6 ms·

It is nice that he mentioned Light Table. I would not be surprised if Swift ends up really benefitting Clojure adoption indirectly. I think one of the big hangu

by gw 12y ago

It is nice that he mentioned Light Table. I would not be surprised if Swift ends up really benefitting Clojure adoption indirectly. I think one of the big hangups for newcomers is that, if you don't have experience with a Lisp, it's often difficult to understand the benefits of interactive development. If a large amount of new programmers become exposed to it, they'll be more open to other options that provide similar or better interactivity.

ibdknox 12y ago
The only thing that makes me sad is that because of their focus on secrecy, they're doomed to relearn all that we learned along the way. Having played with the swift playground stuff, just an hour long conversation could've made a big difference. Such is the way of Apple though.
- ovis 12y ago
  I would be very interested to hear what you think might have benefitted from the experiences of LightTable. Are they things you think could evolve convergently in the future, or are they mutually exclusive design choices?
- pohl 12y ago
  That's also a good thing. They'll have learned more from making experimental mistakes than from having been told something.
- cpr 12y ago
  Well, don't hesitate to give them feedback now. They're staying it's still quite in a fluid state and will change before release. Plus I'm sure they'd always be grateful for feedback, and if it improves the lot of millions of developers (they say now 9M registered Apple devs), you've done the world a great favor. “There is no limit to what a man can do so long as he does not care a straw who gets the credit for it.” Charles Edward Montague
- bsdetector 12y ago
  On the other hand by keeping it secret and internal they avoided making bad decisions due to defensiveness and stubbornness. An example of this is Go; they don't actually take input from outsiders (generics) and when they have had to give in to outside ideas they implement it poorly just to be different (exceptions as panic). They would rather people put in a no-op printf to avoid the unused package error rather than letting it be a warning because of their own self-righteousness. Apple's secretive process may not be ideal, but there are far worse processes out there.
  - cespare 12y ago
    > On the other hand by keeping it secret and internal they avoided making bad decisions due to defensiveness and stubbornness. That is a non-sequitur. The rest of your comment appears to be low-grade trolling.
  - enneff 12y ago
    Wow, I'm not sure how to begin responding to that. Go has taken a lot of suggestions from the open source community. Check the Go 1 mailing list discussions for examples. But panic existed before the opensource release and is not a substitute for exceptions, and nobody has actually proposed a viable generics implementation. The unused variable/package thing is a fundamental to the project's goals of working well at scale, and tools like goimports alleviate the pain (sometimes the answer is tools, not language changes). Overall I'm pretty dismayed by your characterisation of Go as an open source project. We have a lot of great contributors from outside Google and your ignorant comments do them a great disservice.
    - bsdetector 12y ago
      I realize I'm in the Lion's Den. I feel that if you stepped outside the Google employee and Hacker News bubble you would see Go stubbornly refusing to have modern features, as I do. I've been programming for 30 years and have written code that is in every Linux distro, and yes I did read the Go mailing list occasionally in the early days. But if it makes you feel better to call my opinions ignorant then I hope that's working out for you.
- mcintyre1994 12y ago
  When he said they hope to redefine how computer science is taught, that's a great goal (and to an outsider Swift looks great) but have they spoke to an academic about that idea? If they take the Microsoft approach of being nice to schools that'd be awesome - much better to teach kids Swift than how to use Office. So if they're talking about schools, there's probably some private ones that would be interested but it'd take a crazy effort to crack teaching in schools on a wide scale. Universities, not a chance. It might be great but it's not going to be an academic's choice of teaching language however good it might be for that.
  - gress 12y ago
    A load of colleges (e.g. Stanford) already teach iOS development. Swift and playgrounds will automatically chance how this is taught.
    - mseri 12y ago
      And with those nice closures, generics, monadic patterns, ... They even have a nice way to introduce useful abstract concepts at lower level classes (not that they couldn't make it without... It's just funnier and with higher incentive if you can build ios apps out of it)
  - pjmlp 12y ago
    Don't generalize, there are lots of good universities out there.
- shurcooL 12y ago
  > The only thing that makes me sad is that because of their focus on secrecy, they're doomed to relearn all that we learned along the way. But you're not focused on secrecy, so you can publish the lessons you've learned along the way publicly. That way, everyone, including Apple, can skip some of the mistakes. Wouldn't that be the case?
- sjwright 12y ago
  > because of their focus on secrecy, they're doomed to relearn all that we learned That's wrong, condescending and misses the point. Wrong because the developers were not cut off from learning what was learned by others. Condescending in the way it implies that there's a royal "we" of people that should be consulted whenever any programming language is conceived. And it misses the point because some types of learning are learned better by making the mistake yourself rather than accepting the wisdom of authorities. Anyway, I don't see how Swift's life cycle is any different from other most other languages, except that its inevitable celebrity has been tempered by its parents' protectiveness. The number of people expressing interest in nurturing a language in its formative stages grows in proportion to its viability. Swift went from experiment to viable as soon as it got chosen as the path forward from ObjC. > just an hour long conversation could've made a big difference Apple has huge numbers of programmers on staff, including many compiler hackers, many kernel hackers, many with practical experience building and maintaining APIs and developer tools. They employ literally thousands of programmers who will be end users of Swift. To suggest that your opinion is more valid than theirs is hubris.
  - ibdknox 12y ago
    Hm. I said it was sad, not wrong. Let's say you take direct inspiration for something you're working on and you know someone's been there and thought a lot about it. Wouldn't it make sense to ask them about it? It's certainly true that some mistakes are better learned by making them yourself, but a whole lot aren't. Hell, half the time it's just stuff you're too close to see anymore. The Swift playground is strikingly similar to many of the things we've done in Light Table. It's wonderful that Apple is taking that and running with it and I want these things to end up out there and make things better for devs. But I could've helped them skip some of the crap along the way and I've worked with other large organizations to help them do exactly that. > To suggest that your opinion is more valid than theirs is hubris. My opinion is no more valid, but given that their work looks fairly like our own, I certainly have the benefit of past experience. This isn't about hubris. I have a unique perspective in this particular case, one that no one else will have, as the creator of one of the things they were "heavily influenced" by. I'd rather they took advantage of that so that they can continue to push things even further and not fall into some traps that we did at Microsoft and with LT itself. In any case, I'm sorry I seem to have offended you. My goal is not self aggrandizement, it's just to help do my part in making things better for us all.
superzamp 12y ago
> if you don't have experience with a Lisp, it's often difficult to understand the benefits of interactive development Also with Javascript and browsers' built in consoles you can have good times developing interactively
- pjmlp 12y ago
  Yes, although I am not a big fan of web apps, modern browser developer tools brings me back a familiar feel of Lisp and Smalltalk development experience.

by gw 12y ago

ibdknox 12y ago

The only thing that makes me sad is that because of their focus on secrecy, they're doomed to relearn all that we learned along the way. Having played with the swift playground stuff, just an hour long conversation could've made a big difference. Such is the way of Apple though.

ovis 12y ago
I would be very interested to hear what you think might have benefitted from the experiences of LightTable. Are they things you think could evolve convergently in the future, or are they mutually exclusive design choices?
pohl 12y ago
That's also a good thing. They'll have learned more from making experimental mistakes than from having been told something.
cpr 12y ago
Well, don't hesitate to give them feedback now. They're staying it's still quite in a fluid state and will change before release. Plus I'm sure they'd always be grateful for feedback, and if it improves the lot of millions of developers (they say now 9M registered Apple devs), you've done the world a great favor. “There is no limit to what a man can do so long as he does not care a straw who gets the credit for it.” Charles Edward Montague
bsdetector 12y ago
On the other hand by keeping it secret and internal they avoided making bad decisions due to defensiveness and stubbornness. An example of this is Go; they don't actually take input from outsiders (generics) and when they have had to give in to outside ideas they implement it poorly just to be different (exceptions as panic). They would rather people put in a no-op printf to avoid the unused package error rather than letting it be a warning because of their own self-righteousness. Apple's secretive process may not be ideal, but there are far worse processes out there.
- cespare 12y ago
  > On the other hand by keeping it secret and internal they avoided making bad decisions due to defensiveness and stubbornness. That is a non-sequitur. The rest of your comment appears to be low-grade trolling.
- enneff 12y ago
  Wow, I'm not sure how to begin responding to that. Go has taken a lot of suggestions from the open source community. Check the Go 1 mailing list discussions for examples. But panic existed before the opensource release and is not a substitute for exceptions, and nobody has actually proposed a viable generics implementation. The unused variable/package thing is a fundamental to the project's goals of working well at scale, and tools like goimports alleviate the pain (sometimes the answer is tools, not language changes). Overall I'm pretty dismayed by your characterisation of Go as an open source project. We have a lot of great contributors from outside Google and your ignorant comments do them a great disservice.
  - bsdetector 12y ago
    I realize I'm in the Lion's Den. I feel that if you stepped outside the Google employee and Hacker News bubble you would see Go stubbornly refusing to have modern features, as I do. I've been programming for 30 years and have written code that is in every Linux distro, and yes I did read the Go mailing list occasionally in the early days. But if it makes you feel better to call my opinions ignorant then I hope that's working out for you.
mcintyre1994 12y ago
When he said they hope to redefine how computer science is taught, that's a great goal (and to an outsider Swift looks great) but have they spoke to an academic about that idea? If they take the Microsoft approach of being nice to schools that'd be awesome - much better to teach kids Swift than how to use Office. So if they're talking about schools, there's probably some private ones that would be interested but it'd take a crazy effort to crack teaching in schools on a wide scale. Universities, not a chance. It might be great but it's not going to be an academic's choice of teaching language however good it might be for that.
- gress 12y ago
  A load of colleges (e.g. Stanford) already teach iOS development. Swift and playgrounds will automatically chance how this is taught.
  - mseri 12y ago
    And with those nice closures, generics, monadic patterns, ... They even have a nice way to introduce useful abstract concepts at lower level classes (not that they couldn't make it without... It's just funnier and with higher incentive if you can build ios apps out of it)
- pjmlp 12y ago
  Don't generalize, there are lots of good universities out there.
shurcooL 12y ago
> The only thing that makes me sad is that because of their focus on secrecy, they're doomed to relearn all that we learned along the way. But you're not focused on secrecy, so you can publish the lessons you've learned along the way publicly. That way, everyone, including Apple, can skip some of the mistakes. Wouldn't that be the case?
sjwright 12y ago
> because of their focus on secrecy, they're doomed to relearn all that we learned That's wrong, condescending and misses the point. Wrong because the developers were not cut off from learning what was learned by others. Condescending in the way it implies that there's a royal "we" of people that should be consulted whenever any programming language is conceived. And it misses the point because some types of learning are learned better by making the mistake yourself rather than accepting the wisdom of authorities. Anyway, I don't see how Swift's life cycle is any different from other most other languages, except that its inevitable celebrity has been tempered by its parents' protectiveness. The number of people expressing interest in nurturing a language in its formative stages grows in proportion to its viability. Swift went from experiment to viable as soon as it got chosen as the path forward from ObjC. > just an hour long conversation could've made a big difference Apple has huge numbers of programmers on staff, including many compiler hackers, many kernel hackers, many with practical experience building and maintaining APIs and developer tools. They employ literally thousands of programmers who will be end users of Swift. To suggest that your opinion is more valid than theirs is hubris.
- ibdknox 12y ago
  Hm. I said it was sad, not wrong. Let's say you take direct inspiration for something you're working on and you know someone's been there and thought a lot about it. Wouldn't it make sense to ask them about it? It's certainly true that some mistakes are better learned by making them yourself, but a whole lot aren't. Hell, half the time it's just stuff you're too close to see anymore. The Swift playground is strikingly similar to many of the things we've done in Light Table. It's wonderful that Apple is taking that and running with it and I want these things to end up out there and make things better for devs. But I could've helped them skip some of the crap along the way and I've worked with other large organizations to help them do exactly that. > To suggest that your opinion is more valid than theirs is hubris. My opinion is no more valid, but given that their work looks fairly like our own, I certainly have the benefit of past experience. This isn't about hubris. I have a unique perspective in this particular case, one that no one else will have, as the creator of one of the things they were "heavily influenced" by. I'd rather they took advantage of that so that they can continue to push things even further and not fall into some traps that we did at Microsoft and with LT itself. In any case, I'm sorry I seem to have offended you. My goal is not self aggrandizement, it's just to help do my part in making things better for us all.

superzamp 12y ago

> if you don't have experience with a Lisp, it's often difficult to understand the benefits of interactive development Also with Javascript and browsers' built in consoles you can have good times developing interactively

pjmlp 12y ago
Yes, although I am not a big fan of web apps, modern browser developer tools brings me back a familiar feel of Lisp and Smalltalk development experience.