PEP 526 - var annotations and the spirit of python

Steven D'Aprano steve+comp.lang.python at pearwood.info
Wed Jul 4 02:45:48 EDT 2018


On Tue, 03 Jul 2018 19:28:43 -0700, Jim Lee wrote:

> On 07/03/18 16:51, Steven D'Aprano wrote:
>>
>> I love watching pedantically precise people panic and dig themselves
>> into a hole. Since I'm an extremely pedantic person myself, I can
>> recognise it in others -- especially when they're not as precisely
>> correct as they think they're being.
>>
>> It was two numbers, not three, and not even close to "vastly"
>> different.
> 
>    Ah, I see we're not going to leave it alone.  In that case,
> "indefinite" is a "number", in that it was a quantity you cited along
> with the other two.

It would be nice if you consider the context of words rather than 
criticising them in isolation.

In isolation, "10,000" and "some vaguely indefinite number between 
roughly one thousand and one hundred thousand, give or take some 
unspecified quantity" *wink* are clearly not the same. Any fool can see 
that.

But in context, my description of the number as being "indefinite" was 
clarifying the meaning of the earlier use of 10,000 as a vague 
quantifier, not the specific number 9999+1. It hardly counts as a 
separate reference to a number when it is explicitly referring back to 
the first mention of the term "10,000" to clarify that it shouldn't be 
read as a specific number.

Which you then treated as a specific number.


>> Both numbers I mentioned (ten thousand hours, a couple of years) are
>> within the bounds of acceptable precision to each other: their
>> (figurative) error bars overlap.
> 
> Who is defining acceptable here? 

You *did* read the article I linked to, right?

Given the factors that they mention -- the wide range in hours practiced, 
the lack of an upper limit, the variation between individuals, the lack 
of any hard dividing line between "expert" and "non-expert" -- what sort 
of precision do you think is acceptable?

    10,000 ± 6 hours ?

If you want to argue for a specific level of precision, I look forward to 
hearing your reasoning. Until then, I'm going to consider "ten thousand 
hours" to be a hand-wavy figure, not much more precise than "a shit-ton 
of hours".


> And you are once again creating a
> strawman argument.  My quibble was *not* with the difference between
> 10,000 hours and 2 years.

Quote:

    So we went from 10,000 hours, to infinite, and then back to
    a couple of years?

followed by:

    I was simply pointing out that you used three vastly
    different numbers in almost the same breath to describe 
    how long it takes a person to master something.

Perhaps I have misunderstood you. If so, apologies.

If you weren't complaining about them being "vastly different", what 
precisely was your complaint?



-- 
Steven D'Aprano
"Ever since I learned about confirmation bias, I've been seeing
it everywhere." -- Jon Ronson



More information about the Python-list mailing list