@bankim: thanks for the kudos! i have to admit i'm one of those mutant engineers that actually likes teaching; hopefully my experience using Python really helped with giving you that perspective as opposed to someone who's just teaching you Python syntax which you can basically learn on SO or online elsewhere!

@kqr2: this is what led me to my opening remark on how different learning styles require different styles of writing ... believe it or not DiP has gotten good reviews... just check Amazon. the problem i alluded to is that it's not for everyone.

Here's are similar comparisons between my book and Mark's as documented by Eric Wilson both on his blog as well as in an Amazon review... see http://amazon.com/review/RKG44D8GQYLNL and http://wilsonericn.wordpress.com/2011/09/26 for details. So yes, some people don't like it, but to others, it's just what they needed.

@mapleoin, @rhizome31, @wilduck: yep, that's another list, for "expert Python readers". i would like to add that at some point, but my editorial constraints were 1500 words or so. i'd also like to review all those before making suggestions too. i think you're right in that there's probably some overlap with mark's book, but our writing styles are very different (check Amazon reviews for more details), and i've tried to add more contemporary applications in this edition of Core Python.

@ashconnor: that's another valid type of reading list. in this one, i think i compiled 3 discrete lists instead of your more progressive idea, which is a good one. i'd like to see what's in your list. as far as the overlap is concerned, yes, you're right about that as far as the material goes, however the delivery is different and suited to a variety of reader types.

It's easier to switch away and not be locked-in if you develop your app using Python/Django (see my comment above).

in addition to AppScale also look at the TyphoonAE project... these are open versions of the App Engine backend that claim to be fully-compatible to the SDK/client-side.

also to fight vendor lock-in, Python developers can create Django apps then run them on App Engine using django-nonrel or on any traditional hosted stack supporting Django; or move between then w/minor (settings.py) changes; don't forget data migration as well.

Apparently it's not only for hobbyists... it's the leading favorite amongst enterprise CIOs too, as seen in this Morgan Stanley survey which just came out last week: http://goo.gl/LcbS

Don't forget to use django-nonrel so that you can move apps b/w App Engine and any normal ISP supporting Django apps. You only need to change your database in settings.py! http://allbuttonspressed.com

that is changing... cloud/hosted SQL announced in May 2010:

http://code.google.com/appengine/business/#features

http://googlecode.blogspot.com/2010/05/announcing-google-app...