>
>>> Aleksej Saushev
wrote:>>>> Elizabeth D Rather
forth.com> writes:>>>
>>>>> As others in this thread have said, the two keys are:
>>>>>
>>>>> 1. Comments! Every word needs a comment (the conventional place
>>>>> is just after the name) describing what the word expects on the
>>>>> stack and what it leaves there.
>>>
>>>> This is a very serious drawback actually.
>>>
>>>> Consider math. Do you think that it isn't obvious, what, say,
>>>
>>>> (-b + sqrt(b^2 - 4ac))/(2a)
>>>
>>>> expects? Or it results in?
>>>
>>> No, it's not obvious. If you want one of the roots of a quadratic,
>>> say so. And, as I've pointed out several times before, quadratics
>>> should never be solved this way. <7944nb$lfq$1@
korai.cygnus.co.uk>
>
>> Thus you've shown that you understand perfectly, that the
>> expression relates to quadratic equations, and this doesn't
>> involve any comments on my side. This refutes original
>> statement, that "every word needs a comment."
>
> Erm, no, it doesn't. I understood it; that doesn't make it obvious.
>
>> That this isn't proper way to solve quadratic equation is quite
>> another issue, which has nothing to do with the original statement.
>
> If I saw this I would expect a comment -- at length -- to explain why
> the correct way to solve a quadratic was not used!
This doesn't come from requirement of having a comment per word,
you could read it in preambule, e.g. "we calculate here
effective concentration using quadratic approximation per A. K.
Petrov (13), only positive branch (highest root) is meaningful."
This kind of information is generally enough in both articles
and high school books.
--
CE3OH...
