4 ms·
I was thinking about Python3/Perl6 as he was talking and glad when he brought it up. It's important to note, the world has transitioned to Python3. Django remo
by djsumdog 8y ago
I was thinking about Python3/Perl6 as he was talking and glad when he brought it up.
It's important to note, the world has transitioned to Python3. Django removed the py2 comparability imports a while ago and the Python3 Wall of Superpowers stopped updating a while back when it was over 90% green.
For most devs, all new Python code that's green field is Py3. There is still a lot of py2, yes .. a lot in commercial/closed apps that have embed python in there, along with scientific applications. But I think that will change eventually too.
Perl6 is the opposite end of the spectrum. Massive Perl codebases will always be Perl5. I don't see any enterprise people going to Perl6. If you're like Booking.com and still using Perl for everything, you code is probably so decrepit, old and lacking meaningful testing you should just burn your entire company to the ground so your devs are no longer in miserable agony.
He has a point with Python3: it is a large community. Even with that large community, and even with how small many of the syntax changes were, it took a long time to get to py3! Scala is nowhere near as big.
Scala needs to move carefully with Scala3, so that Scala3 is more like Python3 and doesn't end up like Perl6.
- sheepmullet 8y ago> and still using Perl for everything, you code is probably so decrepit, old and lacking meaningful testing you should just burn your entire company to the ground What makes you think that? The perl community pioneered a lot of the current best practices around testing. A lot of the old perl code bases are still around because there is no compelling reason to move away from perl.
- lizmat 8y ago> If you're like Booking.com and still using Perl for everything, you code is probably so decrepit, old and lacking meaningful testing you should just burn your entire company to the ground so your devs are no longer in miserable agony. That's a lot of assumptions: * Booking.com still using Perl for everything * Code being decrepit * Code lacking meaningful testing * Developers in agony Do you have some inside information to know that all of these are true? You also don't seem to understand that companies are in it for the money, not to provide nice new development projects to make developers happy. If you are programming for a living, you should know that programs are written to make money. By the way, torturing the developers to make a better product, is a well known meme among Perl 6 core developers. > doesn't end up like Perl6. Perl 6 is ending up quite nicely, thank you.
- pas 8y agoPython3 is/was hard, because it forced some problems to the forefront, that were under the mat with Py2. There was no automatic upgrade path, because what was a silent undefined behaviour in Py2 became an unhandled edge case in Py3. And a lot of very pragmatic people just wanted nothing to do with this, after all, their Py2 program worked for them, so why are now they forced to make it work again in some hypothetical scenarios they don't really care about? Any maybe that's something Py3 could have handled with a py2 interop (at module load time you set assumptions that lift your py2 code into py3 with those assumptions). But python is not that big on these hacks, it instead just adds 10 more incomplete libraries into its std and calls it progress!