[Vm-dev] Pony for the Pharo VM

Robert robert.withers at pm.me
Thu Apr 9 10:21:07 UTC 2020

On 4/9/20 5:59 AM, Ronie Salgado wrote:

> The point of this is that the language safety net is not enough, and in fact it could actually be dangerous because it can instill a false sense of safeness.

I totally agree! This is precisely the foundation of language politics in the Smalltalk world, if I may attempt to speak for everyone! ;) The necessity for proven program safety are sufficient unit tests. There is no other way to detect the errors that matter other than making test assertions. As you said "the language safety net is not enough", whereas robust unit testing can prove the program works correctly. This means that early-binding type systems are an onerous requirement and a complete waste of time as far as ensuring program safety. As such the readability of the code is enhanced through proper naming, not the draconian type specifications and proofings. This late binding feature, really unique to Smalltalk, in combination with a live image makes for a completely different design experience. I specify design as that is where serious errors are primarily introduced, domain errors. Smalltalk empowers the designer to address those errors. Perhaps others could share what about late-binding is a positive for them, in Squeak?

Getting a #doesNotUnderstand: exception is sufficient to detect a type protocol issue.

Another option for you is perhaps you can come up with a Pony bytecode encoder set and run Pony on the Squeak Cog Spur vm...

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.squeakfoundation.org/pipermail/vm-dev/attachments/20200409/e25fa5ca/attachment.html>

More information about the Vm-dev mailing list