November 2003
From: Paul Murphy
Subject: Are "how to" books archaic?
At present, your focus appears to remain on "how to" books. Some of those are quite good; I have six here. But growth in that group has got to be coming to an end as the rate of change in technology continues to increase along with Internet access to free, and current, technical information.
So how do you see your market progressing? Do O'Reilly buyers become what they read, living encyclopedias of dated technical arcana? Or are you simply assuming that demographic replenishment will keep the company going?
I find your question a bit hard to parse. Do you think "how to" books aren't working because the life cycle of a book is out of step with the rate of change, leading people to online sources? Because, of course, the fact that technology is changing so fast means that people need more "how to" information all the time in order to keep up.
I agree very much with your premises, though. The rate of change is increasing. We're entering a period of greater change in technology than at any time in the past few decades. Everything we thought we knew is wrong, as we reach the end of the personal computer era and enter the network era. The Internet as we've known it was really just a networking add-on to the PC, but now we're starting to see the Internet itself emerge as the platform. There's a fundamental paradigm shift, as fundamental as the shift from earth-centric Ptolemaic astronomy to the solar-centric Copernican vision.
Paradigm failures abound. Linux critics say that Linux is not yet as easy to use as Windows or Mac OS X. Linux partisans point out the progress of OpenOffice, Gnome, and the Gimp. Neither one points to the great, easy-to-use applications of the Internet era--Google, amazon.com, PayPal, maps.yahoo.com--all of which run on Linux or equivalents (maps.yahoo.com runs on FreeBSD). The browser frontend may live on a PC, but the application lives somewhere in the space between the two.
Even more traditional PC applications no longer reside solely on a local device. Rip a music CD, and the artist and song names are automatically looked up on CDDB, an Internet site. This is the new face of web services--application components residing remotely.
Distributed peer-to-peer applications like Napster, Kazaa, and BitTorrent, or distributed computation applications like SETI@home, take the model even further, residing on millions of cooperating computers.
So yes, the rate of change is increasing. But this will lead to more need for "how to" books, not less. I believe that most developers haven't yet grasped just how completely the programming model is about to change. Of course, the leading developers have known this for a long time. But the great bulk of the market is just entering this uncharted territory.
That being said, I agree that books aren't always the answer. Particularly when things are moving very quickly, the development cycle of a book doesn't work. That's why we started publishing online in 1993, and why the O'Reilly Network is so core to our business.
What's more, as the new paradigm takes hold, I believe that there will be further opportunities to expand what we do online. Part of the long-term strategy behind Safari is once we have thousands of books online, we can build new kinds of services against what is now the most comprehensive technical information database available for computers and software.
Tim
Return to: Ask Tim
Showing messages 1 through 10 of 10.
-
what makes a good howto book
2003-11-12 13:31:15 anonymous2 [Reply | View]
What makes a quality technical book isn't simply how good the "how to" instructions it contains are but also how well the author is able to explain they the "why" (which is a lot harder to do!). Understanding (and therefore problem solving ability) comes from the discussions of "why" - the "howto" part provides the hands-on instruction. I've kept many a book whose code or command listings are technically obsolete because the author provided a fantastic exposition of the subject matter. I also find the latter much easier to read in book form than on a computer screen.
-
Books and other media
2003-11-12 12:24:32 paulmartin3 [Reply | View]
I believe that physical books will co-exist with the web and other media for some time to come. I would argue from an analogy which seems obvious to me, but which I haven't seen anywhere else.
In the scientific community (often cited as an example of open source at work) there are two primary sources of information: textbooks and the scientific literature itself. Textbooks, well known by students, are more easily accessible to the general public, and tend to contain larger quantities of tutorial material. The information tends to be dated, but when thousands of students take the same standard courses every year, this is not really a problem. In addition, there are obvious financial incentives to writing for this medium, a problem which has yet to be solved for the web.
The scientific literature, on the other hand, is more like the web. It is fast, immediate and current. Publishing is done to establish a reputation, rather than for financial rewards. The quality of the information here is mixed, but errors are spotted quickly enough for those who are adept at scanning the medium as a whole. This makes it appropriate for experts, but confusing for the neophyte. Without the financial incentives, it is a poorer medium for tutorial material.
The analogy isn't perfect, but I believe that both media will survive. There will likely be quite a bit of dynamic activity as each sorts out its niche in the coming years.
-
what's next for howto books?
2003-11-10 01:54:02 ramonf [Reply | View]
We're talking about knowledge absorption here, and how relevant information is selected and organised for this purpose.
So that information has a shorter shelf-life nowadays? True. That the rate of arrival of new information has increased dramatically? Also true. But this doesn't make "howto" books obsolete. I would have rather thought that it endangers NON-howto books actually.
For the exabytes of information being released nowadays, the need to distil knowledge into howto style is, if anything, more urgent. More tools and more complementary styles are needed for this purpose. At the risk of flattering, I think O'Reilly are doing sterling work in this direction.
-
what's next for howto books?
2003-11-10 14:19:48 Tim O'Reilly [Reply | View]
Thanks. Books like the Hacks series and the Cookbook series both try to use more of a web-style "useful bits" model rather than "single connected narrative." We think that's a good start towards giving people just what they want when they want it. Similarly, we've tried to do things with In a Nutshell and Pocket Guides/Pocket References to boil down essential reference info into the shortest possible compass for sophisticated users. Lots of experimentation to find the right style and approach.
-
How-to books & programming
2003-11-09 17:06:24 anonymous2 [Reply | View]
`. . . most developers haven't yet grasped
just how completely the programming model
is about to change.' Thus Tim O'Reilly.
I'm no programmer, but that remark leads
me to suspect I've missed something important.
Can you say a bit more? -
How-to books & programming
2004-06-12 19:18:41 David Lents [Reply | View]
This is a test -
How-to books & programming
2003-11-10 14:22:23 Tim O'Reilly [Reply | View]
As Dave Stutz remarked, "software above the level of a single device will command high margins for years to come." We're moving away from the era of software targeted to a single device into a world in which software routinely incorporates remote data sources via web services, may have an n-tier architecture in order to support multiple front ends on top of a robust server-based back end, etc. In short, I'm just pointing out that there's a huge amount to learn. You're not programming your father's PC any more, so to speak. Or even your father's web site. We're starting to program the net itself. -
Wow that was a lot of jargon
2005-10-13 10:21:57 GoClick [Reply | View]
Wow that was a lot of jargon and buzzwordary to cram into one paragraphy.
I think what he's trying to say is, we're not just making programs to run on people's computers anymore. There is a shift into making (generaly) web applications that take the place of those desktop style applications.
The main difference with these newer style applications is their more useful because their always up to date and they allow you to access your data and their functionality from any computer with a modern web browser.
There aren't just the web applications the end user sees either tho. There are the "behind the screens" applications that do things like harvest data (google news) that allow the developers to more efficiently interact with eachother which leads to more innovation. -
How-to books & programming
2003-11-10 11:21:35 Mary Hubben [Reply | View]
Tim talks about this in an eWeek interview at www.eweek.com/article2/0,4149,1187612,00.asp .
