Patch uploaded what's next?
Rolf
some weeks ago I uploaded a patch adding new probability
distributions. I even made the recommended changes immediatly. Nothing
happend ever since. Did I miss something?
Simon King
Best regards,
Simon
Rolf
my mistake. It's here http://trac.sagemath.org/sage_trac/ticket/11572
I even added some notebooks to test the new capabilities.
http://www.sagenb.org/home/pub/2887
http://www.sagenb.org/home/pub/2886
Waiting for further suggestions.
Best
Simon King
first of all, I am sorry that I can not review it, myself, since
probability distributions are very far from my own field.
More experienced developers may correct me, but I thought that one of
the responsibilities of the owner of a ticket is to approach potential
reviewers, if it is apparent that nobody else has noticed the ticket.
If the owner forgot about the ticket (holidays and stuff), I think it
is perfectly fine if the author him- or herself asks on one of the
sage lists (probably sage-devel is a good choice, or a specialised
list if that exists for the given topic).
If you know a potential reviewer, then you could ask that person by e-
mail, of course. Note that you can add people as "Cc" on a ticket; so,
if something changed, they'll be informed.
I can understand your frustration. Some of my patches rotted on trac
for months, for instance #9138, #11115 and #11342 (sorry for the
ad...).
Perhaps I am glorifying the past, but it seems to me that the
development process used to be "snappier" a few years ago.
Best regards,
Simon
William Stein
This might be partly my fault since I purposely have been working (a
lot) less directly on Sage during the last 2 years. I had to take a
break to avoid burnout, and get momentum back on some of my other
important work (e.g., research papers).
When I was preparing for my talk [1] in Budapest last month I started
feeling very inspired to spend much more time on mainline Sage again,
and have been starting to work more on it.
Working on getting the Sage trac ticket review process under control
(see [2]), probably following the standard mathematics journal model,
would be very valuable at this point in the development of Sage.
[1] http://wstein.org/talks/2011-07-13-FoCM_plenary/
[2] http://trac.sagemath.org/sage_trac/report/10
-- William
kcrisman
> > is perfectly fine if the author him- or herself asks on one of the
> > sage lists (probably sage-devel is a good choice, or a specialised
> > list if that exists for the given topic).
has asked to receive notifications about this type of ticket.
In the case of this ticket, there were some disagreements as well
about how to structure things in this module.
> > Perhaps I am glorifying the past, but it seems to me that the
> > development process used to be "snappier" a few years ago.
>
> This might be partly my fault since I purposely have been working (a
> lot) less directly on Sage during the last 2 years. I had to take a
> break to avoid burnout, and get momentum back on some of my other
> important work (e.g., research papers).
ownership of Sage by people other than William Stein now, which is
very positive. However, that said, having his energy is very
helpful :)
There's also the issue of people not having as much time to review as
before? We definitely need more reviewers in the mix.
Jose Guzman
I am particularly interested in the implementation of different probability distributions in Sage. The ticket seems like a very important contribution to Sage (I was myself trying to implement the binomial and some other strange distributions). I guess, with your implementation I will be able to solve some problems that I was only able to work with Mathematica. The sage notebooks are every illustrative, and they all look awesome!. I am not an expert in sage development, but I promise I will have a look.
Feel free to contact me.
Best.
Jose Guzman
I am very interested in the implementation of different probability distributions in Sage. I was myself trying to implement some distributions (log-normal, binomial). I found your ticket look like an excellent contribution to Sage. The sage-notebooks look awesome!!
I am not an expert in sage development, but I promise I will have a careful look to your tickets.
Best
Jose.
Ivan Andrus
>>> If the owner forgot about the ticket (holidays and stuff), I think it
>>> is perfectly fine if the author him- or herself asks on one of the
>>> sage lists (probably sage-devel is a good choice, or a specialised
>>> list if that exists for the given topic).
>
> I think that "owner" is little more than a random name of someone who
> has asked to receive notifications about this type of ticket.
Oops. I usually make myself owner of tickets that I am working on, mostly because that's how we did it where I used to work.
> I don't think you should blame yourself. I think there is a lot more
> ownership of Sage by people other than William Stein now, which is
> very positive. However, that said, having his energy is very
> helpful :)
>
> There's also the issue of people not having as much time to review as
> before? We definitely need more reviewers in the mix.
Another trick is to review someone else's ticket and then ask them to review yours. e.g. kcrisman has reviewed so many of my tickets that I would be obliged to review anything he asked me to (assuming I were competent to do so of course). Unfortunately, I can't help with probability distributions, or I would ask you to review some of mine :-)
-Ivan
P.S. kcrisman, I've been planning to work on 8473, but can't seem to find the time. Thanks for your patience.
kcrisman
On Aug 17, 4:26 am, Ivan Andrus <darthand...@gmail.com> wrote:
> On Aug 16, 2011, at 10:51 PM, kcrisman wrote:
>
> >>> If the owner forgot about the ticket (holidays and stuff), I think it
> >>> is perfectly fine if the author him- or herself asks on one of the
> >>> sage lists (probably sage-devel is a good choice, or a specialised
> >>> list if that exists for the given topic).
>
> > I think that "owner" is little more than a random name of someone who
> > has asked to receive notifications about this type of ticket.
>
> Oops. I usually make myself owner of tickets that I am working on, mostly because that's how we did it where I used to work.
there is always a "default" owner of a ticket, depending on its
component.
> > I don't think you should blame yourself. I think there is a lot more
> > ownership of Sage by people other than William Stein now, which is
> > very positive. However, that said, having his energy is very
> > helpful :)
>
> > There's also the issue of people not having as much time to review as
> > before? We definitely need more reviewers in the mix.
>
> Another trick is to review someone else's ticket and then ask them to review yours. e.g. kcrisman has reviewed so many of my tickets that I would be obliged to review anything he asked me to (assuming I were competent to do so of course). Unfortunately, I can't help with probability distributions, or I would ask you to review some of mine :-)
>
> -Ivan
>
> P.S. kcrisman, I've been planning to work on 8473, but can't seem to find the time. Thanks for your patience.
much requested, as you are well aware.
Rolf
> about how to structure things in this module.
a misunderstanding. I thought we agreed upon quick release and
postponement of these structural questions.
Besides the present release works perfect (as far as I can say) and it
would be no problem at all to contruct two new classes for continuous
and discrete distributions by simply inheriting all functionality of
the present class plus an error message if the requested distributions
doesn't belong to this class. That's all about it. 5 minutes work no
more. But I sincerely wonder if this is really necessary, Does a sage
user really needs such a reminder? As long as we don't come up with
new ideas of additional functionality to be put into these
hypothetical classes, I think even these 5 minutes are waste of
time :-)
Best Regards
Rolf
promise.
Best Regards
Rolf
> careful look to your tickets.
sounds good. Looking forward to.
Best regards