Re: Guidance on the implications of the Lyon Amendment

--000000000000293e40061d581ad6
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Jul 16, 2024 at 4:31=E2=80=AFAM Russ Allbery <[email protected]> wrot=
e:

> Hi folks,
>
> I've started a few conversations in the past year about pushing forward
> the minimum Perl version of podlators, which is a "way upstream" package
> with lots of transitive dependencies.  In the v6.0.0 release, I pushed th=
e
> minimum Perl version forward to 5.12.
>
> This broke a release process for brian d foy, resulting in the following
> issue:
>
>     https://github.com/rra/podlators/issues/35
>
> Some of this is my fault in ways that I can fix: I should have used an
> alpha version (weirdly enough, I have somehow managed to be a core
> maintainer for 25 years without using alpha versions, which is entirely o=
n
> me and I need to start), and I should have gotten announcements into a fe=
w
> more places.
>
> Some of this may be my fault in ways that I'm not willing to fix, but I'm
> not sure.
>
> Some of it makes me wonder if I misunderstood the previous discussions an=
d
> the implications of the Lyon Amendment for the usability of CPAN tools an=
d
> for what prerequisites have to be met before bumping the minimum Perl
> version of way upstream packages to versions of Perl up to 5.16.  I feel
> like I'm hearing different things from different people, all of whom are
> far more plugged in to the community than I am.
>
> If folks would be willing to take a look at that issue, particularly folk=
s
> who were involved in some of these previous discussions, and let me know
> if I have completely misunderstood or am barking up entirely the wrong
> tree, I would greatly appreciate it.
>
> Thank you!
>

It's a maximum minimum version. It allows one to require a newer version,
it doesn't require you to do so.

There's always a cost/benefit ratio to take into account, and IMO for old
modules it rarely makes sense to bump the required version. The benefits
are often dubious, and there's a real cost for downstream users. For newer
modules it often makes more sense.

Leon

--000000000000293e40061d581ad6
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr">On Tue, Jul 16, 2024 at 4:31=E2=80=AFAM R=
uss Allbery &lt;<a href=3D"mailto:[email protected]" target=3D"_blank">eagle@=
eyrie.org</a>&gt; wrote:</div><div class=3D"gmail_quote"><blockquote class=
=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rg=
b(204,204,204);padding-left:1ex">Hi folks,<br>
<br>
I&#39;ve started a few conversations in the past year about pushing forward=
<br>
the minimum Perl version of podlators, which is a &quot;way upstream&quot; =
package<br>
with lots of transitive dependencies.=C2=A0 In the v6.0.0 release, I pushed=
 the<br>
minimum Perl version forward to 5.12.<br>
<br>
This broke a release process for brian d foy, resulting in the following<br=
>
issue:<br>
<br>
=C2=A0 =C2=A0 <a href=3D"https://github.com/rra/podlators/issues/35" rel=3D=
"noreferrer" target=3D"_blank">https://github.com/rra/podlators/issues/35</=
a><br>
<br>
Some of this is my fault in ways that I can fix: I should have used an<br>
alpha version (weirdly enough, I have somehow managed to be a core<br>
maintainer for 25 years without using alpha versions, which is entirely on<=
br>
me and I need to start), and I should have gotten announcements into a few<=
br>
more places.<br>
<br>
Some of this may be my fault in ways that I&#39;m not willing to fix, but I=
&#39;m<br>
not sure.<br>
<br>
Some of it makes me wonder if I misunderstood the previous discussions and<=
br>
the implications of the Lyon Amendment for the usability of CPAN tools and<=
br>
for what prerequisites have to be met before bumping the minimum Perl<br>
version of way upstream packages to versions of Perl up to 5.16.=C2=A0 I fe=
el<br>
like I&#39;m hearing different things from different people, all of whom ar=
e<br>
far more plugged in to the community than I am.<br>
<br>
If folks would be willing to take a look at that issue, particularly folks<=
br>
who were involved in some of these previous discussions, and let me know<br=
>
if I have completely misunderstood or am barking up entirely the wrong<br>
tree, I would greatly appreciate it.<br>
<br>
Thank you!<br></blockquote><div><br></div><div>It&#39;s a maximum minimum v=
ersion. It allows one to require a newer version, it doesn&#39;t require yo=
u to do so.</div><div><br></div><div>There&#39;s always a cost/benefit rati=
o to take into account, and IMO for old modules it rarely makes sense to bu=
mp the required version. The benefits are often dubious, and there&#39;s a =
real cost for downstream users. For newer modules it often makes more sense=
.</div><div><br></div><div>Leon<br></div></div></div>

--000000000000293e40061d581ad6--