...I *can* defend an argument that the proposal must
take a stance on the keyword proposed. The existence of people (me,
as an example :-)) whose vote would be changed by the choice of a
keyword implies that not taking a stance leaves those people with no
basis for making their decision.
As I see it, there are two camps to whom we need to be "fair" in this:
1) Those who would reject the syntax if it used an ugly word, and
2) Those who would've preferred Guido to pick an alternative keyword
rather than reject the whole proposal, when it's technically trivial to
change.
Group #2 is why I was taking a "soft stand" on the keyword issue (and
I'm it's leading advocate). In addition, when I wrote the first draft
there wasn't much of a consensus over acceptable keyword(s)--there is
more of one now for "using". But honestly, I can't find a heuristic for
selecting one group to disappoint over the other.
So I'm going to pay attention to pressure from another quarter to
resolve this. I left a lot of things open from the beginning, in order
to let Guido have more freedom. He's explained to me that that isn't a
good idea. So I'm making the proposal fit the existing implementation
more closely. This means the following changes:
1) I took out the phrase in the intro about "separating concerns", and
"partial change is better than none". Guido's a smart cookie, and can
decide to separate them if he so chooses.
2) I struck the paragraph at the end of Argument III (keyword choice)
completely, where it talked about the keyword choice being "less
important", an "independent choice" and "easy to change".
3) I changed "recommend keywords" to "recommend a keyword" throughout.
This matches the phrasing of the other arguments, where we "recommend a
suite", for example.
As I said, these are due to Guido's leanings, not anyone else's. But
they tie in nicely; the upshot for Group #1 is that the stance on
keyword choice is now much more hard-line.
Paul, does that settle your issue with the keywords?
Robert Brewer
MIS
Amor Ministries
fu******@amor.org
