Re: Review: Boost Test Library

From: williamkempf_at_[hidden]
Date: 2001-02-16 10:57:18

--- In boost_at_y..., "Ed Brey" <brey_at_a...> wrote:
> From: "David Abrahams" <abrahams_at_m...>
> > The only excuse for these acronyms, AFAICT, is that it gets 
cumbersome to
> > refer to a library as "The Boost Foo Library" over and over 
within the
> > library's own documentation. I didn't want an acronym for the 
Python lib,
> > but eventually found it hard to make the docs read easily without 
one. Can
> > anyone think of a good alternative?
> 
> We can take advantage of being within the Boost "namespace" within 
our docs
> and just say things like "See the Test library for me details, and 
look at
> the cool Python library."  I thinking capitalizing the name of the 
library
> will make things clear enough.
> 
> Data point: When I posted my review with subject "Review: Test 
library", was
> there any ambiguity regarding what the review was of?

Within the Boost eGroup I'd expect there to be zero confusion.  This 
may or may not be the case within the libraries.  Many Boost 
libraries exist both as a part of Boost, and as seperately available 
libraries from other sites.  Even within the Boost archives, many 
users are searching for only a single library and are pointed to 
Boost where they d/l the package with the intention of only using the 
one library.  If the documentation for that library references 
another library within Boost and "shortens the name" by removing 
Boost from the title, this user may be confused by what library is 
being referred to.  Worse still, not all Boost libraries go by the 
name "Boost _____ Library", which will lead to further confusion when 
not using the full and proper library name.

In the case of the Boost Thread Library, the documentation has to 
referrence other thread libraries from time to time, so referring to 
itself as the Thread Library in the docs is still likely to cause 
confusion despite the capital letters (which a simple typo could 
cause to be an even worse problem).

I'm not sure what the "proper" solution should be, but it doesn't 
seem to be a simple one to me.

Bill Kempf