Addison Wesley
Cambridge University Press
Wiley
OReilly
Wrox
Microsoft Press
Prentice Hall

Apress

A free pdf version:Cross-Platform GUI Programming with wxWidgets

By Julian Smart and Kevin Hock with Stefan Csomor

http://www.informit.com/content/images/0131473816/downloads/0131473816_book.pdf

Obtained a copy of Software Engineering (by Ian Sommerville) 8 Ed. It touched all the aspects of software engineering and provided a good overview on classic to modern software engineering aspects. It at last cleared my confusion over Project Management and Software Engineering:). It also slowed down my pace towards AOSD since I want to spend more time on practical domain since I am not a researcher.

Here I am trying to analyzing some opinions of PSP fitting into life of real world.
PSP opinion: You should spend at least half as much time reviewing a product as you spent producing it.
Onega: I never see a schedule is planned so.
PSP opinion: For a million LOC, good quality code has n defects and each defect will cost m hours, so the total cost is m*n; for a million poor quality code, there are a * n defects, so the cost is a * m * n hours.
Onega: The efforts for a defect varies dramatically from weeks to minutes, and it is not unusual that one line of code will fix several defects.
PSP opinion: PSP process will produce better output than without it.
Onega: That is because extra training is arranged. If the same budget is arranged to train programmers on design skills and coding skills, I'll bet that the output would be even satisfiable.
PSP opinion: statistics proved that PSP process output is very prominent.
Onega: people tend to perform better under monitoring. If management team hire external/internal team for the same level of guiding/monitoring, I would expect about the same level of output.
Conclusion: there is no magic under certain methods. The training, experience and efforts make difference! Given the same budget, management play an important role because it determines how much output can get from each member. PSP is not silver bullet!

http://www.artima.com/cppsource/backyard.html presented some rarely used scenario of C++. RTTI Sniping in order to provide a fast implementation replacing dynamic_cast. Vptr Jamming in order to change type of an object (between types of the same allocated size) Pointer Hydration in order to speed serialization.

I have a glance on this book by Eric S. Raymond. My feeling is that the community is a great source for excellent ideas, and good authors can collect them into a book. I say quite a few books claimed that some ideas are from the news group.
From the coding skills standpoint, I don't see anything new -- most of the ideas I can found in "Code Complete 2" and "Refactoring Cookbook". But it does contains many valuable analysis on existing technologies, the competitor is "Patterns of Enterprise Application Architecture" by Martin Fowler. Martin is also talking of existing architectures, I didn't see much overlap between them. So if you like one, most probably you'll the other.

Just got the book "C++ Coding Standards 101 Rules, Guidelines, and Best Practices" by Herb Sutter and Andrei Alexandrescu. This book is beyond C++ tutorial. If you feel familiar to the name "Andrei Alexandrescu"? Check "Modern C++ Design" and "Loki".
Herb Sutter is author of "Exceptional C++" and "More Exceptional C++".

An in depth book introducing new technology of 1996. I am surprised to read the presence of OpenDoc, after a little further reading of OpenDoc in internet, it was dead for some time...
Interesting metaphor:
RPC functions like procedural language, functions are exposed.
CORBA functions like object oriented language, objects are exposed.
CORBA/OpenDoc is technically superior to OLE, but the result is not controlled by technique.

Imperfect C++
Author: Matthew Wilson
I thought I understand C++ well, because I know boost, and I have read effective and more effective C++, and I know loki, etc. But when I open this book, I hope I have not said those to others:).
Those who thought Java/.NET solved the pitfalls of C++ should also read it, then they'll learn that Java/.NET is also far away from perfect.

代码大全（第二版）（英文影印版） 7-121-03362-3 Steve McConnell
I have read it, but it is nice to keep one for the convenience of communication with others.
Python编程（第三版·英文影印版）（上、下册） Programming Python, 3e Mark Lutz Introduced python2.5
Python seems popular, unfortunately it is beaten by lua and ruby during comparison, but I am not sure how precise those report are.
http://lua-users.org/wiki/LuaVersusPython

The Mythical Man - Month
The author mentioned cost of communication, but his mind is still waterfall.
Architecture worked out design and provide implementation suggestion, then hire
programmers to implement it.
"The Project Workbook" is proposed by the author for team communication, while according to my experience most team members are not familiar to documentation, even though all are required to review and give comments.
Chapter 8 Calling the Shot provided excellent statistics data on project estimation. I agree with them, but management team has no time to read this book, thus it was difficult to convince them. Most of the time, quality is sacrificed, although nobody will acknowledge the fact.

Chapter 9 Ten Pounds in a Five-Pound Sack : This does not apply to PC software development since memory is not a bottle-neck any more. Most of the time we are facing implementation efforts and efficience compromise.

There are a lot of dilemas in software engineering practice: The code of early stage is of poor quality, when they are reviewed, the author receives a lot of comments, but most of the time the code will be commited to main branch before "smell" code has  been revised. Once it is commited, it is out of the control of developer, so we have bad code in main branch. We can avoid this IN THEORY -- code must pass code review and QA verify before committing, that is THEORY.

The core of software engineering process is not methodology, it is execution and process control! Most people believes that a good system can automatically leads to good result.

Service Oriented Architecture
A Planning and implementation Guide For Business and Technology
This is not a book for Engieers, it is a book best for sales and people that talk more that do something. After a glance of it, I don't get much idea from it. SOA is a prevail trendency at present, the only advantage in my mind is its popularity and interoperability. Every technology advocator will claim their technology saves your cost, improve ROI, etc. It's better to know it and understand it before accept it.

I want to have a turtorial on XSLT, but this book is not the right one. It provided a lot of small scripts, I think those little tips can be easily found in internet by google.

Found a good book, "Patterns of Enterprise Application Architecture" by Martin Fowler. It talks of real world design solution. Another book, "Patter-Oriented Analysis and Design Composing Patterns to Design Software Systems" by "Sherif M. Yacoub and Hany H. Ammar" is a little bit too abstract. I wanted to buy some books on XML and corba, but I didn't found what I want.