doc revisions in: feature/issue-211-stan2-manual

[Bob Carpenter]

I pushed all the doc revisions I had. There were a ton
of fixes.

I added a new chapter at the start of the programming
chapter on general programming techniques.

I added a lot of explanation and also command skeletons
and more examples in the command-line Stan chapter. But
most of that's still text from Michael.

I added the sampling syntax to all the distributions and
also indexed it.

- Bob

[Michael Betancourt]

Looks good -- there are a few tweaks I'd like to make. Should I fork and pull-request on the feature branch?

> --
> You received this message because you are subscribed to the Google Groups "stan development mailing list" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to stan-dev+u...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.

[Daniel Lee]

I'd say make the changes on the branch directly. It'll be easy to tell what was changed in the commit log.

[Bob Carpenter]

Definitely just edit the branch I already have going.
The whole point is that we want the pull request checks
before merging into develop, not before pushing updates
on a branch. (Though we could go recursive in the future
when we have 20 or 30 developers :-))

- Bob

On 10/8/13 8:15 AM, Daniel Lee wrote:
> I'd say make the changes on the branch directly. It'll be easy to tell what was changed in the commit log.
>

> On Oct 8, 2013 5:49 AM, "Michael Betancourt" <betan...@gmail.com <mailto:betan...@gmail.com>> wrote:
>
> Looks good -- there are a few tweaks I'd like to make. Should I fork and pull-request on the feature branch?
>

> On Oct 8, 2013, at 6:41 AM, Bob Carpenter <ca...@alias-i.com <mailto:ca...@alias-i.com>> wrote:
>
> > I pushed all the doc revisions I had. There were a ton
> > of fixes.
> >
> > I added a new chapter at the start of the programming
> > chapter on general programming techniques.
> >
> > I added a lot of explanation and also command skeletons
> > and more examples in the command-line Stan chapter. But
> > most of that's still text from Michael.
> >
> > I added the sampling syntax to all the distributions and
> > also indexed it.
> >
> > - Bob
> >
> > --
> > You received this message because you are subscribed to the Google Groups "stan development mailing list" group.
> > To unsubscribe from this group and stop receiving emails from it, send an email to

> stan-dev+u...@googlegroups.com <mailto:stan-dev%2Bunsu...@googlegroups.com>.

> > For more options, visit https://groups.google.com/groups/opt_out.
>
> --
> You received this message because you are subscribed to the Google Groups "stan development mailing list" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to

> stan-dev+u...@googlegroups.com <mailto:stan-dev%2Bunsu...@googlegroups.com>.

[Michael Betancourt]

Just being git-polite. ;-)

[Bob Carpenter]

On 10/8/13 1:36 PM, Michael Betancourt wrote:
> Just being git-polite. ;-)

Thanks. I appreciate it. I hate dealing with
merge conflicts.

- Bob

>
>> On Oct 8, 2013, at 6:20 PM, Bob Carpenter <ca...@alias-i.com> wrote:
>>
>> Definitely just edit the branch I already have going.

...

>>> On Oct 8, 2013 5:49 AM, "Michael Betancourt" <betan...@gmail.com <mailto:betan...@gmail.com>> wrote:
>>>
>>> Looks good -- there are a few tweaks I'd like to make. Should I fork and pull-request on the feature branch?

...

[Michael Betancourt]

I pushed my changes -- mostly respecting the new organization (deviating only to clean up all the dangling subsections). The optimization template text is a bit choppy, but I was too tired to fix it.

The full argument hierarchy was originally meant to just be a formatted version of the help-all output so that the doc could be quickly updated whenever the args were tweaked. We'll just have to be careful to keep them up to date now.

[Bob Carpenter]

On 10/8/13 8:05 PM, Michael Betancourt wrote:
> I pushed my changes -- mostly respecting the new organization (deviating only to clean up all the dangling subsections). The optimization template text is a bit choppy, but I was too tired to fix it.
>
> The full argument hierarchy was originally meant to just be a formatted version of the help-all output so that the doc could be quickly updated whenever the args were tweaked. We'll just have to be careful to keep them up to date now.

I didn't have the energy to change the way the help
was formatted to match :-)

We can go either way if there are major changes --- starting
from where we are now, or starting from the formatted version
of help-all.

- Bob