Call for signatories for J2

[Robert Brewer]

The J2 proposal is as complete as it will ever be.

http://www.aminus.org/rbre/python/pydec.html

The patch is nearly complete; only the __future__ declaration and some
document tweaks remain. It should be updated on SourceForge today or
tomorrow.

This is a call for all who wish to sign the proposal, either for,
against, or abstaining. Please sign by either posting on
comp.lang.python (replying to this is fine), or sending email to
(e-mail address removed) (for some of you reading this, here's a link:
mailto:[email protected]). Please include your full name. If you reject
the proposal, please also include a _short_ description of why you are
rejecting it. You will not receive a reply regarding the rejection or
description; it will be included for Guido's information purposes only.

Please read the proposal carefully before signing. In particular,
carefully read the statement (for, against, or abstaining) to which you
are signing your name; each is explicit and means exactly what it says.

Thanks again to those who have contributed in all forms. Hopefully, we
can put this decorator issue to rest soon and move on.


Robert Brewer
MIS
Amor Ministries
(e-mail address removed)

P.S. I'll be out of town from Friday morning to Sunday evening (U.S.
Pacific Time), so get your name in early. Given sufficient volume, I'll
promote this to Guido just before I leave; otherwise, I'll wait 'til
Sunday night to finish compiling names.

 

[Fernando Perez]

The J2 proposal is as complete as it will ever be.

http://www.aminus.org/rbre/python/pydec.html

The patch is nearly complete; only the __future__ declaration and some
document tweaks remain. It should be updated on SourceForge today or
tomorrow.

This is a call for all who wish to sign the proposal, either for,
against, or abstaining. Please sign by either posting on
comp.lang.python (replying to this is fine)

My signature : FOR

My Name: Fernando Perez. Regular email: fperez AT colorado DOT edu

I am the author of IPython, one of the affected tools by the @proposal; IPython
was referenced in the J2 paper. So I guess this is my 'official' stance on
the matter. I like it for 2 reasons:

a) It will save me a bunch of silly work changing ipython (especially the
docs), and it will save my ipython users having to change their work habits.

b) I just like it; I think the paper authors did a fantastic job justifying the
idea with a set of very coherent reasons. By now it's clear no syntax will be
deemed 'perfect' by everyone, but I found the arguments in the proposal clear,
intelligent and convincing enough to be happy going along.

Thanks to those who were willing to put enough effort to turn a potential
disaster for the language into a rational and productive discussion.

Best,

f

 

[Francesco Bochicchio]

Assuming that simple python users qualify, you can add my name in favor
of your proposal.

If people are using this thread for signing themselves, maybe
it would be better to create subthreads for pro/against/indifferent,
just to make counting more simple.

Ciao

 

[Jamie Radcliffe]

A newbie votes FOR.

Jamie Radcliffe

 

[Michael Sparks]

Robert Brewer wrote:
....

The patch is nearly complete;

Patch against current CVS passes all the tests* and has been uploaded to
SourceForge. I'm currently looking at how __future__ declarations work, and
that will be the next step, but based on useful feedback this looks
relatively simple to do.

* Well, put another way, the only tests that fail are those that fail with
an unmodified CVS tree.

This is a call for all who wish to sign the proposal, either for,
against, or abstaining. Please sign by either posting on
comp.lang.python (replying to this is fine),

Please consider this a vote FOR - I doubt that'll surprise however ;-)

Regards,


Michael.

 

[Nicolas Fleury]

For.

Regards,
Nicolas Fleury

 

[Paul McGuire]

For.

Paul McGuire

 

[Paul Paterson]

For

Paul Paterson

 

[David Pokorny]

Vote For.

David

 

[Colin J. Williams]

For J2, this is better than 2.4.2a version, but there are more Pythonic
ways.

Colin W.

 

[Mahesh Padmanabhan]

The J2 proposal is as complete as it will ever be.

http://www.aminus.org/rbre/python/pydec.html

The patch is nearly complete; only the future declaration and some
document tweaks remain. It should be updated on SourceForge today or
tomorrow.

This is a call for all who wish to sign the proposal, either for,
against, or abstaining. Please sign by either posting on
comp.lang.python (replying to this is fine), or sending email to
(e-mail address removed) (for some of you reading this, here's a link:
mailto:[email protected]). Please include your full name. If you reject
the proposal, please also include a short description of why you are
rejecting it. You will not receive a reply regarding the rejection or
description; it will be included for Guido's information purposes only.

Please read the proposal carefully before signing. In particular,
carefully read the statement (for, against, or abstaining) to which you
are signing your name; each is explicit and means exactly what it says.

Thanks again to those who have contributed in all forms. Hopefully, we
can put this decorator issue to rest soon and move on.


Robert Brewer
MIS
Amor Ministries
(e-mail address removed)

P.S. I'll be out of town from Friday morning to Sunday evening (U.S.
Pacific Time), so get your name in early. Given sufficient volume, I'll
promote this to Guido just before I leave; otherwise, I'll wait 'til
Sunday night to finish compiling names.

I vote for J2.

 

[Michele Simionato]

Uhm... the choice between @ and J2 is tough for me. "@" is somewhat
less intrusive (I automatically ignore it, look at the function body
and then go back at the decorator) whereas J2 focus too much the
attention on the using clause. This is probably because it reminds
me a try: except: block, where the attention is focused on the try:
part. It is probably matter of time to get used to it.

OTOH I think that J2 is more consistent with the rest of Python and
it avoids the introduction of the pie sign which conflicts with Leo
and IPython, so there are strong arguments in its favor.

Conclusion: I would be happy both with "@" and J2 (or even with
"|" instead of "@"). However, the work done by Robert Brewer and
Micheal Sparks is impressive and deserve to be supported, so
count me as a mild "FOR"


Michele Simionato

 

[Alex Martelli]

This is a call for all who wish to sign the proposal, either for,
against, or abstaining. Please sign by either posting on
comp.lang.python (replying to this is fine), or sending email to

For.

I wish we'd use a better keyword than 'using' which would have a
bazillion other possible and useful future interpretations. Of the
possible keywords mentioned on the site, several (by, per, through, via)
appear better to me, and I'd particularly love the three-letter ones as
they'd align well with 'def' -- my own preference would be 'per'.

However, I suspect it's too late to submit J2 with anything but 'using'
and I'd rather have J2 with 'using' than the pie-before-def thingy, so,
here's my "tactical" FOR vote.


Alex

 

[David Fraser]

FOR

Thanks for all the hard work

David Fraser

 

[Paul McGuire]

For.

Paul McGuire

I'd also like to echo Alex Martelli's comments regarding 'per'. I can live
with 'using', but the common case of

using:
staticmethod
def foo():
pass

reads better to me with 'per' as the keyword:

per:
staticmethod
def foo():
pass

(Interestingly, @ and 'per' are similar in that "12 @ $

 

[Paul McGuire]

(Interestingly, @ and 'per' are similar in that "12 @ $ahem!

"12 @ $1.50" is shorthand for "12 at $1.50 per item" was what I meant to say
before I was interrupted!

"per" as a Latin word would help its adoption beyond English-speaking
regions. It's meaning as "by" is consistent with the decorator notion that
foo() is modified _by_ the decorator methods before it is fully defined.

And Alex cites "per"s three-letter-ness, as a nice parallel to the
corresponding "def".

-- Paul
(now I'm really through)

 

[Michael Sparks]

Alex Martelli wrote:
....

I wish we'd use a better keyword than 'using' which would have a
bazillion other possible and useful future interpretations. Of the
possible keywords mentioned on the site, several (by, per, through,
via) appear better to me, and I'd particularly love the three-letter
ones as they'd align well with 'def' -- my own preference would be
'per'.

However, I suspect it's too late to submit J2 with anything but
'using' and I'd rather have J2 with 'using' than the pie-before-def
thingy, so, here's my "tactical" FOR vote.

It might (or might not) be too late but just as a check I checked
to see if "per" is used by any projects listed in the proposal. (I
was checking Twisted and Zope for "using" so decided to do "per"
as well)

The irony here is that just like "using" the only project that I could
find out of those list on the proposal, Twisted and Zope X3 using "per"
is python itself. Specifically Idle uses it internally as a shorthand
for "percolator".

Changing the patch to work with "per" would be simple, and I don't know
if Guido reads this group (I suspect not?), but for the record "using"
and "per" appear to have the same risk level to existing code - ie very
low.

"per"'s risk level might be marginally lower because it's only used
in idle's guts, whereas "using" is used in webbrowser's get method
as a sole named parameter. Whilst I can't find anything using the
latter, it's certainly more likely than someone relying on the guts
of a project to remain unchanging.

Best regards,


Michael.
--
(e-mail address removed)
British Broadcasting Corporation, Research and Development
Kingswood Warren, Surrey KT20 6NP

This message (and any attachments) may contain personal views
which are not the views of the BBC unless specifically stated.

 

[John Crichton]

Vote: FOR

Could I also echo Fernando Perez's sentiments in thanking the people
who have worked so hard putting this proposal together.

Best wishes,

John Crichton

 

[Ksenia Marasanova]

I vote FOR J2.

Ksenia Marasanova.

 

[=?ISO-8859-1?Q?Walter_D=F6rwald?=]

Using decorators for docstrings would look much better
with the J2 proposal:

using:
"""
...
"""
def foo(bar):
...

instead of

@doc("""
....
""")
def foo(bar):
...

(Of course this requires special support for a bare string
inside the using suite)

And folding would really be helpful with long docstrings.

Consider this a vote for the J2 proposal.

Bye,
Walter Dörwald