email. However, a large number of electrons were severely inconvenienced. -- This message w/attachments (message) is intended solely for the use of the intended recipient(s) and may contain information that is privileged, confidential or proprietary. If you are not an intended recipient, please notify the sender, and then please delete and destroy all copies and attachments, and be advised that any review
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, Jan 5, 2011 at 2:13 PM, Anderson, Jim wrote: Hear! Hear! Uhmm... sorry if I looked angry or whatever. Email is at times a poor medium of communication because you lose details like tone of voice and body language. I just wanted to highlight something that I think is relevant to anyone who wants to see increased adoption of Perl 6 / Rakudo. Cheers, Daniel. -- No trees were destroyed in the generation of this email. However, a large number of electrons were severely inconvenienced.
Re: Production Release - was Re: Questions for Survey about Perl
There has been requests and talk of a production release for years now. Fancy titles released have come out monthly and quarterly for some time. At some point you have to say it simply isn't a good product or it is going to production how long are we going to hear excuses of my dog died past week and the production release is delayed for a year. Perl 6 at this point seems like a bad dream at best and there really isn't a need since moose and perl 5 have improved. Sent from my iPhone Wendell Hatcher wendell_hatc...@comcast.net 303-520-7554 Blogsite: http://thoughtsofaperlprogrammer.typepad.com/blog On Jan 5, 2011, at 6:13 AM, Anderson, Jim jim.ander...@bankofamerica.com wrote: Hear! Hear! -Original Message- From: Daniel Carrera [mailto:dcarr...@gmail.com] Sent: Wednesday, January 05, 2011 7:15 AM To: Richard Hainsworth Cc: perl6-us...@perl.org Subject: Re: Production Release - was Re: Questions for Survey about Perl Although everything you said is technically true, I must point out that without a definitive release, potential users will tend to avoid the software. For people not involved in the process (i.e. 99.995% of Perl users) it is impossible to know when the software is good enough for use. You may talk about strange attractors and orbits, but I haven't the faintest clue how big the orbit of either Perl 6 or Rakudo is. Therefore, I cannot recommend it to other people, and I will hesitate to use it on anything that is very important. Daniel. On Wed, Jan 5, 2011 at 12:38 PM, Richard Hainsworth rich...@rusrating.ru wrote: So I'd change that to after a production release of a Perl 6 compiler Out of curiosity (because I think it will illuminate some of the difficulty Rakudo devs have in declaring something to be a production release): - What constitues a production release? - What was the first production release of Perl 4? - What was the first production release of Perl 5? - What was the first production release of Linux? - At what point was each of the above declared a production release; was it concurrent with the release, or some time afterwards? Pm Larry responded to a post of mine asking about when Perl6 would be finished - the post was about the time that Pugs was still being actively developed. He pointed to the difference between the waterfall model and the strange attractor model for software development, perl6 progress being measured using the strange attractor model. Many of the questions and answers about a 'production release' imply the waterfall model. The concept here is that some one 'in authority' sets criteria which define 'finished'. Once the software / language / project fulfils the criteria - the edge of the waterfall - it is 'finished'. This has the advantage that everyone knows when to break out the champaign and have a party. It has the disadvantage that criteria of 'finished' can rarely be written in advance because to do so requires precognition, or knowledge of the future. Is there any sophisticated piece of software that is 'perfect', has no bugs, is easy to use? Was MS Vista 'production' quality? Perl 5.0 was quickly replaced by Perl 5.004 (I think), which include references. The strange attractor model implies a process that is never ending, in that there will always be deviations from the solution 'orbit' or 'path'. However, there comes a time when for most normal purposes, the solution orbit will be so 'narrow' that the blips will be not be noticed for most situations. In this respect, qualitative statements such as 'when developers accept it' or 'providers such as ActiveState etc' bundle it are recognition of the strange attractor measure of progress of Perl6. Personally, I think that we are in sight of acceptance for Rakudo Star. This is an implementation of a subset of Perl6. I also believe that when Rakudo begins to implement Sets, Macros and deals with the problems posed by GUI, we will see further changes in the Perl6 specification. It is unlikely that such changes will 'break' Rakudo *. A question that would be useful to ask is: When will Rakudo Star be useful for some of your purposes? a) It is already useful; b) When running precompiled Rakudo * versions for a test suite of example programs is as fast as running Perl5 versions, on average. c) When running (from human readable text to final result) Rakudo * versions for a test suite of example programs is as fast as Perl5 versions, on average. d) When Rakudo * implements a larger subset of Perl6 and/or access well-written C/C++ libraries efficiently, presupposing (c). Another question would be what should be in the test suite of example programs? The example programs are not the test suite, which verifies consistency with the specification. The example programs should be designed - I suggest - to test speed and memory footprint. Ultimately, programmers are interested
Re: Production Release - was Re: Questions for Survey about Perl
It seems you may have concluded something not intended. It is blindingly obvious that the majority of language users, people who do not have the resources (time, skill set, training) to test a language before using it, will only start to use a language when it is recommended by 'those in authority'. I would suggest that the 'popularity' of a language is more a function of how well it is adopted by teachers of Computer Science at universities and colleges. I think the issue of a version number is irrelevant, given the vested interest of the developer to assign a number that will attract users, to such an extent that there is rule of thumb never to use the first release, but to wait until the version 'has matured'. Even if the developers of Rakudo release a V1.0, would that in itself lead to the acceptance of Perl6. I doubt it. A great deal that is needed to demonstrate the stability and strength of Perl6 for 'production' purposes has been included in the design from the very beginning, namely, a MASSIVE test suite. Perl6 also has documentation and specification and even teaching books, even before the language has completely matured. But for Rakudo to be widely adopted as Perl6, it seems to me there have to be stronger criteria than a version number of 1.0, and a test suite that is passed demonstrating adherence to a specification. For my part, I already use Rakudo for nearly all my programming needs - not that they are particularly burdensome or mission critical. The elegance of the language in itself is a powerful reason to use it. I am willing to deal with and work around the problems. Even in this thread higher standards have been alluded to. But what are they? How specifically can they be quantified? Speed, memory, ease of use? I suggest that Gabor's survey is one way of generating more input about what Rakudo has to be in order for it to be considered 'production' quality. Richard On 01/05/11 16:13, Anderson, Jim wrote: Hear! Hear! -Original Message- From: Daniel Carrera [mailto:dcarr...@gmail.com] Sent: Wednesday, January 05, 2011 7:15 AM To: Richard Hainsworth Cc: perl6-us...@perl.org Subject: Re: Production Release - was Re: Questions for Survey about Perl Although everything you said is technically true, I must point out that without a definitive release, potential users will tend to avoid the software. For people not involved in the process (i.e. 99.995% of Perl users) it is impossible to know when the software is good enough for use. You may talk about strange attractors and orbits, but I haven't the faintest clue how big the orbit of either Perl 6 or Rakudo is. Therefore, I cannot recommend it to other people, and I will hesitate to use it on anything that is very important. Daniel. On Wed, Jan 5, 2011 at 12:38 PM, Richard Hainsworth rich...@rusrating.ru wrote: So I'd change that to after a production release of a Perl 6 compiler Out of curiosity (because I think it will illuminate some of the difficulty Rakudo devs have in declaring something to be a production release): - What constitues a production release? - What was the first production release of Perl 4? - What was the first production release of Perl 5? - What was the first production release of Linux? - At what point was each of the above declared a production release; was it concurrent with the release, or some time afterwards? Pm Larry responded to a post of mine asking about when Perl6 would be finished - the post was about the time that Pugs was still being actively developed. He pointed to the difference between the waterfall model and the strange attractor model for software development, perl6 progress being measured using the strange attractor model. Many of the questions and answers about a 'production release' imply the waterfall model. The concept here is that some one 'in authority' sets criteria which define 'finished'. Once the software / language / project fulfils the criteria - the edge of the waterfall - it is 'finished'. This has the advantage that everyone knows when to break out the champaign and have a party. It has the disadvantage that criteria of 'finished' can rarely be written in advance because to do so requires precognition, or knowledge of the future. Is there any sophisticated piece of software that is 'perfect', has no bugs, is easy to use? Was MS Vista 'production' quality? Perl 5.0 was quickly replaced by Perl 5.004 (I think), which include references. The strange attractor model implies a process that is never ending, in that there will always be deviations from the solution 'orbit' or 'path'. However, there comes a time when for most normal purposes, the solution orbit will be so 'narrow' that the blips will be not be noticed for most situations. In this respect, qualitative statements such as 'when developers accept it' or 'providers such as ActiveState etc' bundle it are recognition of the strange attractor measure
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, 2011-05-01 at 19:05 +0300, Richard Hainsworth wrote: It seems you may have concluded something not intended. I was unsurprised at the reaction to your post. [snip] I think the issue of a version number is irrelevant, given the vested Clearly you were wrong. [snip] For my part, I already use Rakudo for nearly all my programming needs - not that they are particularly burdensome or mission critical. The elegance of the language in itself is a powerful reason to use it. I am willing to deal with and work around the problems. I have decided to adopt it for one project. If that is successful, I will switch from 5 to 6. If not, I'll have to consider python or ruby for the next one. Even in this thread higher standards have been alluded to. But what are they? How specifically can they be quantified? Speed, memory, ease of use? [snip] The fact that Rakudo comes with: a) a warning that it is slow b) a list of things which are *not* implemented Is a red flag. Similarly, Moose has warnings about start-up time so I don't use it as most of my perl is command-line scripts. I think it would be useful to freeze rakudo1 as soon as possible but it would be helpful to have some benchmarks so we know *how* slow. I started using perl4 after perl5 was already in use. I stuck with perl4 until I got interested in OO. Rakudo is not listed here: http://shootout.alioth.debian.org/ Fixing that is something I'd like to help with. Note that go was listed *before* it was announced. That tells me that the go authors are, in some small way, more serious about their project succeeding than perl6. -- --gh
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, Jan 5, 2011 at 5:05 PM, Richard Hainsworth rich...@rusrating.ru wrote: It is blindingly obvious that the majority of language users, ..., will only start to use a language when it is recommended by 'those in authority'... I think the issue of a version number is irrelevant 1) You have more or less contradicted yourself. If we agree that Larry Wall is an authority, for example, it is reasonable to wait until he says that the Perl 6 spec is ready, and many will also wait until Rakudo claims to mostly comply with the Perl 6 spec. 2) Version number may not be relevant to you, but it is relevant to others. Therefore, it is relevant to the adoption of Perl 6. , given the vested interest of the developer to assign a number that will attract users, That has not been my experience with FOSS projects. Rather, I think developers shy away from ever saying 1.0. For example, the JED editor has been around for a long time, but its version number is 0.99-19. The Enlightenment window manager too 10 years before they were comfortable saying 1.0. This fear of 1.0 was even the subject of a paragraph in Eric Raymond's The Cathedral and The Bazaar. to such an extent that there is rule of thumb never to use the first release, but to wait until the version 'has matured'. I've heard this in the Windows world, but I think the FOSS world version numbers tend to be lower. For example, I remember that Netscape 5.0 was equivalent to Mozilla 1.0. Even if the developers of Rakudo release a V1.0, would that in itself lead to the acceptance of Perl6. I doubt it. Necessary but not sufficient condition? A great deal that is needed to demonstrate the stability and strength of Perl6 for 'production' purposes has been included in the design from the very beginning, namely, a MASSIVE test suite. How many people, not involved in Perl 6, know that? See the point? I bet that you don't follow the development process of every single software package you use. For any given software package, 99.99% of users do not follow the developers list of look through the test suite. Daniel. -- No trees were destroyed in the generation of this email. However, a large number of electrons were severely inconvenienced.
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, Jan 5, 2011 at 17:30, Guy Hulbert gwhulb...@eol.ca wrote: Rakudo is not listed here: http://shootout.alioth.debian.org/ Fixing that is something I'd like to help with. Note that go was listed *before* it was announced. That tells me that the go authors are, in some small way, more serious about their project succeeding than perl6. So your suggestion to Gabor is to add the question: Do you think that NOT listing Rakudo at shootout.alioth.debian.org means Rakudo is not a serious project? Or did you have some other point? (This is the first time I've seen shootout.alioth.debian.org, I won't claim that it's not a serious shootout just because of that, BTW.) -- Jan
Re: Production Release - was Re: Questions for Survey about Perl
My point is make it a production release so peeps can push it to the powers that be in the corporate world. This has been the longest production build in test in the history of mankind. If this was a real world project it would have been dead sometime ago. Sent from my iPhone Wendell Hatcher wendell_hatc...@comcast.net 303-520-7554 Blogsite: http://thoughtsofaperlprogrammer.typepad.com/blog On Jan 5, 2011, at 9:31 AM, Richard Hainsworth rich...@rusrating.ru wrote: Without the development phenomenon of Perl6, it's difficult to see how Moose and other improvements in perl 5 would have occurred. Despite the frustrations in following the growth of Pugs, then Rakudo, it's been fun, worthwhile and inspiring. A bit like life really. Do you really want it to end? But until it ends, how can you tell what sort of person you are, or what your achievements have been? I love Perl6. Rukudo is great - already. On 01/05/11 17:21, Wendell Hatcher wrote: There has been requests and talk of a production release for years now. Fancy titles released have come out monthly and quarterly for some time. At some point you have to say it simply isn't a good product or it is going to production how long are we going to hear excuses of my dog died past week and the production release is delayed for a year. Perl 6 at this point seems like a bad dream at best and there really isn't a need since moose and perl 5 have improved. Sent from my iPhone Wendell Hatcher wendell_hatc...@comcast.net 303-520-7554 Blogsite: http://thoughtsofaperlprogrammer.typepad.com/blog On Jan 5, 2011, at 6:13 AM, Anderson, Jimjim.ander...@bankofamerica.com wrote: Hear! Hear! -Original Message- From: Daniel Carrera [mailto:dcarr...@gmail.com] Sent: Wednesday, January 05, 2011 7:15 AM To: Richard Hainsworth Cc: perl6-us...@perl.org Subject: Re: Production Release - was Re: Questions for Survey about Perl Although everything you said is technically true, I must point out that without a definitive release, potential users will tend to avoid the software. For people not involved in the process (i.e. 99.995% of Perl users) it is impossible to know when the software is good enough snip
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, 2011-05-01 at 18:02 +0100, Jan Ingvoldstad wrote: On Wed, Jan 5, 2011 at 17:30, Guy Hulbert gwhulb...@eol.ca wrote: Rakudo is not listed here: http://shootout.alioth.debian.org/ Fixing that is something I'd like to help with. Note that go was listed *before* it was announced. That tells me that the go authors are, in some small way, more serious about their project succeeding than perl6. So your suggestion to Gabor is to add the question: No. The subject changed ... Do you think that NOT listing Rakudo at shootout.alioth.debian.org means Rakudo is not a serious project? Or did you have some other point? Marketing. (This is the first time I've seen shootout.alioth.debian.org, I won't claim that it's not a serious shootout just because of that, BTW.) When go was announced a link to 'shootout' was in the announcement. I think I might have seen if before that but, if so, i'd forgotten so it was new to me at the time. What got me interested in perl6 was the april fools announcement about parrot ostensibly by Larry and Guido. Something like 10 years ago. I don't learn new programming languages unless I have something to do with it. I've been looking at what it would take to implement perl6/rakudo versions of the programs on 'shootout', and I think I can do it so I will try to get one or two of them running properly in the benchmarker. The benchmarking program can be downloaded (which I've done) and comes bundled with 2 or 3 python programs, one of which requires python 2.5 and I'm still on python 2.4 (don't ask). However I've figured out how to see the source for example programs, so I'll manually download all the perl5 and C ones and try to get the benchmarker going for those. It'll take me a little while ... -- --gh
Re: Production Release - was Re: Questions for Survey about Perl
On 01/05/11 19:48, Daniel Carrera wrote: On Wed, Jan 5, 2011 at 5:05 PM, Richard Hainsworthrich...@rusrating.ru wrote: It is blindingly obvious that the majority of language users, ..., will only start to use a language when it is recommended by 'those in authority'... I think the issue of a version number is irrelevant 1) You have more or less contradicted yourself. If we agree that Larry Wall is an authority, for example, it is reasonable to wait until he says that the Perl 6 spec is ready, and many will also wait until Rakudo claims to mostly comply with the Perl 6 spec. From what Larry has already said, I dont think he ever will say the Perl 6 spec is ready. The spec and the language are evolving together. That is what the waterfall and attractor stuff was all about. When I said 'in authority', I meant those opinion-makers (from bloggers to journalists to heads of major software developers) who start saying 'xxx is a really cool language'. 2) Version number may not be relevant to you, but it is relevant to others. Therefore, it is relevant to the adoption of Perl 6. And here it seems to me that you begin to prove the point I am trying to make: version numbers are irrelevant as carriers of information about usefulness, stability, or even maturity of product. , given the vested interest of the developer to assign a number that will attract users, That has not been my experience with FOSS projects. Rather, I think developers shy away from ever saying 1.0. For example, the JED editor has been around for a long time, but its version number is 0.99-19. How can 0.99-19 mean anything? Does it mean under 1.0? If so, does this meant that the developers of JED consider it to be unusable or 'not for production purposes'? My entire point is that the version number, in of itself, has no more meaning than what the developers want it to mean. But acceptance is not determined by the developers. The Enlightenment window manager too 10 years before they were comfortable saying 1.0. This fear of 1.0 was even the subject of a paragraph in Eric Raymond's The Cathedral and The Bazaar. to such an extent that there is rule of thumb never to use the first release, but to wait until the version 'has matured'. I've heard this in the Windows world, Though I have been using Linux exclusively for about 5 years now, the Windows world remains an order of magnitude larger. So again, if the point is true in the Windows world, it seems I would win the argument. but I think the FOSS world version numbers tend to be lower. For example, I remember that Netscape 5.0 was equivalent to Mozilla 1.0. Wasnt that due to organisational and ownership changes relating to the development of Netscape? Even if the developers of Rakudo release a V1.0, would that in itself lead to the acceptance of Perl6. I doubt it. Necessary but not sufficient condition? Not even necessary. Why not v0.99-? A great deal that is needed to demonstrate the stability and strength of Perl6 for 'production' purposes has been included in the design from the very beginning, namely, a MASSIVE test suite. How many people, not involved in Perl 6, know that? See the point? I bet that you don't follow the development process of every single software package you use. For any given software package, 99.99% of users do not follow the developers list of look through the test suite. You are again confirming a point I have tried to make. Most people do not themselves try out new languages or indeed anything new until they have read a recommendation from someone they trust. If I want a new camera, I search the internet for reviews - I cant test each one. But once I do settle on a choice, I then want the proof. Just because a reviewer says its good, how do I know he / she isnt paid by the company? The proof that software is stable and robust comes from testing. And testing has been the foundation of the development of Perl6. When - eventually - critics compare Perl6 to some other language and discuss the robustness of the compiler, they will look at the size of the test suites. Richard Daniel.
Re: Production Release - was Re: Questions for Survey about Perl
I have to agree I don't think this is a serious project. In-fact at this point it seems like a bunch of friends working on a hobby in their basement. Sent from my iPhone Wendell Hatcher wendell_hatc...@comcast.net 303-520-7554 Blogsite: http://thoughtsofaperlprogrammer.typepad.com/blog On Jan 5, 2011, at 10:15 AM, Guy Hulbert gwhulb...@eol.ca wrote: On Wed, 2011-05-01 at 18:02 +0100, Jan Ingvoldstad wrote: On Wed, Jan 5, 2011 at 17:30, Guy Hulbert gwhulb...@eol.ca wrote: Rakudo is not listed here: http://shootout.alioth.debian.org/ Fixing that is something I'd like to help with. Note that go was listed *before* it was announced. That tells me that the go authors are, in some small way, more serious about their project succeeding than perl6. So your suggestion to Gabor is to add the question: No. The subject changed ... Do you think that NOT listing Rakudo at shootout.alioth.debian.org means Rakudo is not a serious project? Or did you have some other point? Marketing. (This is the first time I've seen shootout.alioth.debian.org, I won't claim that it's not a serious shootout just because of that, BTW.) When go was announced a link to 'shootout' was in the announcement. I think I might have seen if before that but, if so, i'd forgotten so it was new to me at the time. What got me interested in perl6 was the april fools announcement about parrot ostensibly by Larry and Guido. Something like 10 years ago. I don't learn new programming languages unless I have something to do with it. I've been looking at what it would take to implement perl6/rakudo versions of the programs on 'shootout', and I think I can do it so I will try to get one or two of them running properly in the benchmarker. The benchmarking program can be downloaded (which I've done) and comes bundled with 2 or 3 python programs, one of which requires python 2.5 and I'm still on python 2.4 (don't ask). However I've figured out how to see the source for example programs, so I'll manually download all the perl5 and C ones and try to get the benchmarker going for those. It'll take me a little while ... -- --gh
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, 2011-05-01 at 10:24 -0700, Wendell Hatcher wrote: I have to agree I don't think this is a serious project. In-fact at this point it seems like a bunch of friends working on a hobby in their basement. I'm not sure I said anything to agree with. You seem to misinterpret my intention. [snip] Do you think that NOT listing Rakudo at shootout.alioth.debian.org means Rakudo is not a serious project? Or did you have some other point? Marketing. What I meant was that a serious project pays attention to marketing. The perl6 marketing effort is limited by resources more than go is. [snip] The benchmarking program can be downloaded (which I've done) and comes bundled with 2 or 3 python programs, one of which requires python 2.5 and I'm still on python 2.4 (don't ask). However I've figured out how to see the source for example programs, so I'll manually download all the perl5 and C ones and try to get the benchmarker going for those. Here's what I will attempt to reproduce: http://shootout.alioth.debian.org/u32/benchmark.php?test=alllang=perllang2=gcc I will start by downloading each program in C and perl (there seem to be several C versions -- and sometimes several perl versions available) and just running them appropriately. It'll take me a little while ... I'm fairly busy. I'll report _any_ progress back to the list ... if you don't hear from me by February 1st feel free to nag me. By 'progress', I mean something on github. -- --gh
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, Jan 5, 2011 at 6:22 PM, Richard Hainsworth rich...@rusrating.ru wrote: From what Larry has already said, I dont think he ever will say the Perl 6 spec is ready. The spec and the language are evolving together. That is what the waterfall and attractor stuff was all about. Not relevant. The question is whether people will take Perl 6 seriously without a Perl 6 spec. I think in general no, or at least extremely slowly. Whether Larry Wall is happy with that is for Larry Wall to decide. It's his language. When I said 'in authority', I meant those opinion-makers (from bloggers to journalists to heads of major software developers) who start saying 'xxx is a really cool language'. That's just your personal preference. Other people have other views. Personally I rarely care about what a blog says on these matters, and I would certainly not consider a blogger or a journalist to be some kind of authority. To me an authority has to, at least, know a lot about the subject matter. For example, Rakudo developers. I really don't understand why you would accept the word of a journalist as some sort of authority but don't think that the word of the actual developers, in the form of a release label or version number is relevant. To me that sounds completely backwards. 2) Version number may not be relevant to you, but it is relevant to others. Therefore, it is relevant to the adoption of Perl 6. And here it seems to me that you begin to prove the point I am trying to make: version numbers are irrelevant as carriers of information about usefulness, stability, or even maturity of product. I disagree. I have argued this at length already and explained how version numbers are generally used in the FOSS world. There are only so many times I'm willing to repeat myself. Though I have been using Linux exclusively for about 5 years now, Ok, you are a relatively new user. When - eventually - critics compare Perl6 to some other language and discuss the robustness of the compiler, they will look at the size of the test suites. If they are critics to begin with, the size of the test suite will not impress them. They could just as well conclude that Perl 6 must have a million corner cases and gotchas that have to be tested. I have never seen a language review that I thought was worth reading that made a point out of the number of tests. Daniel. -- No trees were destroyed in the generation of this email. However, a large number of electrons were severely inconvenienced.
Re: Production Release - was Re: Questions for Survey about Perl
'serious project' ??? For some 'serious' people, Perl6 is a 'serious project'. Concepts of 'serious' differ amongst reasonable people. Not a problem if your 'serious' aint my 'serious'. As an aside, it took 358 years to prove Fermat's Last Theorem. Wiles - who proved it - shut himself away for the five years he spent creating the last part of the proof sequence. A number of historical figures have looked at the problem. That to my mind is a 'serious project' and serious people, and Wiles did indeed work on it in a 'basement' as a 'hobby'. It was an obsession and he was afraid of telling people what he was working on. But now we consider him a hero. Rakudo and Perl6 is being developed in the way it is for good and practical reasons. Richard On 01/05/11 20:24, Wendell Hatcher wrote: I have to agree I don't think this is a serious project. In-fact at this point it seems like a bunch of friends working on a hobby in their basement. Sent from my iPhone Wendell Hatcher wendell_hatc...@comcast.net 303-520-7554 Blogsite: http://thoughtsofaperlprogrammer.typepad.com/blog On Jan 5, 2011, at 10:15 AM, Guy Hulbertgwhulb...@eol.ca wrote: On Wed, 2011-05-01 at 18:02 +0100, Jan Ingvoldstad wrote: On Wed, Jan 5, 2011 at 17:30, Guy Hulbertgwhulb...@eol.ca wrote: Rakudo is not listed here: http://shootout.alioth.debian.org/ Fixing that is something I'd like to help with. Note that go was listed *before* it was announced. That tells me that the go authors are, in some small way, more serious about their project succeeding than perl6. So your suggestion to Gabor is to add the question: No. The subject changed ... Do you think that NOT listing Rakudo at shootout.alioth.debian.org means Rakudo is not a serious project? Or did you have some other point? Marketing. (This is the first time I've seen shootout.alioth.debian.org, I won't claim that it's not a serious shootout just because of that, BTW.) When go was announced a link to 'shootout' was in the announcement. I think I might have seen if before that but, if so, i'd forgotten so it was new to me at the time. What got me interested in perl6 was the april fools announcement about parrot ostensibly by Larry and Guido. Something like 10 years ago. I don't learn new programming languages unless I have something to do with it. I've been looking at what it would take to implement perl6/rakudo versions of the programs on 'shootout', and I think I can do it so I will try to get one or two of them running properly in the benchmarker. The benchmarking program can be downloaded (which I've done) and comes bundled with 2 or 3 python programs, one of which requires python 2.5 and I'm still on python 2.4 (don't ask). However I've figured out how to see the source for example programs, so I'll manually download all the perl5 and C ones and try to get the benchmarker going for those. It'll take me a little while ... -- --gh
Re: Production Release - was Re: Questions for Survey about Perl
Guy, Your idea is actually exactly what I was suggesting when I said 'example programs'. I think there are/were perl6 versions for the shootout problems. I am not sure what happened to them. Getting benchmarking will be interesting. Regards, Richard On 01/05/11 20:15, Guy Hulbert wrote: On Wed, 2011-05-01 at 18:02 +0100, Jan Ingvoldstad wrote: On Wed, Jan 5, 2011 at 17:30, Guy Hulbertgwhulb...@eol.ca wrote: Rakudo is not listed here: http://shootout.alioth.debian.org/ Fixing that is something I'd like to help with. Note that go was listed *before* it was announced. That tells me that the go authors are, in some small way, more serious about their project succeeding than perl6. So your suggestion to Gabor is to add the question: No. The subject changed ... Do you think that NOT listing Rakudo at shootout.alioth.debian.org means Rakudo is not a serious project? Or did you have some other point? Marketing. (This is the first time I've seen shootout.alioth.debian.org, I won't claim that it's not a serious shootout just because of that, BTW.) When go was announced a link to 'shootout' was in the announcement. I think I might have seen if before that but, if so, i'd forgotten so it was new to me at the time. What got me interested in perl6 was the april fools announcement about parrot ostensibly by Larry and Guido. Something like 10 years ago. I don't learn new programming languages unless I have something to do with it. I've been looking at what it would take to implement perl6/rakudo versions of the programs on 'shootout', and I think I can do it so I will try to get one or two of them running properly in the benchmarker. The benchmarking program can be downloaded (which I've done) and comes bundled with 2 or 3 python programs, one of which requires python 2.5 and I'm still on python 2.4 (don't ask). However I've figured out how to see the source for example programs, so I'll manually download all the perl5 and C ones and try to get the benchmarker going for those. It'll take me a little while ...
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, 2011-05-01 at 20:51 +0300, Richard Hainsworth wrote: 'serious project' ??? For some 'serious' people, Perl6 is a 'serious project'. Concepts of 'serious' differ amongst reasonable people. Not a problem if your 'serious' aint my 'serious'. For programming languages, there are rankings by number of developers. A Historical Example DateNumber 19791 198016 198138 198285 1983??+2 1984??+50 1985500 19862,000 19874,000 198815,000 198950,000 1990150,000 1991400,000 Taken from the language author's Design and Evolution book. Chapter 7. My wife was sent on a course to learn this language in the early 1990s. So you have about 10 years to get started. -- --gh
Re: Production Release - was Re: Questions for Survey about Perl
On Wed, 2011-05-01 at 21:04 +0300, Richard Hainsworth wrote: Guy, Your idea is actually exactly what I was suggesting when I said 'example programs'. What convinced me that rakudo is worth pursuing was the 3-line dice class with a roll() method. What I do now is 'use fields' and build from templates. I understood fields via 'perldoc -m' in far less time than I've spent reading through Moose docs. I think there are/were perl6 versions for the shootout problems. I am not sure what happened to them. I doubt they were posted on alioth. I think that a pre-requisite is: apt-get install rakudo on ubuntu. Getting benchmarking will be interesting. I hope I have time. I'm planning to compile and run one C and one perl program today and see if the outputs are the same (that's my understanding, so far, of the requirements for alioth). Regards, Richard -- --gh
Re: Production Release - was Re: Questions for Survey about Perl
Let me just give a probably totally irrelevant comment here. I think most of the open source projects have been in use by many people in production environment before the project had a production release. I guess there are still places that think Linux is not good for their production environment. Probably it is true for all the projects Pm mentioned but a lot of others as well. I remember I was using svn from v0.32 or so. In most technologies I am a very late early adopter. I believe Rakudo and Perl 6 will see a gradual increase in use as they improve, get faster, have more modules etc. It will probably happen a long time before any official 1.0 release will be seen. (if ever) It is very frustrating that the progress is so slow and I can't yet use it for my daily work. It would make both my programming life and my marketing life a lot easier if I could use Rakudo at my clients. But can I seriously complain about the slow progress? Have I made a lot (or any) effort to help Rakudo? I wish I had some time contributing to the effort. Gabor http://szabgab.com/