| |
Description:
|
|
|
Bioinformatics
Software Engineering: Delivering Effective Applications will be useful
to anyone who wants to understand how successful software can be
developed in a rapidly changing environment. A handbook, not a
textbook, it is not tied to any particular operating system, platform,
language, or methodology. Instead it focuses on principles and
practices that have been proven in the real world. It is pragmatic,
emphasizing the importance of what the author calls Adaptive
Programming - doing what works in your situation, and it is concise,
covering the whole software development lifecycle in one slim volume.
At each stage, it describes common pitfalls, explains how these can be
avoided, and suggests simple techniques which make it easier to deliver
better solutions."Well thought-out ...addresses many of the key issues
facing developers of bioinformatics software." (Simon Dear, Director,
UK Technology and Development, Bioinformatics Engineering and
Integration, Genetics Research, GlaxoSmithKline) Here are some examples
from the book itself. On software development: "Writing software
properly involves talking to people -- often lots of people -- and
plenty of non-coding work on your part. It requires the ability to
dream up new solutions to problems so complicated that they are hard to
describe." From description to specification: "Look for verbs -- action
words, such as 'does', 'is' and 'views'. Identify nouns -- naming
words, like 'user', 'home' and 'sequence'. List the adjectives --
describing words, for example 'quick', 'simple' or 'precise'. The verbs
are the functions that must be provided by your application. The nouns
define the parameters to those functions, and the adjectives specify
the constraint conditions under which your program must operate." On
how to start writing software: "Handle errors. Take in data. Show
output. Get going!" On testing: "It may not be physically possible to
test every potential combination of situations that could occur as
users interact with a program. But one thing that can be done is to
test an application at the agreed extremes of its capability: the
maximum number of simultaneous users it has to support, the minimum
system configuration it must run on, the lowest communication speed it
must cope with, and the most complex operations it must perform. If
your program can cope with conditions at the edge of its performance
envelope, it is less likely to encounter difficulties in dealing with
less challenging situations." On showing early versions of software to
users: "It can be hard explaining the software development process to
people who are unfamiliar with it. Code that to you is nearly finished
is simply not working to them, and seeing their dream in bits on the
workbench can be disappointing to customers, especially when they were
expecting to be able to take it for a test drive.
" On bugs: "If your
users find a genuinely reproducible bug in production code, apologize,
fix it fast, and then fix the system that allowed it through. And tell
your customers what you are doing, and why, so they will be confident
that it will not happen again. Everybody makes mistakes. Don't make the
same ones twice." And one last thought on successful software
development: "You have to be a detective, following up clues and
examining evidence to discover what has gone wrong and why. And you
have to be a politician, understanding what people want, both in public
and in private, and how this is likely to affect what you are trying to
do. This book cannot teach you how to do all of that, but it can help."
|
|
|