发表于2004/4/15 19:52:00 2969人阅读
History & Company Overview
By Tim O'Reilly
In the Beginning
We've been in business since 1978, originally as a technical writing consulting company. In 1984, we recognized the possibilities in open systems and started retaining rights to manuals we created for Unix vendors. Our thought at the time was simply to license the books to other vendors, but in the second half of 1985, a sudden slowdown in our consulting business led us to try publishing some of our material as independent books. We thought we might even just give the books away to promote our consulting business. Our first titles were really more pamphlets than books--an average of 70 pages. We called them Nutshell Handbooks?, because we were trying to get down to the basics--just what you need to know, in as few pages as possible.
The first books were well received, but our consulting business revived, so we kept at publishing "in the cracks"--whenever any of our writers had downtime between projects. We did it because it was fun, not because we thought it would grow into the major business that it has become.
Then, in January of 1988, at the MIT X Conference, we happened to show some drafts of the Xlib manuals that we were preparing for two of our clients. We were mobbed! We had planned to license the books to vendors as documentation, but it quickly became clear that there was a huge market for them as standalone books.
An Uncommon Publisher
Somewhere in there, we realized that publishing was a far more interesting business than documentation consulting. We devoted more and more resources to it, until publishing became the core of our business. We currently publish more than 300 titles, employ over 300 people, and have offices in the
Our background in the computer business, rather than traditional publishing, has given us a very different approach than most computer book publishers. All of our editors are expected to get their hands dirty with the technology we publish about. Many are former programmers, system administrators, technical writers, or practicing scientists, and all are expected to have written at least one successful book of their own. Because we're close to the industry, we know what books are really needed, and we make sure they tell people what they really need to know.
Here's something I wrote for our catalog a few years ago, under the title "Myths about Computer Books," that tells a little bit more about our publishing approach. . . .
You don't want our books--you want the information they provide.
We really believe this, so we've thrown out some of the assumptions that go into most books and documentation:
·???????? A book has to be a certain length. We write books that are as long as the subjects they cover, ranging from the 78-page CVS Pocket Reference to Practical UNIX and Internet Security, which weighs in at a thousand pages.
·???????? You can't say bad things about software that doesn't work. Nonsense. We feel free to editorialize. Our goal is to talk to you directly in our books, and if that means giving you hacks and workarounds to software problems, we'll do it. If we've done our job, reading one of our books should be like having an experienced user by your side, passing on helpful hints whenever you get stuck.
·???????? Books are not a replacement for documentation. We try to write books that are as complete as documentation, but as interesting and readable as commercial books. Once you buy one of our books, we hope that you never have to look at the documentation again, except for system-specific options that aren't industry-standard. And whenever possible, we'll try to cover them, too.
·???????? Books can't keep up with the pace of change in the software market. We update our books frequently, often making small changes every time we reprint. And we keep our print runs short so that we have the opportunity to revise every six months. One of our early books went through ten editions in less than five years. Many of the changes are in response to feedback from readers as well as software changes.
·???????? There's always room for another book. Most publishers compete with themselves by putting out lots of books on the same subject, figuring that whichever one wins out, they'll do all right (even if the author suffers). We try to find subjects no one else has touched (where users are hurting for information), and we do only one book on each subject.
First, a little history. Because so many vendors used our X books as documentation, we had many requests back in the 80s to provide the books in various online formats--Sun's AnswerBook, IBM's InfoExplorer, or HP's LaserROM. This was perhaps a good sales opportunity, but maintaining our books in a lot of different formats seemed like a pretty uninteresting business. We realized that if online publishing was really to succeed, for us or for anyone, we needed to develop a common interchange format for online books. Publishers could then stay in the business of providing information, and leave it to the vendors to display the common format with their proprietary tools.
In 1991--along with a group of forward-thinking vendors, including Digital Equipment Corp, Hal Computer Systems, and Silicon Graphics--we founded the Davenport Group to explore online publishing issues. The fruit of that work, the DocBook DTD for SGML, has been adopted as a de-facto industry standard by many of the most important vendors in the open systems market (and in 1999, as SGML spawned XML, we published DocBook: The Definitive Guide, written by former O'Reilly employee Norm Walsh and our production tools manager, Lenny Muellner).
At this same time, we were becoming increasingly aware of the power of the Internet. In 1993, before the Mosaic Web browser had been released, we discovered the World Wide Web. The most exciting of the new Internet applications, the Web was based on a client-server hypertext technology that used HTML, a dialect of SGML, as its data format.
As it turned out, our first online product was not an electronic version of our X Window System series online, but a version of the Internet catalog from Ed Krol's The Whole Internet User's Guide & Catalog. It started out as a demo but quickly grew into a revolutionary product, hailed by Wired magazine as an historic event in the information age. That product, the Global Network Navigator, or GNN, was the first Web portal and the first Web site ever to be supported by advertising. We conceived GNN as an "information interface" to the Internet, a point-and-click framework in which reviews, articles, and news bulletins about Internet services become a door to the services themselves. GNN was one of the first Web sites--in fact, we had to dig deep to find the 300 Web sites in the first online version of the Catalog.
In late 1993 (eons ago in Internet time), we realized that people who were excited about GNN couldn't get to it easily. They'd say, "This is so cool. How do we get it?" And the answer was a long story, involving instructions for getting on the Internet, downloading software, and finally getting to use the Web. We realized we needed a one-stop solution. We teamed up with Spry, a Seattle-based software company, to create the first integrated Internet-access product, Internet in a Box. This was a combined software and information product, including Spry software, GNN, and a custom version of The Whole Internet.
Before long, the Internet caught up with our vision. Internet access providers sprung up like mushrooms, and the online services got into the game. We sold GNN to America Online, and Spry was sold to CompuServe.
We've continued our forays into online publishing, as we (and our customers) turned more often to the Web for technology information. After GNN, we created WebReview.com, a technology-focused site that we sold to Miller Freeman in 1999. Dale Dougherty, the founder of GNN and WebReview.com, has turned his talents to the O'Reilly Network, a portal for developers that focuses on open and emerging technologies. With sites including XML.com, Perl.com, and OpenP2P.com, the O'Reilly Network covers important new technologies in the trademark O'Reilly style--independent, in-depth, and steeped in the experience of those on the "bleeding edge."
In addition to developments at O'Reilly itself, we've also been involved in starting a number of other companies to exploit technologies we thought were interesting but that weren't part of our core mission of delivering information on leading-edge technologies. In 1996, we launched Movie Critic, a site that featured one of the first implementations of collaborative filtering. In true Internet Age fashion, we spun Movie Critic out into a collaborative filtering company, LikeMinds, which later merged with Andromedia, which was then bought by Macromedia. In 1998, we helped found ActiveState in order to make Perl more accessible to Windows users. ActiveState's ASPN product line now provides tools, support, and services for Perl, Python, and XSLT, and Visual Studio plug-ins for these languages so that they can be used with Microsoft's new .Net framework. In 1999, we started CollabNet along with Apache Group cofounder Brian Behlendorf. CollabNet's mission is to take the collaborative software development methodologies pioneered by open source development projects and make them more accessible to corporate
Stirring the Pot
From the beginning, our editors, authors, and developers have been active members of the technical communities whose work we chronicle in our books and Web sites. Over the years, we've extended our support of those communities beyond our publishing program through activism and conferences.
In April 1998, we hosted the first Open Source Summit. This event brought together leaders of many of the significant open source communities, including Linux, Apache, Tcl, Python, Perl, and Mozilla. It was the first time most of the participants had met in person. The
When we published the perennially bestselling Programming Perl back in 1991, we helped to legitimize the language in the eyes of corporate developers. As part of our campaign to support the Perl community, we produced the first-ever conference on "the duct tape of the Internet" in 1997. A few years later, we added conferences on several other important open source technologies, and the Open Source Convention was born. We realized that the people who read our books also want to connect with and learn from each other, so we moved full bore into the conference business. As with our publishing program, our conferences focus on practical, in-depth information taught by those who've mastered (and in many cases, created) high-end technologies. And our conference offerings have expanded to cover key emerging technologies. Our first Peer-to-Peer Conference in February 2001 was a sellout, and we're planning a new Bioinformatics Conference for early 2002.
The Common Thread
So what binds our efforts together? At the core, we create products that we want to use. Whatever form it takes--book, conference, online product--we want anything produced with the O'Reilly name to be useful, interesting, and truthful. And we believe that there are plenty of intelligent, discriminating people in the world who value those qualities as deeply as we do.
Tim O'Reilly is founder and president of O'Reilly & Associates, thought by many to be the best computer book publisher in the world. In addition to publishing pioneering books like Ed Krol's The Whole Internet User's Guide & Catalog (selected by the New York Public Library as one of the most significant books of the twentieth century), O'Reilly has also been a pioneer in the popularization of the Internet. O'Reilly's Global Network Navigator site (GNN, which was sold to America Online in September 1995) was the first Web portal and the first true commercial site on the World Wide Web.
Tim has been an activist for internet standards and for Open Source software. He has led successful public relations campaigns on behalf of key internet technologies, helping to block Microsoft's 1996 limits on TCP/IP in NT Workstation, organizing the " summit" of key free software leaders where the term "Open Source" was first widely agreed upon, and, most recently, organizing a series of protests against frivolous software patents. Tim received Infoworld's Industry Achievement Award in 1998 for his advocacy on behalf of the Open Source community.
Tim has written numerous books on computer topics, most notably UNIX Text Processing (with Dale Dougherty; Howard Sams, 1987), Managing UUCP and USENET (with Grace Todino; no longer in print), The X Window System Users' Guide (with Valerie Quercia), and The X Toolkit Intrinsics Programming Manual (with Adrian Nye), UNIX Power Tools (with Jerry Peek and Mike Loukides) and Windows XP in a Nutshell (with David Karp and Troy Mott).
As an O'Reilly & Associates editor he has had a major hand in the development of many of the company's other titles, including UNIX in a Nutshell, Programming Perl, Sendmail, Essential System Administration, and The Cathedral and the Bazaar.
Tim also conceived an award-winning series of travel books, published by O'Reilly affiliate Travelers' Tales. His company's Patient Centered Guides provide information from "health system hackers"--patient advocates who have experienced the best and worst of what the medical system has to offer, and pass along their experience for sufferers of chronic or life-changing diseases.
Tim has served on the board of trustees for both the Internet Society and the Electronic Frontier Foundation, two organizations devoted to making sure that the Internet fulfills its promise. He is on the boards of CollabNet, Macromedia, the Open Source Applications Foundation, and the Nutch Organization.
Tim graduated from Harvard College in 1975 with a B.A. cum laude in Classics（Tim是学古典文学的优等毕业生J）. His honors thesis explored the tension between mysticism and logic in Plato's dialogues.
An archive of Tim's online articles, talks, and interviews can be found at Tim's archive page
- Caffe 环境搭建中应注意的问题
- Makefile:588: recipe for target '.build_release/cuda/src/caffe/layers/softmax_loss_layer.o' failed
- caffe-fast-rcnn 错误解决途径
- make: execvp: /usr/local/cuda: 权限不够 Makefile:588: recipe for target '.build_release/cuda/src/caffe/l
- 很感谢这篇文张【已解决】Makefile:xxx: recipe for target xxx failed +【顺带搞懂了】如何忽略makefile执行过程中的某些命令的错误而得以继续运行
- Caffe 工程的一些编译错误以及解决方案
- 编译caffe出现错误：make: *** [.build_release/src/caffe/common.o] Error 1
- 编译caffe出现错误：make: *** [.build_release/src/caffe/common.o] Error 1