Re: [review][JSON] Some thoughts on recent reviews

From: Peter Dimov (pdimov_at_[hidden])
Date: 2020-09-23 10:05:09

Hans Dembinski wrote:

> You said yourself that Boost.JSON is not as efficient as it could be 
> during the conversion of "my data type" to JSON, because the existing data 
> has to be copied into the json::value first. I am a young member of the 
> Boost family, but my feeling is that this would have been a reason to 
> reject the design in the past.

I don't think so. As a previous reviewer correctly observed, an apple has 
been submitted, and you're complaining that it isn't an orange.

Were it a bad apple, that would have been a reason to reject. If we didn't 
need apples, if users didn't need apples, if the two most popular fruits 
weren't apples, that might have been a reason to reject. Not being an orange 
isn't.

One possible objection (that has been used in the past) is that if we accept 
an apple, nobody will submit an orange anymore. That's where the calendar 
argument comes in. It's 2020, we didn't have an apple, and nobody has 
submitted an orange. Ten years should have been enough time for orange 
proponents. And this has nothing to do with "urgency".