BKMSENNT.RVW   20020523

"Microsoft Encyclopedia of Networking", Mitch Tulloch/Ingrid Tulloch,
2002, 0-7356-1378-8, U$79.99/C$115.99
%A   Mitch Tulloch info@mtit.com www.mtit.com
%A   Ingrid Tulloch
%C   1 Microsoft Way, Redmond, WA   98052-6399
%D   2002
%G   0-7356-1378-8
%I   Microsoft Press
%O   U$79.99/C$115.99 800-6777377 www.microsoft.com/mspress
%P   1313 p. + CD-ROM
%T   "Microsoft Encyclopedia of Networking, Second Edition"

As soon as I sent him the draft of my review of the first edition,
Mitch asked me to hold off, as the second edition was in the works. 
He stated that he was addressing the issues I had identified, and that
the second edition would have fixed them.  His means of dealing with
the problems are ... interesting.

The scope of the encyclopedia is stated to cover networking concepts,
the Internet, and Microsoft products.  The primary audience is no
longer limited to novices pursuing the MCSE (Microsoft Certified
Systems Engineer) designation, although the Microsoft emphasis is
still fairly clear.

The Microsoft orientation and bias is not quite as explicit as it was
in the first edition, but is still evident.  The errors in dealing
with the redirection (>) and pipe (|) symbols have been eliminated:
the section on "Numbers and Symbols" no longer defines any symbols. 
"Access control" and "clustering" are stated to be "[a]ny technology"
performing the respective functions, but, after a single initial
sentence in this generic fashion, there are two pages that relate only
to Microsoft products.  Impersonation is still defined only in terms
of assisting Windows client/server communication, which is startling
in view of the importance of impersonation as a security exploit.

Now, is it reasonable to complain about a Microsoft emphasis in what
is, after all the *Microsoft* networking encyclopedia?  Well, yes,
when it gets in the way of real information.

A number of entries have little apparent function.  There are, for
example, a number of listings for variant flavours of Ethernet, and
these items seem to describe only different vendor products.  In
addition, there is a great deal of repetition, fluff, and padding in
the writing.  The text often says the same thing over again in a
slightly different way, but this neither develops the topic, nor
really assists the novice user in understanding complex subjects.

Basic networking concepts are covered and, generally, the material is
reasonable, if uninspired.  However, a number of the fundamental ideas
are covered in such a way that the newcomer will not gain a full
understanding of the idea.  In many cases it is difficult to say that
the explanation is in error, but the abstraction could certainly have
been presented in a better way.  "Bursty" traffic, for example, is
described in terms of transferring video files, and any self-
respecting MPEG is going to be big enough to occupy a pipeline with
less capacity than an OC-192 for longer than a mere "burst."

While many entries are longer than the paragraph or two one might
expect from a dictionary, the content doesn't deliver much more
information.  Frame relay, for example, is described in terms of
packet switching, but the discussion of error checking, which
differentiates the two technologies, is almost lost in the sales
pitches for vendors of the service.

As one has come to expect from a Microsoft product, security and
privacy concerns are downplayed at every turn.  The best possible
construction is put on issues such as Authenticode and cookies. 
Again, while the descriptions are not necessarily erroneous, counter-
examples are easily generated.  A cookie, for example, cannot give out
your email address, as the book says.  Unless, that is, you have input
your email address to a Website, and the site has stored the
information in a cookie.  This is a fairly common occurrence.

The entry for virus is pretty appalling.  The aren't quite as many
errors as there were last time, but there isn't anything to help,
either.

Would this book help someone study for the MCSE?  Probably.  One of
the major difficulties in writing the exam is clearing your mind of
how things work in the real world, and sticking to the Microsoft
terminological party line.  Would it help anyone else?  Possibly, but
there are many other works more complete, readable, and reliable.  The
"Microsoft Press Computer Dictionary" (cf. BKMSCMDC.RVW) is much
better: a fairly solid reference over a wide range of issues.  It is
unlikely that anyone with more than a passing acquaintance with
networking will find much of value in this encyclopedia.  Certainly,
with the wide variety of excellent and reliable communications
dictionaries available one wonders at the need for this.  For general
networking there is Newton (cf. BKNTTLDC.RVW), for authority there is
Weik (cf. BKCMSTDC.RVW), and for history and background there is
Petersen (cf. BKDTTLDC.RVW).

copyright Robert M. Slade, 2001, 2002   BKMSENNT.RVW   20020523