Lambda the Ultimate

I've not read the book, only the sample chapters on the website. Those seem to barely touch the surface of their languages, though I did like the introduction to Icon.

I think it was meant to introduce more than teach.

The extracts are only the first few pages of a chapter.

Full disclosure: I was one of the technical reviewers for the book, and my book Debug It! is published by the same guys.

The average LtU reader is not the target market for this book. It provides an introduction to each of the seven languages - more than just "Hello World", but definitely not the whole of the language. It's far more about opening people's eyes to alternatives and providing a "flavour" of different approaches.

I'd bet that anyone reading this site has already had their eyes opened :-) But with luck, the book will sow seeds in a few fertile minds that will lead to them becoming LtU readers. And it's definitely a good book to have lying around in case you know someone who you think might be ready to break out of the Java monoculture.

While it's not the best wide read on the market, I enjoyed it. Bruce summarizes each language clearly, and shows a number of solid examples of the interesting bits from each facet of the language. It's a light read for this audience, but still worth the price of the eBook IMO.

Now if you're looking for a deeper look into general language flow and design, I recommend Michael L. Scott's 'Programming Language Pragmatics'. It covers aspects of over forty different languages from the perspective of language design. I've read this one a number of times now, a text that has been instrumental in a number of DSL designs I've contributed to over the years.

As has been pointed out before: “Keeping up with events in the worlds of Messrs Eckel and Tate is going f-a-a-a-r beyond the call of duty.”

Beauty lies in the eye of the beholder, but what Tate calls “a beautiful example of method_missing in action” I would rather consider poor and ugly: Towards the end of the chapter on Ruby, he uses method_missing to encode the functionality of mapping from representations of Roman numerals to integers.

I have read up through the chapter on Ruby and I feel this book presents some of the most interesting features of Ruby and serves as a sort of a heads up on how you would proceed to program in Ruby. Imagine a book that sorted out the mundane aspects of a language leaving only its most interesting features and you have this book. While I couldn't build a Ruby application as of now, I would certainly know how to plan out doing so as I know the capabilities of the language. (The meta programming aspects of Ruby seem to be as important in application design as considerations of object orientation.)

I'm in the middle of the book, and my impression is this: as with most books, if you put in the effort, you'll get something out of it. The difference is that it's much more obvious with this book that you need to do some extra legwork.

As others have pointed out, exercises include things like "go find and read the official documentation for this language"---a nontrivial assignment! Nevertheless, the book attempts to walk you through enough of the language so that one not-completely-trivial example of an idiosyncratic use of the language can be demonstrated.

In Ruby, for example, the author implements a class that, when included in another class, provides a method to populate the latter with methods for accessing CSV data stored in a file. The intent is to demonstrate that Ruby's metaprogramming facilities can be used to implement back door multiple inheritance.

Of the seven languages in the book, I came in proficient with Haskell and familiar with the semantics (but not syntax) of Prolog (having worked through The Reasoned Schemer). Sure, that meant the Haskell chapter was pretty much useless for me, but in a sense it gives me insight into the relative merit of the other chapters. In particular, it's clear that there's just not enough space to cover everything (as others have mentioned, Monads are completely skipped!), and the Haskell chapter is treated like a lesson in functional programming using Haskell syntax. Nevertheless, to someone who has no experience with FP, it's certainly exposure to something new and different.

The book is not particularly dense, which is good inasmuch as it makes for nice light reading; obviously this is at the cost of more comprehensive coverage of the material.

Overall, I think this book is worthwhile as a way of dipping your feet into a few different languages so that you can pick and choose which ones to explore further. Perhaps a better title would have been "seven programming paradigms in seven weeks," since the book serves more to teach about the paradigms than about the languages used in the teaching.