GnuPG 2.4.4 still using legacy packets?

[Loup Vaillant]

>> To me, and I believe most people, "obsolete" does *not* mean "we can 
>> drop support right now".  It just means it is time to deprecate it.  
>> The way that happens for wire formats (at least those who have a 
>> version marker) is always the same:
> 
>  From Merriam-Webster's online dictionary:
> 
> "obsolete: no longer in use or no longer useful."

The second part, "or no longer useful", is what I had in mind.
Applied to a wire format, the obvious consequence is "it is time to 
deprecate it".  Because of course we're not gonna break compatibility 
overnight.

---

While we're listing dictionary definitions, "obsolete" can also be a 
transitive verb.  Word reference says:

""" to make obsolete by replacing with something newer or better;
""" antiquate:
""" > Automation has obsoleted many factory workers.

That's how RFC 9580 used the word:

""" [...] the Legacy packet format was obsoleted by [RFC2440] in 1998.

- Considering the definition of "obsolete" as a transitive verb,
- considering RFC 2440 introduced the new packet format 27 years ago,
- considering that the new packet format dominates the old,

RFC 9580 is correct: RFC 2440 made the old format antique, and enabled 
other implementations to support the new, which is the first step for 
deprecation.

---

> Yes, if something is obsolete it is to be dropped. You are confusing 
> obsolete with obsolescent:
> 
> "obsolescent: going out of use : becoming obsolete"
> 
> If it's obsolescent, deprecate it. If it's obsolete, get rid of it.

That interpretation of "obsolete" is too strict in my opinion.  It would 
mean that if I cannot reasonably get rid of the old format, then it is 
not obsolete.

And as a matter of fact, readers can't get rid of it, because GnuPG 
still outputs it.  So under that interpretation, Werner just has to say 
"nope, not gonna change", and just like that the redundant, discouraged, 
old format is declared "not obsolete".

That would be a weird power to have.

Loup.