Boost logo

Boost :

Subject: Re: [boost] [process] Formal Review starts today, 27 October
From: Paul A. Bristow (pbristow_at_[hidden])
Date: 2016-10-28 04:57:37


> -----Original Message-----
> From: Boost [mailto:boost-bounces_at_[hidden]] On Behalf Of Antony Polukhin
> Sent: 27 October 2016 20:28
> To: boost_at_[hidden] List
> Subject: Re: [boost] [process] Formal Review starts today, 27 October
>
> 2016-10-27 20:41 GMT+03:00 Klemens Morgenstern <klemens.morgenstern_at_[hidden]>:
> > Am 27.10.2016 um 10:47 schrieb Johannes:
> >>
> >> Hello,
> >>
> >> following you can find my little review of the process library.
> >> Unfortunately, I have no more time to dive into more details of the library.
> >> But from what I have read and seen on the list I think that the library
> >> would be a good addition to boost. The author seems to be quite
> >> knowledgeable about the library domain and I think that he will bring it
> >> into the state that will be consensus after the review.
> >>
> > Thank you for your review. It was surely a mistake to only read the sources
> > of the doc for errors, so I missed too many things; I'm really confused why
> > so many links didn't work. I fear I cannot change the documentation during
> > the review, but you're certainly right, it is a weak point. Nevertheless,
> > I'll go through and address your points, so other reviewers might have some
> > questions answered.
>
> It is allowed to fix links/typos in the docs during review. Just make sure that:
> * the docs are always available
> * that sections do not change their names and addresses
>
> In that way people will get the fixed docs and links to sections will
> remain valid during the whole discussion.

The way Quickbook works, I believe it is impracticable to make *all* links to work unless the docs are in the right place in the
full Boost source code tree (or as I have done previously, a skeleton Boost tree containing all the essential stuff).

So readers must checkout/download the docs using github into readers Boost tree to get the documentation in all its glory. However,
I don't think that this is necessary to review the library.

This problem goes away when it is formally accepted and built by the Boost docs build process.

There are much more important issues to worry about in a review?

Paul

---
Paul A. Bristow
Prizet Farmhouse
Kendal UK LA8 8AB
+44 (0) 1539 561830

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk