xml-dev - RE: good book on XML

["Box, Don" <dbox@develop.com>]

> -----Original Message-----
> From: Peter Murray-Rust [mailto:peter@ursus.demon.co.uk]
> Sent: Monday, February 21, 2000 1:43 AM
> To: 'xml-dev@xml.org'
> Subject: Re: good book on XML
> 
> I was recently asked to review three XML books for the Times Higher
> Educational Supplement (the UK weekly magazine for HE). Among 
> others things
> I noted the value of public online reviews (e.g. at 
> amazon.com - anyone can
> post) from which I was able to find a lot of useful information (one
> reviewer had listed a number of typos in one book). 

As an author and techical-book-wonk, I have a different opinion of
amazon.com reviews. I periodically scan the reviews of books written by
friends or colleagues, and I find the reviews to be fairly random and
subjective. The fact that the majority of amazon.com reviews are anonymous
and/or hotmail makes things worse, although amazon has recently taken steps
to make posting anonymous reviews more difficult. My favorite amazon.com
story was regarding Jeff Richter's "Advanced Windows" book, which for a
while started getting a one-star review every three days from "A reader". It
was obvious to anyone familiar with the book that some crackpot was stalking
Jeff. However, the oblivious book buyer looks at the most recent 4 or 5
reviews and passes the book up. 

> I also 
> commented that
> fixed-date paper books were likely to be of increasingly 
> limited value and
> that the resources on the WWW itself were extremely 
> important. We have the
> opportunity in XML to create a new approach to "books" since 
> we control the
> technology of publication. An XML "book" is no longer static, but
> distributed over time, place and society. 

I agree that the "regurgitate the docs/specs" kind of book is of limited
value given the current technology churn rate in XML. However, if someone
has a story to tell, I think a book can be written to last. Does the fact
that many of the technologies that appear in Gamma et al have fallen out of
favor make the book less valuable? I don't think so. Similarly, even though
CORBA interest is on the decline, the Henning/Vinoski book is still on my
top ten list of recommendations to anyone building distributed apps.  I
think that the traditional publishers (Morgan Kaufman, Prentice Hall,
O'Reilly, Addison Wesley) generally are doing a good job of putting out
books that remain relevant for more than 12 months. 

An interesting case study in publishing is wrox. Wrox is not a classic book
publisher. Most (but not all) wrox books are really a collection of articles
written by multiple individuals. Yes, there are exceptions (Grimes, Mohr,
Esposito), but the average wrox book cover has more head-shots than Cindy
Crawford's portfolio. 

In essence, wrox is in the magazine business, however, instead of
subsidizing the publication with ad revenue, the cover price is roughly ten
times higher than a traditional magazine. That doesn't mean that wrox stuff
is better or worse than "traditional" books, it just means that one cannot
generalize the wrox model to all books on the market. 

All of this stated, I see the virtual book approach is more of a viable
alternative to wrox-style books than to the traditional
"beginning-middle-and-end" kinds of books like Henning/Vinoski or say
Stephen Mohr's book from wrox. However, why couldn't this effort be
coordinated via Tim O'Reilly's XML.com site? Currently, XML.com typically
publishes short articles, but the folks on the masthead seem pretty plugged
in to the XML world. I for one would rather see something coordinated by
folks with a grounding in traditional publishing than an ad hoc effort
(which we should reserve for more important things like standardizing APIs
;-) ;-)).

DB
http://www.develop.com/dbox

***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@xml.org&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/threads.html
***************************************************************************