New Python implementation
Michael Torrie
torriem at gmail.com
Tue Feb 16 14:51:21 EST 2021
On 2/16/21 10:58 AM, Ben Bacarisse wrote:
> Attempts at a universal compiler stalled in the 1980s (though there may
> have been some new developments since I stopped looking) because
> expressing the semantics of different languages is so very hard. In
> fact, much of the interest in pursuing the idea came from benefits that
> would be derived simply from having a language's semantics formally
> described.
>
> I don't think there is anything to see here. If the author had come up
> with some new ways to tackle any of the problems, he would be telling> people about these, not saying "be patient" (and bad-mouthing CPython).
Indeed, in all seriousness if he is successful, I look forward to
reading his PhD dissertation, because it would be worthy of a PhD,
especially if he made some breakthroughs in metacompiler design. His
comments don't give me hope, though.
Seems a bit paradoxical to me to, on the one hand, express a desire to
build a Python implementation, but on the other hand, mock Python as a
toy language. Why bother with Python at all?
I wish him luck and maybe he'll eventually come back to the community
with something to show and impress with.
More information about the Python-list
mailing list