Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:118793
Content-Type: text/plain;
	charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 13.4 \(3608.120.23.2.7\))
In-Reply-To: <CA+p1Fabc4bNK1XLy66D5Yc8CRCKf1LqD5jZNFtxhO5ecs0w+5Q@mail.gmail.com>
Date: Tue, 11 Oct 2022 01:09:03 -0400
Cc: David Rodrigues <david.proweb@gmail.com>,
 PHP Internals <internals@lists.php.net>
Content-Transfer-Encoding: quoted-printable
Message-ID: <5EBAAEBD-4F9E-4832-BBDD-6E67D8758490@gmail.com>
References: <CAEsg9X3LGYQiF6PGgfhY9s=4iD5v0u2scFRyekNtJfcKDMijtQ@mail.gmail.com>
 <CA+p1FaZqJKFqHdRVqP9xqpHRJoJpKEXbBeo6qTtoeXhZs8oUkQ@mail.gmail.com>
 <CAEsg9X1AztFeF_AvN57wc+CqfAefzsfmy3OG8ZmeRqvk-fcR3w@mail.gmail.com>
 <CA+p1Fabc4bNK1XLy66D5Yc8CRCKf1LqD5jZNFtxhO5ecs0w+5Q@mail.gmail.com>
To: David Gebler <davidgebler@gmail.com>
Subject: Re: [PHP-DEV] Experimental features
From: mike@newclarity.net (MKS Archive)



> On Oct 10, 2022, at 8:04 PM, David Gebler <davidgebler@gmail.com> =
wrote:
>=20
> This is what's bothering me. Either these "experimental" features have
> passed RFC and will be part of the language, as you've said above, or
> they're actually experimental i.e. not finalized and implementing =
behaviour
> and syntax which may be subject to change or even dropped entirely
> depending on the feedback cycle.=20

If they are not experimental and their usage is finalized, then they are =
in-fact approved and they would be absolutely no different from the =
existing RFCs today.

You cannot have experimental features unless those features have the =
potential to not be included in the future language and/or those =
features have the potential to be improved before they are finalized. =
Otherwise, what's the point?

BTW, I googled several different phrases trying to find a horror story =
in Node.js because of their experimental features, but the only thing I =
was able to find was people glowing about the features that are =
currently experimental and/or the excitement when a feature graduated to =
fully supported, for example:

https://nodesource.com/blog/experimental-features-in-node.js/

Maybe you can find examples of downsides of having experimental features =
in Node.js (or some other language/platform) that I was unable to find? =
If so, it would be instructive to learn from their experiences.

> Using the trivial json_validate() example (although as others have =
said, a
> simple new function is probably a bad example),=20

(As an aside, I disagree that a new function would be a bad example. I =
think it would be a good use-case for experimental features because of =
how decoupled it would be from the rest of the language, and it would =
help to see if a lot of people use the function before committing to =
including it.)

> If it's the former, the library can just check if PHP >=3D 8.3 and =
fall back
> to userland implementation / polyfill for earlier versions. This is =
just
> what we have now, where approved new features are part of a versioned
> release.

Not all potentially useful functions can reasonably be implemented in =
userland PHP.

> If it's the latter, everyone touching the experimental version, =
directly or
> indirectly, may have to rip out and change code later which renders =
the
> work they've put in to use the experimental feature in their code =
pointless.

That is not a hard software engineering problem. Software developers =
have been isolating 3rd party dependencies for decades. There's no =
reason why developer could not isolate (most) potential experimental =
features in the same way.

> I'm inclined to suggest if you opt in to an experimental language =
feature,
> you are by definition making at least an implicit declaration that you =
know
> what you're doing,

You speak of people "knowing what they are doing" as if it were a binary =
thing, i.e. they know what they are doing or they do not. Let me use an =
analogy for you here to try to make the counter point.

There are a handful of doctors of the world's leading heart surgeons.  =
They absolutely "know what they are doing."  But if I have brain cancer =
and need a tumor excised, I would not let any one of those doctors touch =
me unless they were also celebrated as a brain surgeon. I am guessing =
you would feel the same.

Similarly, someone can be an excellent software developer but not have =
strong DevOps or sysadmin skills. In my 10+ years working with PHP I =
met, knew, and worked with *many* people like that. And for many years, =
*I* was one of those people.

> so the idea people using these features wouldn't have
> the skills to spin up a container or install something on a VM doesn't
> convince me, personally.

Not every developer is a polymath. Some are excellent at writing =
software but cannot configure a server or a VM or Docker container to =
save their life.  In my LinkedIn I could probably give you a list of 100 =
developers what that would apply to. =20

IOW, they are a lot of excellent heart surgeons out there but they just =
don't know how to fix a brain.

-Mike