Related link: http://www.zdnet.com/zdnn/stories/news/0,4586,2689187,00.html

Related link: http://www.zdnet.com/zdnn/stories/news/0,4586,2689187,00.html

While most of the world holds its breath, waiting to see what’s in store for Napster, a number of diehard MP3 fans have gone ahead and started their own “Open Napster” servers, using freely distributable software. As of this past week, the RIAA is now after them, too, via their ISPs. Can the recording industry bully Internet providers into shutting the Open Nap servers down, by making examples of a select few? Or will the RIAA ultimately find unable to slay this Hydra?

John Perry Barlow made a surprise appearance on the P2P conference stage this morning. He offered a number of pungent observations, including his oft-heard maxim that “Information is not a noun, but a verb.” If physical commodities are nouns and information is a verb, perhaps peer-to-peer is a preposition. It’s the ineluctable, nearly invisible line that ties everything around it together. And if you follow that metaphor, and look again at Dave Winer’s statement “The P in P2P is people,” then perhaps what’s really important in P2P is the 2.

We’ve had many high points in this all-too-brief conference (in fact, for me it’s been a three-day high), but the morning events that began with Larry Lessig’s keynote generated more energy by far than anything to date. Lessig immediately showed his deep caring for, and knowledge of, the various disciplines relevant to the social meaning of new technologies. Computing history, the technological dilemmas confounding current policy, and, of course, the laws of intellectual property he wove together expertly into an exhortation that, “The public should be aware of the extraordinary system of control that is being rammed down our throats in the name of the Constitution.”

What we’re losing is the public domain, the right to build on the work of others, and the subtle mesh of rights that copyright law calls “fair use.” These include educational uses, citation for the purposes of commentary or criticism, and (in technology) the ability to investigate a vendor’s product in order to hook into it or compete with it.

After a rousing ovation, Lessig was joined by Barlow, Clay Shirky, journalist Dan Gillmor, and Tim O’Reilly. The level of flighty rhetoric on all sides merely echoed the energy aroused in the audience. Often I’ve noted that people care about policy and demonstrate a desire to influence the political sphere, but rarely do they know how to harness and carry through that desire. Lessig and the panel provided that focus today.

The defiant tone cycled on into the morning session that followed. Titled “Business and Social Implications of Decentralized Systems,” it stayed firmly on the topic of overcoming censorship and allowing people to speak their minds (at least during the portion I attended). I had to miss part of the panel in order to see the one on industry standardization. While the topic was much more engineering oriented, the feeling of community pressure was just as strong. In fact, people were much more contentious around the question of technical standardization than they were around Lessig’s call to political action.

Intel is still getting flack for its attempt to organize the P2P Working Group as a consortium with fees and an operating style much like the World Wide Web Consortium. Several partipants at today’s discussion insisted they’ve made a 180-degree turn, that they’re not trying to restrict discussion to paying members, that they’re active on the oft-cited decentralization mailing list, that they’d be willing to open the planning meetings if the community wants it, and that they’re reaching out to project leaders all over the P2P space and getting input from everyone. Much has happened even in the past week, and Tim O’Reilly assured us that “it’s not done yet.” Bob Knighten said that they were starting with the basic question of what needs to be standardized. He told me later that the open-source library they released for peer-to-peer security is an “experiment” and that they look forward to others putting forward their packages. The “town meeting” (open discussion) that ended the day started right off with the same question of whether the working group was excluding people through its fees or members-only meetings.

I was wondering why Intel was slammed so hard by people in the P2P field, when Tim Berners-Lee got away with creating the W3C with pretty much the same structure. (Very rarely has anybody grumbled about the W3C–and usually just when they weren’t making progress on something.) I have an answer for this question: it has to do with the context in which the W3C started versus the current P2P field.

When Berners-Lee proposed the W3C, Netscape had defeated Mosaic (through superior technology, to be sure) and emerged as the ferocious lion dominating the WWW savannah. While Berners-Lee’s organization was somewhat closed, it was seen by everyone as a relatively well-organized salvation from the monopoly situation that loomed.

By contrast, the P2P arena is currently completely open; there’s no way to tell what the relationships are among the players or who will win. People want some coordination and standardization, but they’re not going to put up with the faintest attempt to draw a line and say who’s in or who’s out. Intel, like Sun with their JXTA proposal, is coming into a very different environment from Berners-Lee. It’s also, of course, a different era in computing history, characterized by a triumphant open source movement.

I found the interaction at this conference–the 2 in people-2-people–to be exhilerating and intensely productive. People really made the conference; it was a Symphony of a Thousand (850 registrations and 150 speakers, journalists, etc.). Audience comments were as insightful and provocative as the speakers.

There will never be another peer-to-peer conference like this, the first. But there will be another O’Reilly peer-to-peer conference this coming September 17-20 in Washington, D.C. The conference will be totally different from this one, I’m sure, because the field will be different. What both will turn out to be like remains unknown until we all turn up there.

Related link: http://digitalmass.boston.com/news/daily/02/021501/oneclick_patent.html

In a pre-trial decision, a Federal Court will allow Barnes & Noble to use one-click technology on its site while awaiting the September trial that disputes Amazon’s one-click patent claim.

Nobody seems to be creating much infrastructure for peer-to-peer applications. At least, that’s how it should be, as we heard this morning from Bill Joy (of BSD and Sun fame) in his P2P conference keynote.

You’ve heard the clichés “Less is more” and “The government that governs least, governs best” and “Everything should be as simple as possible but no simpler.” Well, Joy’s talk suggested to me that peer-to-peer should have as few standards as possible, but no fewer.

I didn’t enter the Westin’s grand ballroom all groomed to be won over by Bill Joy, archangel of computing though he may be. I had heard several cynical reports (such as a weblog by Rael Dornfest) of Sun’s recent Sun ONE announcement, which several people thought was mostly a hyped combination of old features. I knew Joy was going to make a major announcement in his keynote, but expected nothing better than “ONE squared.”

But there does seem to be solid good sense behind JXTA, the odd buzzword that Sun derived from the word “juxtaposition.” Bill Joy explained that their long tunneling through the dictionary during the Java years have left them with a lot of trouble finding more words that begin with J. He apparently did not think it appropriate to use the work “jockeying,” even though this is obviously what the new initiative is trying to accomplish relative to Microsoft’s .NET and Intel’s recent release of an open-source library for
peer-to-peer application infrastructure.

Good sense, even so, because Joy and his management staff claim to be disavowing world domination. First, JXTA is meant to be small. Like a well-designed language, Joy said, it should provide just enough to fulfill all the requirements at its own level and leave room for people to innovate at a higher level. He listed three tasks:

1. To allow pipes from one peer to another.

2. An ability to group peers and form heirarchies of groups.

3. To allow monitoring: knowing what’s going on across the network, instituting policies, and maintaining control.

Second, Sun is trying to avoid a repeat of the grumbling its Java licensees have expressed over its control of that language. Project manager Mike Clary avows that Sun is “just another participant out there that’s helping to influence the technology.” JXTA will be placed under an Apache-style license.

Their goal? Quite modest, said Joy: “We want to develop a platform so we can produce some successful apps and build a community around it. If there are other communities, that’s fine.”

And one of the most perceptive statements of the morning, I thought, came from the marketing manager, who deliberately called Sun’s current sandbox the “peer-to-peer space” rather than the “peer-to-peer market.” She said, “There’s no “peer-to-peer market any more than there’s a client/server market.” I think it’s critical to recognize, as she does, how broad the buzzword “peer-to-peer” is.

Attempts at world domination, which Tim O’Reilly warned against during the discussion following the keynote, came up in various guises several times through the day. During a panel about Web services, for instance, when asked how end users could trust the reliability and honesty of a service, Dr. Steve Burbeck of IBM suggested “a Better Business Bureau or Dun & Bradstreet model for a trusted rater of Web services.” It seems that rampant decentralization needs to be countered by some centralization. Other panelists, however, suggested that peer rating (as is done imperfectly now on eBay) could fill the credibility gap.

Freenet project manager Brandon Wiley seemed set on world domination, as he described how all Internet protocols are being rewritten by his team to run over Freenet. All your applications could work just like before, except that you’d feel like you were Alice in Wonderland after quaffing the bottle that said “Drink me.” I cannot describe the bizarre effect of his hilarious and insightful talk.

Jim Gallagher ranted (his own words) against another form of world domination, caused by ICANN and a trademarked-centered Domain Name System. In general, I thought, even the talks that had obvious marketing content showed some novel approaches to solving old problems. If someone is to come to dominate the world, the smart guys at this conference have just as much a right as anyone.

It’s common in an emerging technology area for journalists and venture capitalists to bemoan the lack of a clear business model. You can hear those questions being asked about P2P, about Napster, Gnutella, Seti@Home, Groove Networks and other such efforts. The focus is too often on picking winners. Yesterday, at the O’Reilly P2P Conference, Larry Cheng of Battery Ventures did a good job of handicapping the P2P field, analyzing which companies had good technology, an experienced management team and a credible set of reference customers.

Right now, anyway, it’s too early to be picking winners (unless it’s really your job to gaze into a crystal ball). What’s going on in the P2P space is research and development into the shape of the Internet. This research
is coming from a variety of sectors: well-funded, commercial companies; seat-of-the-pants startups; well-known universities; and distributed, open source projects some of which are no more than proposals on SourceForge.

So, at O’Reilly P2P Conference, the kind of question to ask is — what is being learned? How do they see the current state of the Internet? Many are here saying that the Internet is broken, that even the process of sharing a document via a Web site can involve getting permission from an ISP, paying to register a domain name and other such bureaucratic steps. Clay Shirky pointed this out in his keynote, explaining how Napster by making it easier to share files creates “usability at the network layer.”

One area that P2P focuses today is discovering what Shirky calls “resources at the edge of the network.” These resources are devices, people, and programs. These resources are under-connected, and under-utilized.

P2P seeks to connect these resources without necessarily establishing a central control point. However, some centralization seems to be necessary in most of the P2P systems, and this is an interesting area of research. Clay Shirky said that we should ask of any P2P system if it is “decentralized enough.”

For now, it’s good to have a lot of different approaches because each is an investigation into solving important problems. Moreover, each is a kind of an assault on the established way of thinking. And for some of us, it is a little bit odd to think of the Internet or the Web as the establishment; but it is (and it isn’t). Many of the P2P efforts are valuable critiques on the capabilities of the Internet, and its emergence as a mass medium.

Of course, all the P2P players are hoping that their research leads them to establish a new way of doing things. At that point, the winner will be obvious and the business model will be crystal clear.

People at the O’Reilly P2P conference are telling me something over and over, but I can’t quite hear it yet. There’s some unspoken thread running underneath their fervent descriptions of their projects, where they believe some peer-to-peer model is critical to success. It’s something about people having to customize a service that is so complicated it could never be formalized by a central staff. Whether it’s a matter of customizing a large order or storing digital content for later retrieval, peer-to-peer seems to offer an escape hatch where traditional centralization would tie everybody in knots.

This illustrates what’s so great about the conference: it gives one a chance to re-examine the fundamental premises and purpose of one’s work. Many conferences feature people standing in tight clumps arguing about the best strategy for integrating RMI and COM or some such narrow topic. While you get plenty of that here, you also get questions like “What can allow people to make the best use of the Internet?” or “What could people accomplish if they could share all the resources on a million computers?” Not in a fluffy or superficial way, but a serious exploration of a serious technical question teaming with social implications.

Just a few impressions follow of the sessions I saw today.

The keynote by Clay Shirky (along with a background introduction by Tim O’Reilly) was complemented beautifully by a later morning panel on “P2P in the Enterprise.” Tim and Clay painted broad, rainbow-colored visions of the future. The following panel basically covered the question, “How does anybody make a living doing this?”

Clay gave yet another in his evolving definitions of P2P (”aggregating resources at the edges of the network”), offered several reasons why Napster works (I’m sure he’ll publish those on his own) and warned us of the ominous centralization represented by the databases at sites like AOL Instant Messenger and Napster. If each P2P application sets up its own name service, as they do, not only does the world of P2P applications become fragmented, but a public resource becomes subject to the whims of those in control of the database.

In another blue-sky session that followed, Johnny Deep of Aimster suggested that their combo of instant messaging and file downloading could take care of the last mile problem that infrastructure providers haven’t been able to economically solve.

I couldn’t attend most of the afternoon sessions, but the final CTO Panel of distributed computation pointed up some of the amazing spread of approaches that all fall under the P2P buzzword. Take the issue of security, for instance. When you’re downloading software to do computations, you have to worry whether it’s malicious (or just buggy), while the person giving you the computation to solve has to worry whether you’ll substitute your own software and return bad results. This can be a major security issue. On the other hand, projects like Aimster and Groove base their security on the trust among a small group of collaborators who know each other.

The lack of popularity of the public key infrastructure (PKI) came up, and was ascribed to the trouble it takes to get people to set up their keys. This ties back to Shirky’s keynote, because he stressed that successful systems hide the complexity of the Internet and offer a system that anybody can use–and that just works.

They tell me that the Westin St. Francis is San Francisco’s most elegant hotel. Its formality does not fit the image of the people I’ve been dealing with in the P2P world over the past six months, many of whom are scruffy graduate students. In fact, some are even undergraduate students (although these tend to be less scruffy than the graduate students). But I am being unfair to my beloved authors and august companions, some of whom are well-regarded experts in computing.

I’m honored with a corner room on the 11th story, from which I can see the corner rooms of two neighboring 11-story buildings. The room is small and there’s a special electricity surcharge on the bill to make outsiders share the pain with Californians.

Still, the twisty staircases, opera-house presentation space, and dim hallways (which make the hotel a perfect setting for a 1930s-style mystery movie) should provide a congenial setting for all the novel ideas that will emerge over the next three days. I am psyched to hear new things. At least twice a week for the past several months, somebody somewhere has sent me an article that provides a perspective I haven’t yet seen on the wide-ranging P2P phenomenon.

An example of what we don’t need comes from a book called Meta-Capitalism, which I saw heavily marketed in the airport. The authors’ thesis is that we are entering a stage of accelerated capitalism (wasn’t the orignal version destructive enough?) during which successful companies decapitalize and divest themselves of productive facilities. So who the hell will have the capital and the productive facilities? Unsuccessful companies? Let no one at this conference fall into such babble. Check this space for something more significant tomorrow.

Tara Calishain uses the net as a tool for market research. As a consequence, she tracks what’s happening among search engines and related services on her site, ResearchBuzz. She has relied heavily on Usenet newsgroups, and the search interface at Deja.com, aka DejaNews. When she learned that Google had bought Deja,
this expert user had this to say:

Even though I speculated on this back in October (“Deja News Search Engine for Sale: News and Irresponsible Speculation,” I’m still stunned to hear that Google has purchased Deja.com, including, they say, the entire Usenet archive dating from 1995.

A beta version of the news search is available at groups.google.com , which you can also get to at http://www.deja.com.

Longtime Deja users (like moi) will be annoyed to see that the advanced search suffers from some shortcomings. You can no longer search by title, apparently. You may not confine yourself to a certain group of groups (forsale, standard, complete, etc.) It is not clear if you can exclude newsgroups (for example, do a search for resume and exclude all jobs newsgroups.) Google also doesn’t support hierarchical browsing at the moment either (keep repeating to yourself, “It’s only a beta. It’s only a beta…”)

If you’re a Google user the search results will look familiar. The search results consist of title and a few words from the message. The bottom line of the search result contains the newsgroup to which it was posted, date of posting, and name of poster. There’s also a link to review the entire thread.

The current archive only goes back to August of 2000, but Google co-founder Sergey Brin told me they hope to get up the archives from 1995 within the next 90 days. He also noted that there’s some functionality currently missing (the incompleteness of the advanced search, for example, and the lack of Web-based posting) but that Google does intend to restore the data archives to all their glory. “Twenty years from now this will be very valuable data. We want to keep it safe and make it available.”

Google does plan to integrate text-based advertising into the Deja searches, both with the premium spots (the colored boxes at the top of the page) and the AdWords program (the text ads in small boxes down the right side of the page) but they didn’t have current specifics available. (Searches for general keywords cars and flowers found no advertisements, so I would guess it’s not implemented yet.)

I’m relieved to see that Deja is getting what looks like a good home. I can’t wait for the promised improvements in the search interface.

It is good news that these Usenet archives appear to be in the hands of a company who understands their value.

Related link: http://slashdot.org/articles/01/02/06/2030205.shtml

In a Slashdot interview, David Korn discusses
his own Korn Shell, and its place in Unix history and philosophy. He compares ksh with other shells (bash, c, bourne) and considers different implementations of the Korn shell under Linux. Most versions of the Korn Shell are derived from an older code base, and the newer version (ksh93) has been proprietary to AT&T until recently. Source and binaries for the Korn Shell (ksh93) are now available for download from AT&T. Korn’s recent work has been develoted to the UWIN project, an effort to create a Unix environment on top of Windows.

The W3C has never been quite sure if it is a standards body or not. It is an industry consortium consisting of member organizations who participate in working groups that produce specifications, which must be approved by the membership. What’s difficult to know is how much the W3C feels its obligation to serve the interests of the entire Web community, or to cater to the needs of its members.

This largely unresolved question sometimes puts the W3C in an awkward position. What happens, for instance, if W3C members fail to implement a W3C specification to the letter? Or more specifically, when Microsoft fails to implement fully the specifications that affect the browser, what can the W3C do about it?

Not much, judging from the W3C’s recent Note Common User Agent Problems. This Note, produced by three W3C staff members, usefully points out “some common mistakes in user agents due to incorrect or incomplete implementation of specifications.” But it is clear from the Note’s introduction that the W3C is uncomfortable pointing out these problems, perhaps because it might be perceived as criticizing one of its member organizations. How else to read the following?

This document is a Note made available by the W3C for discussion only. Publication of this Note by W3C indicates no endorsement by W3C or the W3C Team, or any W3C Members. There is no commitment by W3C to invest additional resources in topics addressed by this Note.

In short, this Note is merely informative. It has no teeth. The W3C seems to accept a fairly passive role in enforcing, or even encouraging, compliance with its specifications.

This sheepishness creates a lot of uncertainty on the Web. We have to wait to see how well a W3C specification will be implemented, if at all, before deciding if the specification is real. The dominant implementation is what’s really real; it becomes the defacto standard. In the case of the Web browser, what matters is how IE actually works, not how the W3C specifications say it should work.

The authors of this Note say that they did not want to “incriminate specific user agents” and presumably, the members that developed them. Yet, that is precisely what they should do to monitor compliance and point out deficencies in implementations. Name names. Document which browsers cause the problems.

The W3C should be proactive in determining whether common problems are the fault of the developer’s implementation or an ambiguity in the language of the specification. The W3C needs to be a fearless champion of its own standards, defending the Web against those who fail to comply.

AudioRequest makes an MP3 jukebox (ARQ), a blackbox computer that is intended to be a home theatre component. With a single CD drive, this device will convert your CDs to MP3 files and store them on its 17 or 30G hard drive. A user, on the company’s bulletin board, said accurately that he bought the ARQ not to replace his CD player, but to replace his CD collection.

AudioReQuest has two Intel-Celeron-based models: the ARQ1-20 unit with a 17G hard drive at $799.95 and the ARQ1-30 unit with a 30G drive at a cost of $1199.95. This jukebox computer, which only has a small LCD window on its front panel, can interface with a PC or connect directly to a network. It has a lot going on in the back panel: 1 Ethernet port, 1 parallel port, 1 keyboard port, 2 USB ports, and 1 VGA out port. You can connect it to your TV set with either the S-Video or composite video connections. It also has audio in and out ports. Because of the lack of controls on its front panel, you interact with the AudioRequest via a remote control device or use AudioRequest’s software on your PC to play music, or to upload or download MP3 files. Unfortunately, the software interface emulates the remote control, and is relatively clumsy to use.

When I bought an AudioRequest unit last year, I thought it looked a lot like a standard computer. But I didn’t know much about what was really inside the Jukebox and what software it ran. The company’s production information was rather light on specifics.

Then, just recently, I came across
the
ARQ Hacking page, which told me everything the company didn’t think I needed to know.
Christopher Richmond and friends started the page to describe what they found when they opened the AudioRequest black box and began exploring inside. They not only looked but they began to try changing things, on their own.

Now, I will probably never follow their guidance and open the Jukebox, and in the process, void my warranty. But I learned a tremendous amount from their various experimentations. Chris describes how to expand the ARQ by replacing the hard drive with a larger one (80 G). He describes overclocking the 366 Celeron processor and monitoring the rise in temperature. He even finds out how to access the BIOS, which he discovers is password-protected. But he cracks that. “After trying lots of AudioRequest specific information (including the cat names: Elsie and Abu), I was out of ideas. I did some searching on the NET and found the following program: bios320.exe,” a utility that “allows you to crack the BIOS password if you’ve forgotten it…ahem….”

I had wondered what operating system ran the ARQ. From the ARQ Hacking, I learned that it was QNX, a real-time operating system based on a microkernel. Of course, Chris could not leave well enough alone. He installed Red-Hat Linux on the ARQ with reasonable success.

“Why did I do this?” Chris asks rhetorically. “To see if it could be done…for fun… …to learn something new.” This is a wonderful hacker quality.

(with thanks to Rob Flickenger)

The recent launch of our OnLAMP open source web platform interest site was received with some consternation by certain parties at this year’s LinuxWorld convention. O’Reilly Net staff on hand at the convention reported being asked many concerned questions from people using alternatives to the Linux / Apache / MySQL / ( Perl | Python | PHP ) hegemony.

“What about *BSD?” they were asked. “What about PostgreSQL? How about Roxen, or Ruby?” Meanwhile, back at the office, one member of our staff went out and registered the OnBAMP.com domain, and another quipped, “Maybe we should call the site OnBLARMP.com?”

But underneath all the questions and the witty remarks lies a real confusion at the heart of the Open Source community. When does advocacy of one particular application or platform or OS become detrimental to the growth or use of another similar package? Am I, by pointing out all the things I like about Linux, or about Perl, somehow implying that there’s something wrong with *BSD or Python? Not necessarily. Yet this assumption, this “tribalism” (as Mark-Jason Dominus puts it), seems to infect our community every time there’s an option between two or more tools that do similar things.

This confusion persists even here at the O’Reilly Network, where concern has been expressed that, with OnLAMP.com, we’re binding ourselves and our efforts to a particular cross-section of the available technologies, almost willy-nilly. But I don’t think that’s really the case. Let’s have another look at the “LAMP” acronym.

LAMP stands for Linux, Apache, MySQL, and P-for-Perl, PHP, or Python. Now, let’s ask a silly question: What features do these packages have in common? First, they fulfill roles integral to any web-based application platform; namely, OS-level services, application-level services, databasing, and scripting. Second, they each excel at the task they were intended for, in terms of performance, flexibility, reliability, and scalability. Third, and possibly most importantly, they are all Open, if not Free outright. At the O’Reilly Network, we believe that these three features are crucial to providing web-based (or other network-based) application services, anywhere, anytime.

However, nowhere have we stated, nor should we ever state, that these are the only tools for the job. The *BSD operating systems are just as Open as Linux, and can be more or less useful to you personally than Linux, depending on your preferences. Apache may be the most popular webserver on the ‘Net, but perfectly good alternatives exist (e.g., Roxen) that are just as Open. PostgreSQL may not always be as fast as MySQL, but it has features that MySQL doesn’t. Perl may be faster and more widely supported than Python, but Python has Zope. Or maybe some of the newer languages, like Ruby or Pike, have fixed things you didn’t like about the more established ones. Or maybe you just like Tcl. And so on.

As such, there isn’t really any reason you can’t (or shouldn’t) use BAMP, (i.e. *BSD in place of Linux), LAPP (PostgreSQL in place of MySQL), or LRMP (Roxen), or even the unfortunate-sounding LAMR (with Ruby). The idea of interchangeable tools isn’t new in the UNIX world, either. For almost thirty years, the *NIX communities have with one voice propounded the UNIX Tools Philosophy, which is usually stated as: “Each tool should do one job, and do it well.” Well, things are more complicated these days than they were back when cat and sort and grep were young. It’s easy to argue that you don’t need an alternative to cat that does the same thing, but works differently. However, when you get into domains as large and as complicated as “operating systems” and “network servers” and “application development languages”, things become a little blurred. In fact, in practice, the issues become very blurred, at least to the point where maybe, just maybe, there might be more than one way to seize the proverbial cat by the tail.

So, if the UNIX Tools Philosophy implies the old saw about “using the Right Tool for the job”, then the LAMP Tools Philopsophy exhorts us to “use any Right Tool for the job,” so long as it helps you get the job done. Which is really what LAMP is all about, ultimately: Using the tools to get the job done, and done well, every time. This situation, and Open Source advocacy in general, is not, to borrow a term from a recent Clay Shirky article, Pareto Optimal; that is, we can recommend and advise on the use of one operating system or scripting language, without implying anything to the detriment or expense of another. (For deeper insight into this idea, see Dominus’s aforementioned article.)

Obviously, then, the choice of the “LAMP” acronym is merely a matter of convenience: It indicates examples of the tools that people can use to get their job done, and, furthermore, it’s easy to pronounce in English. By the term LAMP, we at the O’Reilly Network mean to embrace all of the Open Source tools that people use in building network application platforms, and to emphasize the ways they can be combined, as parts of a functioning whole. And, besides, LAMP is “all the rage in Europe.” =)