Scala 3 versy impressive, It can replace Python i think

I read some document for Scala 3, I like it’s changes. Better than Scala2.
I learn Scala 2 for more than 5 years.

I like Python and happy to see Scala 3 more likely Python.
it like feel i just write python code.

And this is very powerful.

If scala can learn more and more from Python, we can see how things change.

The scientist and AI learning , compute, why not using a fast scala , python is good, but too slow.
Scala 3 fast, static typed, and quite similar to python 3.

whoops. great works.

5 Likes

The narrative so far has been that Scala 3 will not be like Python 3, an upgrade that divides the community.

I think the current estimate is that Scala 3 will succeed in this regard.

2 Likes

Coming from using F# daily in my work, would like to thank all devs making Scala 3.
It seems concise with almost every thing I would want in a language.

I have been experimenting with it for bit more than a month. There are few minor nitpicks such as type inference where it seems compiler should be able to infer it though I think this could be improved later, case class naming instead of just record, modules instead of object, I have seen arguments for and against but its not really big of a deal to me.

I never found such fears particularly credible, hence I find the achievement a pretty low bar. With the exception of macros, it seems that there’s very little of Scala 2.13 that won’t compile in Scala 3.0. I would be interested to know whether it was typically easier to port Scala 2.9 => 3.0, vs 2.13 => 3.0.

Although Scala 2 macros are going away strenuous efforts are being made to offer alternative mechanisms for nearly all Scala 2 Macro capabilities. The fact that Macros were labeled as experimental ultimately proved irrelevant. It could be argued that the Scala 3.0 is kind of misleading and that the real Scala 3.0 was Scala 2.10 and the coming release should be called 3.4.

Personally I think the major / epic release distinction is unhelpful and a relic of an earlier phase in the evolution of software development. I think we should follow the Java pattern of release names or go to an Ubuntu calendar based release naming scheme.

In fact, at the SIP ski retreat and covid party in Italy at the beginning of last April, they decided to call the Dotty release “Scala X”, but that turned out not to be the road map. I think when MacOS rolled over to 11, that scheme lost its appeal.

Eventually, they decided it makes the most sense to call it Scala 3, because of 3-space indentation. Projects stuck on 2-space tabs can use Scala 2. There is still only one correct way to line up the asterisks in Scaladoc comments.

I was sanguine about the community hanging together until they settled on given ... with.

This will wreak havoc on code bases that switch to 3-char indentation, where you expect everything to line up with your vals and defs.

I still hope they trim object to obj, though my hope may be based on the irrational exuberance of the season.

I have to check whether my SIP for that change is in “pending” state, “under review”, or “ignored”.

That SIP also includes trimming extends to ext. If they also bring back optional extends and accept given ... extends as recently proposed, then we’ll be able to write given ext {}. Probably I’d use that syntax exclusively for extension methods.

The SIP process also includes dispositions “dormant” and “rejected”, which are aliases for “deferred until a student implements it for Dotty.”

Now I have to update my SIP for Scala 3, to trim with to mit so that we can be well on our way to 3-char keywords and 3-space indentation.

It’s generally acknowledged by now that anything not expressed in 3 chars is dubious. So it’s === not ==, for not while, ??? not TODO, etc.

Inexplicably, they recently lengthened Not to NotGiven, but I chalk that up to pandemic fatigue.

It may be some years before “scala” is shortened to “ska”. That might be Scala XXX.

In any case, it’s very exciting to start a New Year and try out a new Scala. That feeling may sustain us for some time.

3 Likes

I fear said mod might be considered one step beyond … i.e. Madness.

1 Like

Remarkably, @propensive was onto something

:joy:

MMS Forth proponents have argued so for decades.