Wednesday, May 25, 2005

On Writing

Some recent questions have been about writing, the process of writing and the benefit. I thought I’d take a look at them today

To the people who have written books... has this helped your career in anyway?

Others may post their feedback in the comments section, but for me — I’d have to say definitely yes for me personally but your mileage may vary. For me personally, it definitely increased my own visibility. In that regards, it did not “help” the career so much as “send it in a direction”. That point is subtle, the writing of a book changed my path. Is that good? Well, for me in hindsight, absolutely. I really enjoy what I get to do now. The books established a level of credibility, not just because I wrote them — but for what they contained. It is not just the very act of writing that makes it credible, it is the act of writing credible material that can be acted on.

But, one of the reasons for writing the first book was to see if I could. It did come with a certain sense of achievement after it was finished. I used to joke that my life had three phases — BB, DB and AB. Before Book, During Book, After Book. Just doing it, as part of the career itself, was a challenge.

All in all, if the material produced is quality material, I cannot see anyway this would hurt your career (as long as you don’t under perform at work while doing it of course!). It does not hurt the resume, it could lead to other opportunities, and it will expose your name to a broad range of people that might not otherwise have heard of you.

How much influence does the publisher have on the book?

This will vary by publisher I’m sure, and your current level of exposure. The process typically goes like this, you submit an outline and a concept. A little back and forth, maybe some market research, a lot of “gut feel” and a refinement here and there and you get accepted or not. So, now you have an outline — an outline isn’t very confining at all.

So you are free to write whatever you want, but the publisher will be sending it out to a technical review team for commentary. (I still feel sorry for a certain Java programmer who was trying to write a database book and had me as a technical reviewer — they did not get to write what they wanted, how they wanted — but I do think they learned something new about the database by the time they were done). The technical review team, as I’ve said before, if they are good will rip pieces of the work to shreds (that is their job). They probably have more influence than the publisher — simply because they are supposed to be subject matter experts in the field the book is on, whereas the publisher (editor) might not be.

The editor I work with knows quite a bit about Oracle now — but when we started, he didn’t. Still, he has never written code for money, so he has somewhat of an academic background in the technology only. That is why I rely on people that have written code, managed databases for money (it is their way of making a living) to review my material. They have a huge say in what is ultimately said in the book and how it is said.

But the publisher will want the general outline to be followed, the page counts to be what was estimated (that is hard, I just put “35” as the page count for every chapter myself — and hope some come up short to balance the ones that come up long)

How much say do the publishers have on deadlines?

Well, you sign a contract with them and the contract typically includes a delivery schedule. However… Well, you know how it goes. Basically, if you are the sole author they are sort of reliant on you to deliver and you can only type so fast. Now, they generally have clauses in there that would allow them to bring in other authors to finish or help — but you would have to be extremely late in order to have that kick in.

But you get to negotiate the time frame up front. They do not set it, you negotiate it. If you are not happy with it, you have the option of walking away. Don’t sign anything until you agree with it. Have a lawyer look at it (I did not, I was naive about it back in the beginning, a lawyer would have changed the wording of a thing or two and with WROX going out of business and the contract being sold — I would have liked the wording to have been changed. Hindsight is 20/20 though). And if the contract has an option for future works, meaning they have the right to publish your next 1 or 2 books, get that stricken from the contract immediately. Don’t even go there, don’t let them do that to you.

About sole authorship — I would recommend that if possible. It makes a more cohesive book all in all. The book I worked on with Sean Dillon and Chris Beck worked out so well only because we all sat within 20 feet of each other at work and play poker together. On a multi-author book, where the publisher is assembling the team, you might find the authors to be spread throughout the world, with radically different opinions on the best way to accomplish something. That leads to a book where chapter 4 contradicts what chapter 15 says to do and so on. And the different writing styles, examples, and so on will drive the reader nuts.

One question just for Tom, do you have any control over your Expert one on one book or do you sign over full control to the publisher when you write it?

I have a fair degree of input. Control is too strong a word, they own the “rights” to it, but I have lots of “artistic input”. Take the two volume approach — my suggestion (recently, after seeing near zero percent chance of finishing the entire thing before the fall). Their approval. The CD-ROM of the 8i book, my suggestion, their implementation. The content of the 9i/10g release — all about me, they cannot make me write that which I do not desire to write about.

So far, I’ve found (even when just beginning) the publisher tries to work with you. They do not necessarily have an agenda, they are somewhat reliant on us out here to know what is needed in the market to a degree.

What in your opinion is the hardest thing to do if you are a writer?

Besides write? For me, it is getting it started. On a thread in asktom, someone was asking how to get really good at SQL. They were looking for ways to practice. My response was “I cannot ‘practice’ SQL, or C, or PL/SQL, or PL/I or JCL or SAS or any of the technologies really”. I need a real problem to solve. Making up an excuse to use something is really hard. So, in writing the book — coming up with the proper examples is perhaps the hardest part. It has to be real, but not too real (you cannot print someone's application, they get upset). So it is the “making it real” without faking it. None of the examples are faked, they are all real and run on real machines with real CPU’s (singular and plural). But coming up with something that represents a real world situation takes time. And it takes review — have to let the reviewers poke holes in it.

For example, while doing the update of Expert One on One Oracle, I was doing the chapter on memory structures. I had a couple of points I wanted to make. Two of them were

  • in 9i, automatic PGA memory management cannot take place with shared server connections
  • the “A” stands for automatic, the algorithms are not really documented, but the gist is “as the number of users needing workareas goes up, the size of the workareas go down”

At first, my example showed that shared server could not, did not use automatic PGA memory management and that the setting of sort area size and so on was very important still for these connections. And I just stated “as the load goes up, memory allocated to each session goes down”. I wasn’t entirely happy with the chapter for some reason, just a nagging feeling so I had Tony (the editor) read it and suggest something. His idea was “I believe you when you say shared server won’t use it, the example doesn’t really provide much benefit beyond showing that fact, but I don’t understand the memory allocated goes down as the user load goes up — that would be a much more useful example”.

Unfortunately, he was right (and it was obvious), so a rewrite was called for. That would have been enough right? No, not really, the reviewers got the first cut of the example and I had to change it yet again.

So, I guess it is the getting started and getting it refined — but definitely, getting it started. What is going to grab someone, what is the most important thing you want to say, what do you want them to take home with them and think about.



Blogger Alberto Dell'Era said....

And do you feel now as motivated as when you were writing the 1st edition - ie the 1st Edition was a challenge, your "opera prima"; what's this new book to you ?

Wed May 25, 12:44:00 PM EDT  

Blogger Richard Byrom said....

Thanks for writing this post Tom - it gives an interesting insight in to the process of writing books. What I would like to ask is was the book writing process financially rewarding and I mean purely from the money generated from sales (not in terms of the book opening up new career opportunities which is a given). Would certainly like to know more about the financial side i.e. what was your cut in percentage terms after other costs like paying off the publisher etc and would you consider "self publishing" as a better route to take if you wanna make some dosh.

Most of the people I have spoken to seem to think that writing books is not something that's going to make you wealthy from book sales alone.

Added this post to


Richard Byrom

Wed May 25, 12:59:00 PM EDT  

Blogger Joel Garry said....

It is not just the very act of writing that makes it credible, it is the act of writing credible and actionable material.

Hmmm, a dictionary defines actionable as Giving cause for legal action.

No wonder some people are so pugnacious online regarding some authors :-)

Wed May 25, 02:15:00 PM EDT  

Blogger Thomas Kyte said....


I see, so it does. I've abused that term here. I'll have to change my use of it in the future (any ideas for a good replacement word?). I updated the blog to reflect that -- it was misleading.

What I was trying to imply is instead of giving broad, sweeping generalizations that may or may not apply -- give advice that one could take action on, that you can act on, that when you read it -- you know what to do and when to do it.

Using the book metaphor, I reject reviews from people that say "I didn't like it". So what? Why? All of it? Some of it? One little bit of it? What do I need to change? How can I change it? Give me information that I can act on.

Don't tell me "materialized views can make queries go super fast when you have a slow query".

Do tell me

o this is what a materialized view is
o this is how they work
o this is what they are designed to solve
o this is a list of considerations you need to be aware of when implementing them
o this is how to see if they are being useful
o here is a set of times that they are quite simply not appropriate
o here are the caveats you must be aware of when using them.

The first one is a broad sweeping generalization (and not a made up on, one I just read recently).

The second actually gives you information you can act on to decide if a materialize view might be useful in your case.

Wed May 25, 02:28:00 PM EDT  

Blogger Alberto Dell'Era said....

any ideas for a good replacement word?

Practically Effective ?

Wed May 25, 02:41:00 PM EDT  

Anonymous Anonymous said....

"Practically Effective"
-> "Practical"

Wed May 25, 02:57:00 PM EDT  

Anonymous Anonymous said....

Cool blog entry

Wed May 25, 03:21:00 PM EDT  

Anonymous denni50 said.... about?

"Simulated behavior/actions".

anyway, here's my commentary on your books and writing:

Good insight on the do's and don't's of writing a book and what your experience has been from novice to 'not so' novice....however I gather most of us will never endeavor in such a possibility.

you wrote:
"The books established a level of credibility,not just because I wrote them — but for what they contained. It is not just the very act of writing that makes it credible, it is the act of writing credible and actionable material."

(first and foremost you are being way too modest...without you these books would never have come into is because YOU WROTE them these books are what they are).

Those words sum it up! many others can claim the same conviction with their books. Not only does your wealth of knowledge and vast experience shine through, even more is your determined precision for technical truth and integrity.

You may not have reaped enormous financial gain from your books,but you have earned the utmost respect, admiration and 'fame' from your peers and followers from all walks of the Oracle world.

Your invaluable efforts and contributions have had an enormous impact on the Oracle Community especially for those in my category(novice and 'not so' novice)dealing with the daily challenges of managing such a complex system.

Enjoy the rewards of your labor, in whatever shape or form they are derived,you definitely shine in the technical limelight!

waiting impatiently for the new edition.


Wed May 25, 03:29:00 PM EDT  

Blogger Alberto Dell'Era said....

... your determined precision for technical truth and integrity.

You may not have reaped enormous financial gain from your books,but you have earned the utmost respect, admiration and 'fame' from your peers and followers from all walks of the Oracle world.

I agree 110% with denni50.

You may stop here, and I'm sure many of us will remember you until retirement and beyond :)

Wed May 25, 03:54:00 PM EDT  

Anonymous Doug said....

Tom said ...

"Well, you sign a contract with them and the contract typically includes a delivery schedule. However… Well, you know how it goes."

It reminds me of that Douglas Adams quote

Wed May 25, 04:00:00 PM EDT  

Anonymous Anonymous said....


Why should I buy your books (or books from any other author)? Can't I learn everything in your books from the Oracle documentation?

Wed May 25, 04:08:00 PM EDT  

Anonymous denni50 said....

hey Alberto...(chuckle!)

have no plans to retire anytime soon...given the situation with social security in this country I see myself working until I bite
the dust.


Wed May 25, 04:35:00 PM EDT  

Blogger Thomas Kyte said....

Why should I buy your books

You absolutely can learn it all from

o the documentation
o experimentation
o implementation
o education
o reading the theory
o working with others with experience

Absolutely. I learned that way. But I did read a couple of books on the way -- lots of them.

You won't see bind variables stressed as vigorously in the documentation as you will in my book :)

But your question begs the question "why buy any books on any subject, if the subject comes with a manual".

To get perspective perhaps. Learning that recovery is the most important thing a DBA needs be able to do and what happens if they don't test it, play with it, perform experiments (what happens if ...) is something you will not see in the documentation. Or at least not stressed as passionately.

I guess you get them to get some of the experiences of the person writing them.

But yes, you absolutely could get everything you need from

o the documentation
o experimentation
o implementation
o education
o reading the theory
o working with others with experience

but you might consider books to be just like all of the points after "the documentation"

Wed May 25, 06:22:00 PM EDT  

Anonymous Paul said....

Why should I buy your books?

... I read lots of what could be called technical books about bridge (the card game). I prefer specific authors because the subject matter that they cover is worth knowing about and they also write in a pleasing manner. Most of what Tom writes has little relevance to me, but it's the way it's presented to the reader that makes me go to asktom ever day. (Yes, I've got Tom's books too).

Thu May 26, 03:32:00 AM EDT  

Blogger Tom Best said....

Wow. That is very interesting. Paul posted that your books have little to do with his interests, but he just likes the way you write, so he reads them, and asktom! I've never heard of anyone reading something they don't care about just because of how it is written, except for maybe poetry.

Thu May 26, 08:52:00 AM EDT  

Anonymous William said....

From Paul's comment, I understood him to mean that Tom's knowledge of the Oracle domain is wide, and Paul's own areas of involvement much narrower. Therefore much of Tom's writing is of little relevance to his day-to-day work. Same goes for me, I don't have to worry much about recovery, but I deal with lots of SQL queries, table structures and indexes. But I still use AstTom to broaden my perceptions and understanding beyond my own requirements. Sometimes you need to know why a production DBA sees your "good idea" as a resource hog, and AskTom helps me with those insights.

So it's all useful Tom: keep doing it!

Thu May 26, 10:14:00 AM EDT  

Anonymous Anonymous said....


This is a bit off top, but what exactly do you do at Oracle? You appear to be the technical face of the company and spend alot of time giving presentations and explaining how the database works.

Do you manage software development projects? Do you design systems? Or do you manage an Oracle sector? You seem to be fairly different than most VPs. Most VP jobs are not really technical positions. They are more big picture and running an oraganization type positions.

How big is your organization?

One other thing, have you ever met Larry Ellison?

Hope it's ok to ask this...


Thu May 26, 12:07:00 PM EDT  

Blogger Connor McDonald said....

except for maybe poetry

(true story)

I once had a view that was materialized,
On top of a DUAL that was index organized.
But alas I was unable,
To fast refresh enable,
Until DUAL was as a heap reorganized


Fri May 27, 07:41:00 AM EDT  

Blogger Thomas Kyte said....

Connor -- keep the day job :)

Fri May 27, 07:53:00 AM EDT  

Anonymous Resident Poet said....

Feel free to delete if it is consider in poor taste, but I could not resist after Connor's comment:

I once knew a certain DBA
Whose Rules of Thumb lead him astray.
And try as he might,
He could not get it right -
His arrogance got in his way.

Fri May 27, 02:05:00 PM EDT  


<< Home