Re: [MacRuby-devel] OS X10.9 & MacRuby's future...
Francis, I know nothing about RubyMotion but if I understand correctly it uses a compiler and not an interpreter. So I doubt if it can implement Metaprogramming and Reflection. If it does not , then it is not a Ruby . It might be an excellent language but not Ruby. To the best of my understanding Ruby is a very well defined language supported by a large community and works in what seems to me ( I know Ruby for no more than a year) in a very orderly way. David On May 19, 2013, at 3:52 AM, Francis Chong wrote: > David > > Yes? I got an impression it's just matz implementation. BTW, do RubyMotion > even run ruby spec? > — > Sent from Mailbox for iPhone > > > On Sun, May 19, 2013 at 5:04 AM, david kramf wrote: > > Francis, > Ruby is very well defined language with a well defined standard. > David > > On May 17, 2013, at 3:31 PM, Francis Chong wrote: > >> @david depends on your definition on full ruby. I would say standard library >> is part is full ruby, where RubyMotion deliberately remove part of them >> >> @stephen thanks for the update, I should have tested that myself >> — >> Sent from Mailbox for iPhone >> >> >> On Fri, May 17, 2013 at 8:26 PM, stephen horne wrote: >> >> From what I understand, the only thing missing in Rubymotion is eval() >> >> There's an article by Clay Allsop about meta-programming in Rubymotion at >> http://clayallsopp.com/posts/rubymotion-metaprogramming/ >> >> I tested to see if eval() works in desktop Rubymotion apps (I read somewhere >> that the reason it's not included is due to Apple restrictions on run-time >> code evaluation in iOS, rather than a limit of Rubymotion), but it doesn't. >> >> fb >> >>> >>> david kramf 17/05/2013 13:19 >>> >>> Is RubyMotion a full Ruby. Does it support reflection and metaprograming? >>> Thanks, David Kramf >>> >>> >>> >>> ___ >>> MacRuby-devel mailing list >>> [email protected] >>> https://lists.macosforge.org/mailman/listinfo/macruby-devel >>> >>> Francis Chong 17/05/2013 12:15 >>> While I'm really happy about OS X support on RubyMotion, it is not a >>> replacement for MacRuby. >>> >>> IMHO MacRuby is far superior: >>> >>> It offer JIT compiler, you develop orders of magnitude faster as you dont >>> need clean and rebuild every time. >>> >>> You have full ruby compatibility, load standard library as you wish. >>> >>> It loads gems and framework dynamically like what you would expected from >>> regular ruby. >>> >>> You don't have to write new gems, or rewrite them. Many gems just work, >>> even native ones could work. >>> >>> You can use regular technique for meta programming, and generally you don't >>> enter a uncanny valley between dynamic language and static build system. >>> >>> Some of these limitations are inherited from RubyMotion due to iOS >>> restriction, I don't see them going away anytime soon. >>> >>> That said, RubyMotion team is the ones who know most of MacRuby, and their >>> direction is not like MacRuby in past. If you are going to develop Mac app, >>> your best choice is probably go RubyMotion, or just use Objective-C. >>> — >>> Sent from Mailbox for iPhone >>> >>> >>> >>> ___ >>> MacRuby-devel mailing list >>> [email protected] >>> https://lists.macosforge.org/mailman/listinfo/macruby-devel >> >> ___ >> MacRuby-devel mailing list >> [email protected] >> https://lists.macosforge.org/mailman/listinfo/macruby-devel > > > ___ > MacRuby-devel mailing list > [email protected] > https://lists.macosforge.org/mailman/listinfo/macruby-devel ___ MacRuby-devel mailing list [email protected] https://lists.macosforge.org/mailman/listinfo/macruby-devel
Re: [MacRuby-devel] OS X10.9 & MacRuby's future...
Hey David, Francis is right, MRI is leading the way, although there was an ISO standard released last year. The alternative implementations started the ruby specs, years before that so they could get their rubies to work as drop in replacement for MRI. There are other languages with variants that implement only a subset of the syntax for special purposes. To say RubyMotion is not Ruby because it has a limitation sounds weird to me, almost religiously puristic. I love ruby because it focuses on human usage and Laurent has pushed the limit what we can do with ruby on OSX and iOS. Think of what it can do not what it cannot, it's enabling technology and they are celebrating their first anniversary. best ben On 19 May 2013, at 10:04, david kramf wrote: > Francis, > I know nothing about RubyMotion but if I understand correctly it uses a > compiler and not an interpreter. So I doubt if it can implement > Metaprogramming and Reflection. If it does not , then it is not a Ruby . It > might be an excellent language but not Ruby. > To the best of my understanding Ruby is a very well defined language > supported by a large community and works in what seems to me ( I know Ruby > for no more than a year) in a very orderly way. > David > On May 19, 2013, at 3:52 AM, Francis Chong wrote: > >> David >> >> Yes? I got an impression it's just matz implementation. BTW, do RubyMotion >> even run ruby spec? >> — >> Sent from Mailbox for iPhone >> >> >> On Sun, May 19, 2013 at 5:04 AM, david kramf wrote: >> >> Francis, >> Ruby is very well defined language with a well defined standard. >> David >> >> On May 17, 2013, at 3:31 PM, Francis Chong wrote: >> >>> @david depends on your definition on full ruby. I would say standard >>> library is part is full ruby, where RubyMotion deliberately remove part of >>> them >>> >>> @stephen thanks for the update, I should have tested that myself >>> — >>> Sent from Mailbox for iPhone >>> >>> >>> On Fri, May 17, 2013 at 8:26 PM, stephen horne wrote: >>> >>> From what I understand, the only thing missing in Rubymotion is eval() >>> >>> There's an article by Clay Allsop about meta-programming in Rubymotion at >>> http://clayallsopp.com/posts/rubymotion-metaprogramming/ >>> >>> I tested to see if eval() works in desktop Rubymotion apps (I read >>> somewhere that the reason it's not included is due to Apple restrictions on >>> run-time code evaluation in iOS, rather than a limit of Rubymotion), but it >>> doesn't. >>> >>> fb >>> david kramf17/05/2013 13:19 Is RubyMotion a full Ruby. Does it support reflection and metaprograming? Thanks, David Kramf ___ MacRuby-devel mailing list [email protected] https://lists.macosforge.org/mailman/listinfo/macruby-devel Francis Chong 17/05/2013 12:15 While I'm really happy about OS X support on RubyMotion, it is not a replacement for MacRuby. IMHO MacRuby is far superior: It offer JIT compiler, you develop orders of magnitude faster as you dont need clean and rebuild every time. You have full ruby compatibility, load standard library as you wish. It loads gems and framework dynamically like what you would expected from regular ruby. You don't have to write new gems, or rewrite them. Many gems just work, even native ones could work. You can use regular technique for meta programming, and generally you don't enter a uncanny valley between dynamic language and static build system. Some of these limitations are inherited from RubyMotion due to iOS restriction, I don't see them going away anytime soon. That said, RubyMotion team is the ones who know most of MacRuby, and their direction is not like MacRuby in past. If you are going to develop Mac app, your best choice is probably go RubyMotion, or just use Objective-C. — Sent from Mailbox for iPhone ___ MacRuby-devel mailing list [email protected] https://lists.macosforge.org/mailman/listinfo/macruby-devel >>> >>> ___ >>> MacRuby-devel mailing list >>> [email protected] >>> https://lists.macosforge.org/mailman/listinfo/macruby-devel >> >> >> ___ >> MacRuby-devel mailing list >> [email protected] >> https://lists.macosforge.org/mailman/listinfo/macruby-devel > > ___ > MacRuby-devel mailing list > [email protected] > https://lists.macosforge.org/mailman/listinfo/macruby-devel ___ MacRuby-devel mailing list [email protected] htt
Re: [MacRuby-devel] OS X10.9 & MacRuby's future...
Hi Ben, I am writing an OS X project that relies on Metaprogramming and Reflection. Does not seem like RubMotion is an option for me. All the best, David On May 16, 2013, at 10:05 PM, Carolyn Ann Grant wrote: > Thanks, Mark! > > Yeah, I know the price is more than reasonable, Mark, it's just that right > now, we're not in a position to afford much of anything. Without getting too > personal, we're still digging out from the Great Recession, which hit my > family pretty hard. (As they say in DC, "mistakes were made", and I seem to > have gone out of my way to make sure they were doozies!) I agree that HipByte > is likely to work toward their own success; I'll definitely be looking at > them when I can. > > I think at this point, I have to stick to Objective-C, as much as I really > don't want to. Ruby is just so much better! As for why, I need to have > confidence that I'm not investing a large amount of time and effort into > something that I'll have to abandon when OS X 10.9 comes out. I've chased > more than a few promising technologies, only to see them wither on the vine, > so to speak. I've made such a habit of it, that I was beginning to think that > if I was interested in something, it was likely on its way out! At this > point, I simply can't afford to do that again. So while I'm not delighted to > be writing code in Obj-C, at least I know it's going to be around for a few > years. And I don't have to try and figure out what I did wrong with bridge > support files, etc. > > I am disappointed, and I do wish I had the time and knowledge to further > MacRuby, but I have to prioritize what gets my attention and what I'd like to > do but can't. > > Thank you, all! :-) > > /Carolyn > > On May 16, 2013, at 2:38 PM, Mark Villacampa wrote: > >> I'm a longtime RubyMotion user, and MacRuby user before that. I want to >> share my view as to what is the current status of MacRuby and what can >> happen in the future. >> >> The momentum around MacRuby has been inexistent for almost a year and a >> half. That is, since Laurent Sansonetti (the original creator of MacRuby) >> left Apple, and that left the project without maintainers who were being >> paid to work on it. Only Watson and a couple other maintainers have been >> doing maintenance work and fixing a couple of bugs. >> >> Since nobody is being paid to maintain it, and (AFAIK) there is no >> company/individual whose main/critical systems depended on MacRuby, nobody >> has taken over the project. This is pretty much a chicken-egg situation. >> >> That said, a year ago, Laurent launched RubyMotion, a product based on >> MacRuby which introduces many new features, such as an ARC based memory >> model, and iOS support (dropping OSX support). Just a few days ago, in the >> first anniversary of RubyMotion, they introduced OSX support. >> >> Rubymotion is not open source, and the license costs 200$, plus an annual >> renewal fee of 99$. Two reasons that people sometimes argue for not >> investing in RM are: >> >> - "It's closed source, it might disappear at any moment": Actually, >> RubyMotion is probably more likely to stay in the long term than MacRuby was >> at the beginning. Despite Apple being a huge company, MacRuby was kind of an >> experiment that they could kill at any moment. For HipByte (the company >> behind Rubymotion), Rubymotion is its main product and the one that pays its >> employees. They are way more interested in watching RM succeed than Apple >> was in watching MacRuby succeed. >> >> - "It's too expensive": for playing around or releasing a pet project or >> free app that is not one of your ways of income, that might be the case. >> However, for a company or individual that wants to develop a product from >> which they hope to get some revenue, that price is ridiculous. I've seen PHP >> libraries for creating web forms more expensive than RubyMotion (nothing >> against those libraries). We're talking about a static compiler and a whole >> toolchain for developing iOS apps. If you're a student and want to play >> around with RubyMotion, there is a student discount available (send them an >> email for more information). >> >> So my conclusion is: If you want to develop OSX applications and you liked >> MacRuby, invest in getting a RubyMotion license, you probably won't be >> disappointed. >> >> Mark. >> >> On Thursday, May 16, 2013 at 8:01 PM, Christopher S Martin wrote: >> >>> They recently added support for OS X to rubymotion: >>> http://blog.rubymotion.com/post/49943751398/rubymotion-goes-2-0-and-gets-os-x-support-templates >>> That said, since rubymotion is (I believe) based off of macruby with some >>> additions specifically around static compilation of apps, I don't know if >>> the issues around GC/ARC would be any better in rubymotion on OS X, as I've >>> only used it for iOS. >>> >>> >>> On Thu, May 16, 2013 at 10:56 AM, Jeff Dyck wrote: Just wanted to ad
Re: [MacRuby-devel] OS X10.9 & MacRuby's future...
@ben thanks, this is very clear @david you can do reflection and metaprogramming in RubyMotion, but if you porting code from regular ruby (like those use missing standard library like singleton, delegate, or those missing API like eval string, method_define string) , it might need some big change — Sent from Mailbox for iPhone On Sun, May 19, 2013 at 6:44 PM, david kramf wrote: > Hi Ben, > I am writing an OS X project that relies on Metaprogramming and Reflection. > Does not seem like RubMotion is an option for me. > All the best, > David > On May 16, 2013, at 10:05 PM, Carolyn Ann Grant wrote: >> Thanks, Mark! >> >> Yeah, I know the price is more than reasonable, Mark, it's just that right >> now, we're not in a position to afford much of anything. Without getting too >> personal, we're still digging out from the Great Recession, which hit my >> family pretty hard. (As they say in DC, "mistakes were made", and I seem to >> have gone out of my way to make sure they were doozies!) I agree that >> HipByte is likely to work toward their own success; I'll definitely be >> looking at them when I can. >> >> I think at this point, I have to stick to Objective-C, as much as I really >> don't want to. Ruby is just so much better! As for why, I need to have >> confidence that I'm not investing a large amount of time and effort into >> something that I'll have to abandon when OS X 10.9 comes out. I've chased >> more than a few promising technologies, only to see them wither on the vine, >> so to speak. I've made such a habit of it, that I was beginning to think >> that if I was interested in something, it was likely on its way out! At this >> point, I simply can't afford to do that again. So while I'm not delighted to >> be writing code in Obj-C, at least I know it's going to be around for a few >> years. And I don't have to try and figure out what I did wrong with bridge >> support files, etc. >> >> I am disappointed, and I do wish I had the time and knowledge to further >> MacRuby, but I have to prioritize what gets my attention and what I'd like >> to do but can't. >> >> Thank you, all! :-) >> >> /Carolyn >> >> On May 16, 2013, at 2:38 PM, Mark Villacampa wrote: >> >>> I'm a longtime RubyMotion user, and MacRuby user before that. I want to >>> share my view as to what is the current status of MacRuby and what can >>> happen in the future. >>> >>> The momentum around MacRuby has been inexistent for almost a year and a >>> half. That is, since Laurent Sansonetti (the original creator of MacRuby) >>> left Apple, and that left the project without maintainers who were being >>> paid to work on it. Only Watson and a couple other maintainers have been >>> doing maintenance work and fixing a couple of bugs. >>> >>> Since nobody is being paid to maintain it, and (AFAIK) there is no >>> company/individual whose main/critical systems depended on MacRuby, nobody >>> has taken over the project. This is pretty much a chicken-egg situation. >>> >>> That said, a year ago, Laurent launched RubyMotion, a product based on >>> MacRuby which introduces many new features, such as an ARC based memory >>> model, and iOS support (dropping OSX support). Just a few days ago, in the >>> first anniversary of RubyMotion, they introduced OSX support. >>> >>> Rubymotion is not open source, and the license costs 200$, plus an annual >>> renewal fee of 99$. Two reasons that people sometimes argue for not >>> investing in RM are: >>> >>> - "It's closed source, it might disappear at any moment": Actually, >>> RubyMotion is probably more likely to stay in the long term than MacRuby >>> was at the beginning. Despite Apple being a huge company, MacRuby was kind >>> of an experiment that they could kill at any moment. For HipByte (the >>> company behind Rubymotion), Rubymotion is its main product and the one that >>> pays its employees. They are way more interested in watching RM succeed >>> than Apple was in watching MacRuby succeed. >>> >>> - "It's too expensive": for playing around or releasing a pet project or >>> free app that is not one of your ways of income, that might be the case. >>> However, for a company or individual that wants to develop a product from >>> which they hope to get some revenue, that price is ridiculous. I've seen >>> PHP libraries for creating web forms more expensive than RubyMotion >>> (nothing against those libraries). We're talking about a static compiler >>> and a whole toolchain for developing iOS apps. If you're a student and want >>> to play around with RubyMotion, there is a student discount available (send >>> them an email for more information). >>> >>> So my conclusion is: If you want to develop OSX applications and you liked >>> MacRuby, invest in getting a RubyMotion license, you probably won't be >>> disappointed. >>> >>> Mark. >>> >>> On Thursday, May 16, 2013 at 8:01 PM, Christopher S Martin wrote: >>> They recently added suppo
Re: [MacRuby-devel] OS X10.9 & MacRuby's future...
Just because RubyMotion is compiled doesn't mean it can't have metaprogramming and reflection abilities. These features are not orthogonal to each other. It is true that most compiled languages DON'T have these features, but objective-c definitely DOES. For some light reading, check out (if you're curious) the Objective-C runtime functions. It's easy to see how RubyMotion can take advantages of that system. On the topic of "if it doesn't implement feature ‘X’ then its not Ruby": There is not just one implementation of Ruby. The standard library and syntax are defined by the MRI implementation, but JRuby, IronRuby, Maglev and Rubinius all introduce different takes (and then there's mruby, but it's not fair to compare that one). There is an oft-quoted Ruby-ism that applies here: "If it looks like Ruby, and walks like Ruby, it's Ruby" ;-) #colinta ___ MacRuby-devel mailing list [email protected] https://lists.macosforge.org/mailman/listinfo/macruby-devel
Re: [MacRuby-devel] OS X10.9 & MacRuby's future...
Simple question tho: how well is Rubymotion running RubySpecs? On May 19, 2013 4:18 PM, "Colin Thomas Arnold Gray" wrote: > Just because RubyMotion is compiled doesn't mean it can't have > metaprogramming and reflection abilities. These features are not orthogonal > to each other. It is true that most compiled languages DON'T have these > features, but objective-c definitely DOES. > > For some light reading, check out (if you're curious) the Objective-C > runtime functions. It's easy to see how RubyMotion can take advantages of > that system. > > On the topic of "if it doesn't implement feature ‘X’ then its not Ruby": > There is not just one implementation of Ruby. The standard library and > syntax are defined by the MRI implementation, but JRuby, IronRuby, Maglev > and Rubinius all introduce different takes (and then there's mruby, but > it's not fair to compare that one). > > There is an oft-quoted Ruby-ism that applies here: "If it looks like Ruby, > and walks like Ruby, it's Ruby" ;-) > > #colinta > > ___ > MacRuby-devel mailing list > [email protected] > https://lists.macosforge.org/mailman/listinfo/macruby-devel > ___ MacRuby-devel mailing list [email protected] https://lists.macosforge.org/mailman/listinfo/macruby-devel
