wonderings

[Matthias.Lang@REDACTED]

 > The Erlang book and the documentation are great, but I now feel that a book 
 > about the Erlang libraries and the OTP libraries would be very
 > helpful. 

 > More detailed explanations and examples would make life much easier!
 > 
 > Are there any plans? Or maybe there already is such a book? Does
 > anyone else think it's a good idea?

I'd love to be able to buy a really neat, well-written book about more
advanced Erlang topics. I'd also love to get a monthly
newsletter-style thing with nice Erlang ideas, along the lines of the
C++ report. Such a report would be an excellent way to avoid that
"Duh! That whole subsystem was buried in a manual and I re-invented it
because reading manuals from cover to cover sucks" feeling.

The latter would be particularly nice for a less referene-oriented
walk-through of a particular topic, perhaps written by someone with
quite a bit of experience.

I'm not so sure that a book about OTP would be that great a
success---the OTP books are actually quite good, though they are much
nicer to read on paper and, as far as I know, the only way to get
paper copies is to buy a supported copy of Erlang. In any case, 
I mention all the books about Erlang in questions 3.5 and 3.6 
of the all-new Erlang FAQ at

   http://www.ericsson.se/cslab/~mml/faq/x122.html#AEN160

A few of us had some corridor discussions about "a new Erlang book" a
while back. Some thoughts which came out of that were:

- Many people would like a more example-driven replacement for the
  "Erlang book". This would contain an up-to-date description of
  Erlang, contain different (and more) examples, choosing things
  which people are more likely to have come up against, e.g. a
  satellite control system seems pretty irrelevant, but a simple
  WWW server is something many of us have tried or at least thought
  about. 

  It also seems tempting to move many of the longer the code
  examples out of the book and onto a www site.

- Others would like an advanced Erlang book, maybe something like
  "Large Scale Erlang Software Design" (title rip-off deliberate).
  The catch here is that there are very few people qualified to write
  such a book, and most of them are busy working on big systems.

- Writing an advanced book is likely to be strictly a labour of love, maybe
  100 people would understand it. Of course, when Erlang really
  takes off...

Is it possible to write a good book about OTP without it being an
advanced Erlang book or being merely a rewrite of the OTP books?
For instance, I can't see a huge scope for improving what the manuals
say about Erlang applications and Supervision trees than the
manuals have:

http://www.erlang.org/documentation/doc/design_principles/part_frame.html
  
Matthias