[redp]

no flame, only facts
3 edition - 600 pages, sep 2000. w2k was released in 2000. Actually this was only edition which I completed to read and it was great, well-timed and well - fresh

4 edition - 976 pages, dec 2004. xp was released in 2001. So this book was written just too late

5 edition - 1232 pages, jun 2009. vista was released in 2006. I tried to read this book but
- it`s again too late
- most of I wanted to know was in "Writing Secure Code for Windows Vista" 2007
So I readed only 15-20 pages or so and abandoned this hopeless activity

6 edition - 752 + 640 = 1392 pages. w7 was released in 2009

So since 3rd edition delay is suspiciously the same - 3 year. To write 200 pages. Wonderful

[newgre]

no flame, only facts
3 edition - 600 pages, sep 2000. w2k was released in 2000. Actually this was only edition which I completed to read and it was great, well-timed and well - fresh

4 edition - 976 pages, dec 2004. xp was released in 2001. So this book was written just too late

5 edition - 1232 pages, jun 2009. vista was released in 2006. I tried to read this book but
- it`s again too late
- most of I wanted to know was in "Writing Secure Code for Windows Vista" 2007
So I readed only 15-20 pages or so and abandoned this hopeless activity

6 edition - 752 + 640 = 1392 pages. w7 was released in 2009

So since 3rd edition delay is suspiciously the same - 3 year. To write 200 pages. Wonderful

So you basically are disappointed with how quickly the book reflects the currently shipping OS kernels and their internals. I somewhat agree with that, the 6th edition seemed to take forever.
However, this is a fundamentally different statement in comparison to your previous statement where you complained about no new pages being added.

[redp]

However, this is a fundamentally different statement in comparison to your previous statement where you complained about no new pages being added.

1392 - 1232 = 160 new pages
160 / 1392 = 11.5%
I think this is almost undistinguished from term "nothing"

[Buster_BSA]

Well, the book has the cover labeled with "6th edition". Do not know in other countries, but here in Spain a "x edition" is usually just a re-edition of the book due the prior editions were sold so a new impression is required to continue selling the book.

So in my opinion a "x edition" is not a new book, it is just a reedition of the original.

If you see the editions like I do, then adding new material to new editions is just a good thing.

If you you do not see it like this, then I guess you consider the book should be named with a different title every time a new edition appears.

[aionescu]

Without disrespect to redp's opinion (which he is entitled to):

1) Each edition removes content as well, so a simple page diff is not enough. Also, Part 2 is not done yet, so it is not yet accurate to make the computation. I would say about 200 new pages each edition since 4->5->6.

While you might think 200 pages in 2-3 years is "slow", keep in mind the level of research that is required to figure out the minute changes between each release. For example, go see the scheduling section and compare it to older releases... Or read the new ALPC section.

There are also the illustrations (notice most were re-done/new), the technical review process (takes months), the copy editing, and more. As the book grows, this takes exponentially longer (they do not simply re-publish the old 1200 pages and only look at the new 200 -- they start from zero each time).

As others said, much of Windows does not change, so of course 80% of the book is "old" content. This is "by design". But: we do actually go back on the old experiments and re-do them, take new screenshots, etc, all of which takes time.

And finally, none of us are full-time book authors -- you probably know Mark is one of the highest positions at Microsoft now, and I work full-time on other things. Compare this to those who only write books without any real-life job (hint: most authors are not developers, and most developers are not authors -- this book is probably one of the only exceptions).

2) "Writing Secure Code" is an excellent book, but if it gave you the "same" information as Windows Internals, I would say you were probably not the target audience of WI.

3) Anyone who thinks there is money to be made in the Windows Internals book business, especially for the authors (don't know about publisher), really needs a reality check. Based on Amazon sales rank, you can guesstimate the number of sales of the 5th Edition. You can then use public information on royalties usually given to authors. You can then divide that by the number of authors.

[redp]

While you might think 200 pages in 2-3 years is "slow", keep in mind the level of research that is required to figure out the minute changes between each release. For example, go see the scheduling section and compare it to older releases... Or read the new ALPC section.

It seems highly strange that MS employees with access to windows source code/early internal alpha versions claims how hard windows re is (just like mere mortals :) )

There are also the illustrations (notice most were re-done/new)

This can be easy automated

2) "Writing Secure Code" is an excellent book, but if it gave you the "same" information as Windows Internals, I would say you were probably not the target audience of WI.

Sure. Thank you for your advice :twisted:

Also it seems that you are real Ionescu, so I want to ask - why you abandoned your cool blog ?

[rkhunter]

It seems highly strange that MS employees with access to windows source code/early internal alpha versions claims how hard windows re is (just like mere mortals :) )

Are you sure that this is easy task - writing book about Windows Internals in collaboration with MS?

[redp]

Are you sure that this is easy task - writing book about Windows Internals in collaboration with MS?

I never tried
Although I naive assume that it`s much simply to see some aspects in source code vs several hours in windbg/disasm
But I also saw lots of source code which much better to understand in ida pro (including open source) :twisted:

[rkhunter]

But I also saw lots of source code which much better to understand in ida pro

Ha-ha-ha, I hope that you mean not WRK. Because without it amount of work grows by an order.

[aionescu]

While you might think 200 pages in 2-3 years is "slow", keep in mind the level of research that is required to figure out the minute changes between each release. For example, go see the scheduling section and compare it to older releases... Or read the new ALPC section.

It seems highly strange that MS employees with access to windows source code/early internal alpha versions claims how hard windows re is (just like mere mortals :) )

I am not an MS employee, and never have been (I actually worked at Apple for 4 years...). I also do not have source code access, and never have had it. I also don't have access to "alpha" versions other than official preview builds... I read Neowin like everyone else to see what's going to be new in Windows 8/9/etc, and download the same DP/CP builds everyone else downloads. So try looking at KeWaitForMultipleObjects (win7 sp1) in IDA

There are also the illustrations (notice most were re-done/new)

This can be easy automated

The book has a whole illustration team and I'm not sure they'd like to think they could be "automated"... if you really have a solution for this, you should patent/publish it, because I think book publishers around the world would love to know this. In either case, this is outside of the author's control. MS Press decides that in 2011, their diagrams will have a new "style", they will spend the time to re-do all of them, regardless of what we think.

2) "Writing Secure Code" is an excellent book, but if it gave you the "same" information as Windows Internals, I would say you were probably not the target audience of WI.

Sure. Thank you for your advice :twisted:

Also it seems that you are real Ionescu, so I want to ask - why you abandoned your cool blog ?

Lack of time -- my free time has all been spent on the book updates. I hope to get back to it as soon as Part 2 is done. Thanks for the compliment, and likewise, almost half of my Google searches end up on your blog these days (Google Translate is helpful) and I've become a regular reader (also love wincheck).

--
Best regards,
Alex Ionescu