Re: Client/Service Design Style.

John Carter <[email protected]> writes:
> This is such an old idea, such an obvious notion, yet a Good One. All
> my own code is written in this manner and it produces such elegant
> designs it surely must be a standard practice...
[…]
> Any references?

Rereading your post, this may be tangential, but you should look at
REpresentational State Transfer (REST) – as defined by Roy Fielding's thesis,
not the sometimes bastardized pop-technical version of late – is an
architectural style which is based on Client-Server.  Chapter 5¹ of his
thesis is the derivation of the style by the layering of a certain set of
constraints.  Chapters 1² and 3³ are more like an abstract overview of
"Software Architecture" and "Network-based Architectural Styles", including a
systematic classification of the properties of various styles of
architectures.  Those section's references should lead into the literature.

I'd love to see other types of works along the same lines.  Blog postings and
even technical articles can often be so hand-wavy about what the architecture
or design being espoused actually accomplishes with its contraints.

¹: http://www.ics.uci.edu/~fielding/pubs/dissertation/rest_arch_style.htm
²: http://www.ics.uci.edu/~fielding/pubs/dissertation/software_arch.htm
³: http://www.ics.uci.edu/~fielding/pubs/dissertation/net_arch_styles.htm

-- 
...jsled
http://asynchronous.org/ - a=jsled; b=asynchronous.org; echo ${a}@${b}

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)

iEYEARECAAYFAkg2vVAACgkQXze2dBAuga2ELACfbHWFpeY9MV+3FJ0Vr0t9GqHj
fWUAoJ/9W9RzAK5rp/qdaMsFulY5jNla
=2taP
-----END PGP SIGNATURE-----