Fwd: JEPs proposed to target JDK 14: 345, 361, 363, 364, 365, & 367

[Cay Horstmann]

Since I just gave conference presentations urging more participation 
with upcoming Java features, I feel I should comment here and urge 
committers and reviewers to consider their position on JEP 361.

The switch syntax for what is commonly referred to as "match 
expressions" has been designed and implemented as a preview feature. The 
principal characteristic of the design is "orthogonality" of the axes 
expression/statement and disjointness/fallthrough. It is claimed that 
this aids learning, since programmers can reason from their existing 
knowledge about switch statements. However, this orthogonality gives 
rise to a match expression with fallthrough whose utility is 
questionable, and it has been weakened with the introduction of the 
"yield" statement. It is also not clear whether it matches programmer 
expectations--see for example 
https://mail.openjdk.java.net/pipermail/amber-spec-observers/2019-October/001844.html. 


I am unaware of evidence gathered in the preview process that would 
warrant moving the switch syntax out of preview at this point. Have the 
three new forms of switch been adopted in meaningful ways? Have the 
teachability claims been validated by analyzing teaching materials? Are 
other languages successfully arriving at similar decisions? Has it been 
shown in some other way that Java programmers can more successfully 
reason about the four forms of switch, when compared to a non-switch 
syntax for match expressions?

There is a risk in precipitously finalizing what may well become an 
unloved feature, with negative impact for the reputation of both the 
language and the process that drives its evolution. My suggestion is to 
keep the switch syntax in preview until you can share evidence for its 
success.

Cheers,

Cay


-------- Messaggio Inoltrato --------
Oggetto: JEPs proposed to target JDK 14: 345, 361, 363, 364, 365, & 367
Data: Thu, 31 Oct 2019 15:44:04 -0700
Mittente: mark.reinhold at oracle.com
A: jdk-dev at openjdk.java.net

The following JEPs are proposed to target JDK 14:

   345: NUMA-Aware Memory Allocation for G1
        http://openjdk.java.net/jeps/345

   361: Switch Expressions (Standard)
        http://openjdk.java.net/jeps/361

   363: Remove the Concurrent Mark Sweep (CMS) Garbage Collector
        http://openjdk.java.net/jeps/363

   364: ZGC on macOS
        http://openjdk.java.net/jeps/364

   365: ZGC on Windows
        http://openjdk.java.net/jeps/365

   367: Remove the Pack200 Tools and API
        http://openjdk.java.net/jeps/367

Feedback on these proposals from JDK Project Committers and Reviewers [1]
is more than welcome, as are reasoned objections.  If no such objections
are raised by 23:00 UTC on Thursday, 7 November, or if they’re raised
and then satisfactorily answered, then per the JEP 2.0 process proposal
[2] I’ll target these JEPs to JDK 14.

- Mark


[1] https://openjdk.java.net/census#jdk
[2] https://cr.openjdk.java.net/~mr/jep/jep-2.0-02.html