develooper Front page | perl.perl5.porters | Postings from March 2021

Re: on changing perl's behavior

Thread Previous | Thread Next
Christian Walde
March 30, 2021 09:52
Re: on changing perl's behavior
Message ID:
On Tue, 30 Mar 2021 07:00:37 +0200, Dan Book <> wrote:

> Thank you for writing this out. It elaborates the set of considerations that has led me to my conclusions upthread >and elsewhere.
> On Mon, Mar 29, 2021 at 11:08 PM Ricardo Signes <> wrote:
>> 👉 A big question:  Does this mean that the benefit to people writing new code is contingent on future changes >>to the defaults being much, much safer than (say) turning on strict?  Or just as valuable?
> I think this highlights a future-facing problem with the idea of changing defaults as a design, which I have touched >on in other posts. Say we decide it is worth the risk to have strict and warnings by default, through whatever >timeframe or mechanism. That means Perl code written for that version has to be interpreted differently by both >the interpreter and things which are not the interpreter (linting tools, documentation, stack overflow, the humans >reading it). Then we find that something else is "as safe" and "as important" to change in a similar process. Now we >have three versions of the language, three interpreters that operate noticeably differently, and three ways that all of >these things have to read Perl code. This situation (even before we get to three) is in fact a detriment in addition to >a benefit to new users of the language, as almost none of them will be learning exclusively from a clean room of new >code examples. Given the alternative design that avoids  
> these problems and makes it clear to all consumers >(machine and otherwise) what is going on, for me there is no choice at all.

I highly agree with this email. All possible plans hurt newbies, just different plans hurt different demographics of newbies.

And i want to highlight it with this:

On Tue, 30 Mar 2021 05:07:37 +0200, Ricardo Signes <> wrote:

> People writing new code get benefits from writing "no boilerplate" code. They just start writing, they don't need to >remember the correct invocation, and the documentation shipped with their Perl is written for the common >case being "you are using default Perl, so we never need to show any governing pragmas."

I'm pretty sure most people learning Perl do not do so by going `man perl`, but by googling perl tutorial and pasting the first code examples they find on the internet.

Changing Perl defaults may benefit the newbies who learn via `man perl`, but it breaks the process for newbies learning via Google.

With regards,
Christian Walde
Thread Previous | Thread Next Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at | Group listing | About