[swift-evolution] [swift-evolution-announce] [Review #2] SE-0117: Default classes to be non-subclassable publicly

[David Waite]

> On Jul 20, 2016, at 12:44 PM, Tino Heth via swift-evolution <swift-evolution at swift.org> wrote:
> 
> 
>> Am 20.07.2016 um 18:20 schrieb L. Mihalkovic <laurent.mihalkovic at gmail.com <mailto:laurent.mihalkovic at gmail.com>>:
>> 
>>> So my advice: Be glad that you don't see such problems in your real work life, and hope that the extremists who would like to completely remove classic object orientation and cripple Swift to fully match their ideals don't prevail ;-)
>> 
>> That ship has sailed... it is now just a matter of the implementation details... My hopes are now on google forking swift like they did to webkit and dalvik. It won't save the apps, bug it would the servers.
> ;-) nay, that's a little bit to pessimistic for me:
> After all, even with SE-0117, you can still write nice code with Swift — it just will be less fun :(
> 
> Afaics, there is constant pressure to turn Swift into a language for fools, rather than a language that helps us avoiding foolish mistakes… but so far, Swift got more things right for me than any of the alternatives.

I think we’ve all had some disappointments in at least some part of the evolution of Swift - it is only natural, as the difference of opinions and perspectives is why this mailing list exists.

I for one was very disappointed at the addition of the new “private” visibility - I didn’t see a real-world need for smaller-than-file scope, it made ‘fileprivate'  be spelled ugly and I felt it was too early to expand visibility features before we had established whether submodules or multi-module frameworks would be evolution targets. But in the words of a former coworker, it hardly seems a hill worth dying on.

In the end, design is all about trade-offs. I still think Swift has a decent and consistent design and design evolution, and I look forward to see what we enable in the Swift 4/5 timeframe.

-DW
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.swift.org/pipermail/swift-evolution/attachments/20160720/7f5d82d2/attachment.html>