Date: May 28 1999
Currently, I own UNIX in a Nutshell, Essential System Administration, TCP/IP Network Administration, Programming Perl, and Practical C Programming. All of these books are respected works in their genres, and are so because they're very well written, and because they're the kind of books you don't see many people writing these days. Most TCP/IP books I've seen are test preparation guides, and C++ has mostly overtaken the C book market.
Even most of your books on Oracle and Windows like the books on PL/SQL programming, Win32 threads, DNS on Windows NT, and others offer wonderfully detailed insight into subjects that aren't covered much elsewhere.
So, imagine my surprise when I perused your site for new books and found books on subjects like DCOM and Oracle Database Administration. These are subjects that already have books written about them that come right from the source (Microsoft and Oracle Press). I haven't had a chance to thumb through these books, so I'm curious: Why should I buy the O'Reilly books? How are they different from the already established sources?
Mind you, I only mentioned these as examples. There are probably more on your site that fit that description, but those are ones that jumped out at me. I have particular interest in buying a book on Oracle administration, but I haven't decided yet.
I appreciate any response you could give me, public or otherwise.
I'm not personally familiar with what's in the Microsoft Press or Oracle Press books that you mention above, but I can tell you in a general way what distinguishes O'Reilly books from "official" books from the software publishers:
I actually heard a cute story from one of our bookstores: a Microsoft Press representative told them that O'Reilly had an "unfair competitive advantage over Microsoft, because we could tell the truth about their products." Because it's Microsoft, whom everyone loves to bash, we can all chuckle and leave it to that. But I think there's more to this story than that.
One of the reasons that I started O'Reilly was because as a tech writer doing contract work for companies, I couldn't always tell the truth. I'd want to write about bugs and problems so I could offer workarounds, and the company would say "you can't say that about our product!" So when we started publishing our own books, I made that a centerpiece of our philosophy.
I remember talking with one author, who told me he couldn't write one chapter of his book, because the software didn't work as advertised. I told him that that was exactly what he had to write. Part of the job of a book is to do the work for the reader, to find the problems (and to the extent possible, solutions to them) before the reader does. And even if you can't find a solution, it helps to tell readers that something doesn't work so they don't waste time, or beat themselves over the head for being stupid.
In this regard, I'll take the opportunity to plug a book I worked on myself recently (the first in years), Windows 95 in a Nutshell (and its successor, the soon-to-be-released Windows 98 in a Nutshell.) There are obviously hundreds of books on Windows 95 and Windows 98, including official ones from Microsoft Press, but I will guarantee you that NONE of them contain as much useful information as the In a Nutshell.