On 7-7-2026 18:28, Rowan Tommins [IMSoP] wrote:
On 7 July 2026 16:40:00 BST, "Gina P. Banyard" <[email protected]> wrote:
There are no guidelines on what constitutes a "impact assessment".
Especially as qualitative assessments or the cost of keeping it in the language
seem to not count.
I deliberately did not use the words "impact assessment". I said the proposals did not
include "any analysis or discussion of the impact on existing code".
I'm not talking about proposals where the discussion of impact doesn't meet
some particular threshold. I'm talking about proposals where there's not even a
single sentence discussing it. No attempt to think about who it might affect.
Not even a wild guess.
I do not think we should be removing features without thinking about who will
be affected. I'm honestly shocked that that's a controversial opinion
My offer to assist with creating some insights into the potential impact
of the proposed deprecations still stands.
I believe that by strategically using the PHPCompatibility library [1],
creating this type of insight should be fairly straight-forward for ~25
out of the current 36 proposals.
The only reason I haven't done so yet, is that it's pretty
time-consuming to run PHPCompatibility over large data sets and there
are more things on my to-do list than just this.
Side-note: Gina - I do see and appreciate the (hidden) costs of keeping
certain features in the language. But costs goes both ways (cost of
deprecating something across the ecosystem versus the costs of continued
maintenance of the feature in the language).
Smile,
Juliette
1: https://github.com/PHPCompatibility/PHPCompatibility
