Re: sqlite3

[Tomas Hlavaty]

On Tue 14 Dec 2021 at 20:53, Eli Zaretskii <eliz@gnu.org> wrote:
>> From: Tomas Hlavaty <tom@logand.com>
> That is unfair and even hostile.  I don't think you have any reason to
> judge us so harshly.

I am sorry if it sounded harsh.

I certainly value your work on Emacs a lot.

Lars' work too, as I was impressed how quickly he understood and
implemented a suggestion proposed some time ago.

>> I find that really puzzling.
>> That is why I ask if my understanding is correct.
>
> If the above is your understanding, it is definitely incorrect.

Thanks for clarification.

It is valuable experience to see reasoning in such a high profile
project like Emacs.

In this case of sqlite, it seemed that good arguments raised by others
and suggested pure lisp implementation were dismissed without good
reason.  It seems to me that e.g. Qiantan raised better arguments.

I guess it must be hard to develop with so many devils advocates :-)

>> If not, I would expect Emacs to use the proposed superior pure lisp solution
>> and removing the sqlite dependency completely.
>
> No solution is superior in all use cases.  This is software
> engineering, after all.

I was addressing the discussed use-cases, not all possible use-cases.