|
|
Boost : |
Subject: Re: [boost] Formal Review: Boost.Polygon starts today August 24, 2009
From: Paul A. Bristow (pbristow_at_[hidden])
Date: 2009-09-16 10:30:29
A few more thoughts on my (nano-) review of Boost.Polygon.
> Please always state in your review, whether you think the library should be
accepted as a Boost library!
Boost.Polygon *should* be accepted as a Boost library.
BUT it should NOT be called "Boost.Polygon" - as previous noted by Dave
Abrahams - because it is not as fully generic as is possible.
Boost.Polygon2D4590 perhaps ;-)
I do not believe that Boost is *only* about stuff that can be considered for C++
Standardization.
Stuff that is useful (with Good Quality) is also BoostWorthy IMO.
'Boost.Polygon2D4590' clearly does need some people's expectations and needs,
and promises meet my expectations for quality.
This would not preclude other library (or libraries) that aspires to be more
generic.
Although C++ provides tools to make things generic, there are still limits to
what can be done without a price in compile and/or run time. 1D, 2D and 3D
really have major differences that are still difficult to 'program' - as I have
found just drawing plots.
So I conclude from reading the discussions that some more learning is
inevitable. It would be better to accept the real difficulties, and not expect
to do the job in one pass. If necessary, to accept Boost.Polygon2D4590,
Polygon1, Polygon2 ...
Meanwhile I think it would be a bad mistake to not make something useful
available NOW, especially when supported by a major company. If this library is
rejected, its Open Source development will probably cease.
The Best is the Enemy of the Good.
Paul
--- Paul A. Bristow Prizet Farmhouse Kendal, UK LA8 8AB +44 1539 561830, mobile +44 7714330204 pbristow_at_[hidden]
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk