J2 paper 0.2.1

[Robert Brewer]

Shalabh Chaturvedi wrote:
>     3. Nitpick. Since you are selling to GvR, it might be 
> nice to spell 
> his name correctly in the code examples :) Use lowercase 'v' 
> in van when 
> writing the full name.

Thanks!

> III Choosing a keyword:
> 
>     1. "Reads correctly" but only if you know that 'using' 
> preceeds the definition.

By "reads correctly" I mean it's rough English. Given the following from
PEP 8, I think I'll treat the run-on case as one we can ignore:

"Separate top-level function and class definitions with two blank lines.
Method definitions inside a class are separated by a single blank line."

>     2. "Should be easy to search for, in both docs and Google."
>        Hmm, 'Python using' is more likely to show irrelevant results 
> than, say, 'Python predef'.

True. I should probably say something about priorities. The Lang Ref
should index the word "using", which fixes half the problem.

>     3. Because of above I'd favour the 'predef' keyword. It 
> has almost 
> all the advantages you list for 'using', and a few more.

I expect to add another paragraph or two based on alternate keywords
which people support (vocally *and* technically). I would say all of the
"*def" candidates could work. In the interest of generating more content
for that paragraph ;) here's the devil's advocate view of predef:

1) It's not a word. But then "def" isn't either.
2) If it is a word, it's a verb. Dependent suite keywords tend to be
adverbial. Perhaps there's a more adverbial word which means, "set up
these items for later processing". Unfortunately for English, this form
of "predef" would be "predefining". Some synonym of "prepare", perhaps?

I'll think about it some more.

> I don't mind the @ syntax at all, but I'll consider signing 
> the proposal 
> as an alternative since I now think it to be at least as 
> good. Perhaps 
> only if you change the keyword <wink>.

Great! Thanks for your comments.


Robert Brewer
MIS
Amor Ministries
fumanchu at amor.org