Your book looks great and I look forward to reading more of it. I also
applaud your choice of target audience; the lay reader/user/programmer.
Data processing is essentially simple, the complexity of current
SW products are the results of poor development and probably also
I strongly applaud your aim at making a Smalltalk as Smalltalk should
have been and hope to be able to add my two bits worth to your effort.
Targeting the lay reader poses makes it hard to discuss the hard parts.
I strongly believe that a new departure such as Spoon must be
created on a solid foundation. For example in chapter twenty-two:
imprinting and modules. Every object has a unique identity. A copy
of the object can exist in a remote image. But then there are two
distinct objects with the same identity, thus violating the fundamental
rule. This rises a number of fundamental problems that need to be
studied in depth, otherwise we may be building our edifice on shifting
Another concept that needs to be carefully studied is the notion of a module.
I take it it is here a module of objects. There is the Islands in
Tweak, but I find this unsatisfactory because it is not well
Another comment. chapter twenty-three: collaboration: Documentation
is a two-edged sword. It is hard to keep it synchronized with a rapidly
changing program; it is hard to read and often hard to understand
because it is terse and often uses an unfamiliar terminology. I aim at
making the code itself more expressive so that it becomes more readable.
The document should be dated.
estimated completion date: 1 June 2007 is somewhat unlikely
A few figures would have helped me. As it is its target audience
are pure left-brainers.