Aug 5

Tim O'Reilly

Tim O'Reilly

Not 2.0?

Tim Bray writes:

I just wanted to say how much I’ve come to dislike this “Web 2.0” faux-meme. It’s not only vacuous marketing hype, it can’t possibly be right. In terms of qualitative changes of everyone’s experience of the Web, the first happened when Google hit its stride and suddenly search was useful for, and used by, everyone every day. The second—syndication and blogging turning the Web from a library into an event stream—is in the middle of happening. So a lot of us are already on 3.0. Anyhow, I think Usenet might have been the real 1.0. But most times, the whole thing still feels like a shaky early beta to me.
While being completely right in the details (we are quite arguably on 3.0 or even 8.0 if we're thinking about the internet compared to other software versioning), Tim is completely wrong about the big picture. Memes are almost always "marketing hype" -- bumper stickers is a better way to say it -- but they tend to catch on only if they capture some bit of the zeitgeist. The reason that the term "Web 2.0" has been bandied about so much since Dale Dougherty came up with it a year and a half ago in a conference planning session (leading to our Web 2.0 Conference) is because it does capture the widespread sense that there's something qualitatively different about today's web.

Kevin Kelly wrote about this change at length in an article in the current issue of Wired: the key to success in this next stage of the web's evolution is leveraging collective intelligence. And yes, Google's introduction of page rank was absolutely a milestone in this evolution of the web, but what was once an isolated stroke of genius is now being understood as one of the keys to the new paradigm. There's a set of "Web 2.0 design patterns" -- architecting systems so that they get smarter the more people use them, monetizing the long tail via a combination of customer-self service and algorithmic management, lightweight business models made possible by cooperating internet services and data syndication, data as the "intel inside", and so on.

More immediately, Web 2.0 is the era when people have come to realize that it's not the software that enables the web that matters so much as the services that are delivered over the web. Web 1.0 was the era when people could think that Netscape (a software company) was the contender for the computer industry crown; Web 2.0 is the era when people are recognizing that leadership in the computer industry has passed from traditional software companies to a new kind of internet service company. The net has replaced the PC as the platform that matters, just as the PC replaced the mainframe and minicomputer.

You have to remember that every revolution occurs in stages, and often isn't recognized till long after the new world is in place. The PC revolution began in the early 80s, and most of the key PC companies and technology innovations were founded in that decade, but it wasn't till the mid-90s that the new shape of the computer industry was clear to everyone. The Microsoft-Netscape equivalent of the 80's was the debate about whether ATT's entry into the computer industry would dethrone IBM. The crucial choices had already been made, though, that set the course for the Wintel-dominated industry of the 90s. Similarly, the writing was on the wall when Yahoo!, EBay, Amazon, Google and other web giants were started in the mid-90s. We're now at a stage equivalent to the period in the PC market when people were debating whether OS2 or Windows was the operating system of the future.

Perhaps I'm biased, because O'Reilly was the source and has been one of the biggest promoters of the Web 2.0 meme, but I think it captures exactly where we are at this moment: a widespread awakening to the fact that the game has changed. There might be a better name (I tried "internet operating system" on for size starting back in 2000), but the fact that Web 2.0 has caught on says that it's as good a term as any. While the patterns that constitute Web 2.0 are far from completely understood, there's a kind of intuitive recognition of sites that are expressing the new model. (For example, at Esther Dyson's PC Forum last March, after presentations by two startups showing shared calendaring services, I overheard one attendee say to another, "xxx is so Web 1.0, and yyy is so Web 2.0" and the other attendee knew exactly what he meant. A meme is a pointer, and as long as it points in the right direction, so that the listener recognizes what is being pointed at, it works.)

I guess it's the old debate between language purists, and language pragmatists. The right words are the ones people actually use, and this word is catching on.

tags: web 2.0  | comments: 74   | Sphere It

Previous  |  Next

0 TrackBacks

TrackBack URL for this entry:

Comments: 74

  Pat [08.05.05 05:23 PM]

Since when have version numbers ever made sense?

  John Dowdell [08.05.05 07:37 PM]

Thanks, Tim. I keep looking for an authoritative, single-sentence definition of that "Web 2.0" label -- most of the discussions seem to get sidetracked into "Is X 'Web 2.0'?" or "Is Y 'Web 2.0'?" -- it'd be great to get past that, into analyzing implementations.

The closest to what I was seeking here seemed to be this line you had: "There's a set of "Web 2.0 design patterns" -- architecting systems so that they get smarter the more people use them, monetizing the long tail via a combination of customer-self service and algorithmic management, lightweight business models made possible by cooperating internet services and data syndication, data as the "intel inside", and so on."

Is this close to listing the characteristics you check for when you decide whether something "is Web 2.0" or "is not Web 2.0"? Are those four traits usually the significant identifiers?

If so, then would this be accepted by others as the canonical definition of the term? Or maybe there's a better definition elsewhere? Do you think these set of identifiers would be generally accepted?

And if so, then would I be accurate, when talking to others, if I phrased it along these lines? "If a service gets smarter and richer as its audience increases, and if audience input and existing services are the main production costs, then it can serve the majority through many minor markets, and can be called by the label 'Web 2.0'."

That seems sort of a definition by business model, rather than by technology... could you refine such a single-sentence definition? Is there some sentence which would be accurate, useful, and widely accepted?

(I've got no interest in which definition it turns out to be... my main goal is to get the discussions less about labeling, more about working, thanks.)

  Tim O'Reilly [08.05.05 08:29 PM]


As it turns out, on my to do list is an article entitled "What is Web 2.0" You've given me some good food for thought here. I'm going to try to get it done in the next couple of weeks.

  Savanna [08.05.05 09:08 PM]

i have to disagree. =)

i think 'web 2.0' is totally silly. it sounds like such a hyped up techie thing to say. and, as much as i thought your column was well thought out and all, any argument which consists of 'paradigm' when talking about a 'zeitgeist' in the techie world means that it sounds like marketing b.s. to me. =)

if you went up to the average user and said 'that is so web 2.0' to them, they'd look at you and say something like 'oh, they have something new out now?' they wouldn't get it. not to mention that *most* users don't even *know* what an rss feed is yet! 'blog' may have been the word of the year last year, but i know a *lot* of people who still ask me 'so what's a blog?' they just don't get it.

at least with the 'hype' about netscape, people *got* the idea. you told them 'netscape' and it was like 'oh ya! the web!' and if you told them 'google', they would say the same thing starting a few years ago. say 'blog' and people still look at you weird sometimes like you just sneezed or something. and like when you think about it, that's what matters. when devs talk about 'web 2.0', i think they're just trying to come up with some sort of new way to present stuff that people are doing anyway as if it's a whole philosophy behind it.

which basically means: hot air. =)

it's about as catchy as (remember this one?) 'information superhighway'!

ya, that *really* lasted. =)

still, i like your column. you techies do make me smile. =)



  Kim [08.06.05 05:31 AM]

I just hope that Web 2.5 will be the time where standard HTML/CSS will be used and IE7.5 will follow W3C standards :-)

  Eric [08.06.05 06:54 AM]

Four blind men encounter an elephant. One grabs the leg and is convinced it's a tree trunk. One holds the tail and thinks it's a whip. Another touches the elephant's trunk and decides it's a hose while the fourth man pats the side and is sure it's a wall.

The wise man tells them, "All of you are right."

These design patterns all point to something, and in time we're beginning to see the elephant.

But WRT to the the name, I think it's great. Netscape IPO was 1.0, pagerank was a minor but greatly appreciated version change, say 1.4 to 1.5 . But big changes are afoot. WebOS might be a little better at capturing such a paradigmatic shift, but hey as long as we know what we're talking about.

  shel Israel [08.06.05 07:18 AM]

It seems to me that "2.0" implies a new, improved version being released by some entity on a particular daye, and that is not the case with Web 2.0. The vast improvements you describe Tim,will happen through a long steady of emergence of innovations--some small and some very significant. While we are waiting for them all to come together in a next generation Internet. Web 2.0 has slipped into market vernacular. Take a look at launches of new Internet-related products and services. Note how many times, insignificant new offerings claim to be ushering in this new Web 2.0. Soon I expect we will hear promises that the new Web 2.0, as they are bringing it, will make us all taller, smarter, and give us whiter teeth.

  paul [08.06.05 08:09 AM]

So much hairsplitting . . .

While I'm sympathetic with the pushback against a version number (as if "the Web" is a product a single entity), I think the idea of Web 2.0 points out that things are different and arguably better than when most of us jumped into the pool.

I don't think Tim or Dale are using it as a sales tool but simply as a milepost: we have come this far. No claims as to how far there may still be to go -- simply an acknowledgment that things have evolved (or been designed) to a represent a new experience we didn't have 10 years ago when Netscape debuted.

Personally, I'm inclined to say that something is so 20th century: it makes the point and gets a laugh, but is the turn of a century any less arbitrary than an experience's version number?

  Danny [08.06.05 08:35 AM]

Obviously I can't disagree with your definition of Web 2.0, with O'Reilly being the source... But what shifted my opinion of the meme from "hot air" to "something useful" was Ian Davis' suggestion that Web 2.0 is an attitude not a technology. When you say "it's not the software that enables the web that matters so much as the services that are delivered over the web" there are echos of the Andreesen/Berners-Lee argument over the image tag. There's a big difference between what the tech can support and what you do with it. The growth of the Web that led to the dotcom bubble was characterized by mind-numbingly uninspired use of the tech, mirroring real-world (commercial) applications in the new environment. The recent growth of blogs and syndication isn't essentially very interesting tech-wise, but the empowerment of the end user to create content easily is positive (something of a fulfillment of the original plan for read/write browsers). But this is all Web 1.x, where these and other areas gets more interesting is in the sharing of data and interconnection of the services. Individual and corporate presences on the Web are becoming less like discrete islands. Desktop and mobile tools for the Web (browsers, aggregators, widgets) are becoming less like dumb terminals and more like virtual machines contributing as first-class Web entities. I personally suspect what we're calling Web 2.0 right now is a very early phase of a significant transition, one which will take a further 5-10 years to come to fruition (long version: The Missing Webs). I do think there are ways of accelerating progress, in particular encouraging the paradigm shift from the Web as document repository to Web as global database. The Web 2.0 attitude is a good lever point.

btw, here's another elephant

  Tim O'Reilly [08.06.05 10:23 AM]

Shel --

As to the downside of a new marketing meme -- "it will also make your teeth whiter" -- this is certainly true. But that doesn't make the new idea any less valid. When we named "open source", everyone and his mother jumped on the bandwagon, and many of them got it wrong, but nonetheless, we'd created a new name that helped people see something that they hadn't noticed before. We grouped a set of concepts together, gave them a name, and nothing was ever the same again. Names have this power: they allow us to see.

For some additional background on how I think about all this, see two essays I wrote years ago: Remaking the Peer to Peer Meme (which, despite the name, spends a lot of time on how we created the Open Source meme), and Language, Thought and Reality, which is a meditation on Perl through the eyes of Benjamin Whorf.

  Tim O'Reilly [08.06.05 10:40 AM]

Danny -- I love the idea that web 2.0 is an attitude. It's also a sensibility. A sensitivity to the nuances of how the web is changing the possibilities of computing, rather than an attempt to fit to old models. I love the Wallace Stevens phrase: "search[ing] the possible for its possibleness." And yes, I completely agree that this is just the beginning of something very different, not the completion of a new stage. It will indeed take 5-10 years or more to play out.

Savanna -- if you went up to the average consumer, and said "What is open source", you'd get the same reaction, but that doesn't mean that open source software isn't reshaping the computer industry right now!

Eric -- I definitely agree that we're in the blind men and the elephant stage. We know that there's something there, and are feeling our way towards an understanding of just what it is. That in fact is part of what makes this a powerful meme. If it was trivial to understand, we wouldn't all be wrestling with it. The numinous power of new ideas is precisely that they are psychoactive, not just fitting neatly into old forms and concepts but making us stretch to encompass them.

  Tim O'Reilly [08.06.05 10:54 AM]

Savannah --

One more thing: "information superhighway." While the term didn't last, it had an impact. It played a significant role in helping ordinary people to understand the possibilities of the internet, even as the metaphor was outrun by the reality. It also gave us such memorable images as "roadkill on the information superhighway," an idea that still makes people smile with recognition.

  pb [08.06.05 04:40 PM]

This may be simplistic, but as I think about it, also a workable definition, with some more subtle implications.

Web 1 was the period up to the dot bomb. Web 2 is what has been emerging, what has been succeeding, since that big implosion. I believe that this is the way most "average" people -- if not the technical elite -- will perceive the terms.

And while it may be simplistic and, on the surface, reductionist to the point of ignoring both variation within a timeframe as well as technical and design underpinnings, the macrocosm today nonetheless corresponds to the perspective.

Internet organizations succeeding today (as opposed to mere internet outlets, e.g. a non-dynamic product catalog, a newspaper's online version, or government information) are enabling users' collaboration and control, deliberate (e.g. blogging, customization) and unpremeditated (e.g. PageRank), as a central function of their business model. A self-sorting matrix in place of a either strict hierarchy or an undifferentiated mass.

Whether those organizations were born after dot bomb, or whether they rode their way through that low point, their current work reflects the above.

Organizations that imploded, did not offer the above. Not in any substantive, qualitatively significant fashion. And organizations that feel "old school" these days, even if brand new, also reflect that lack.

This is going to sound cheesy, but, perhaps Web 1 was marketshare. Web 2 is mindshare.

The smart organizations have learned that it's mindshare that leads to marketshare.

  Dare Obasanjo [08.07.05 03:52 AM]

The problem with "Web 2.0" is that it has an amorphous definition and thus its definition includes everything and nothing, kinda like "SOA".

When I read blogs claiming that "Web 2.0" is primarily a label attached to the new generation of startups that are sexy to Silicon Valley VCs like Flickr or or Wikipedia definitions that claim that using CSS and semantic XHTML makes a site "Web 2.0" I can tell the term is already becoming meaningless.

The funny thing is that just like with SOA there is a germ of a good idea behind all the hype. The problem is having to shovel past all the BS to get to the useful nuggets of information.

This rant sumarized from my response at

  Tim O'Reilly [08.07.05 08:46 AM]

pb --

I think you've hit on something. The dotcom meltdown was indeed the turning point. Companies that survived it did indeed have something that was different from their peers, and that "something" is now being emulated by new startups. And I think you're very right that defining Web 2.0 as "after the crash" will make a lot of sense to people, especially if we focus on how it was reinvented after the crash to learn from the survivors, and to eschew the ideas that led so many of the early web startups to fail.

  Arjun Singh [08.07.05 01:19 PM]

Web 2.0 might be great as an inside industry term, but I have heard it used with visitors at a tradeshow booth - and there were a lot of confused looks. I think Web 2.0 is a great inpiration and rallying point for techies and tech entrepreneurs but for the general public, even "Internet Operating System" is more descriptive.

  Tim O'Reilly [08.07.05 02:27 PM]

Arjun --

You're probably right, at least right now. But all I have to do is think back to the blank incomprehension of people when we were first showing the web itself back in 1993. The man on the street was just as confused. "What do you mean that this part of what's showing on the screen is a software application, and this other part, with buttons and links and other software like characteristics, is a document that is coming from a computer thousands of miles away?" What's more, that early web was also a muddle of different services: you'd point your web browser at a telnet server to look at library catalogs, at gopher servers, at WAIS...and all through the same browser. Just what was "the web" anyway? See my 2001 weblog entry, SLAC Symposium on the Early Web, for a more extensive account of that experience.

Any new term is confusing at first, till there's a general sense of what it means. And even then, it's a consensus that concentrates meaning out of a cloud of possible meanings, a gravitational core with a gradually attenuated influence, not a hard boundary.

  len [08.08.05 06:33 AM]

Web 2.0: The web as the horseless carriage becoming the automobile. The price for the new features was losing the reliability of the horse while keeping the discomfort of the carriage.

  Clarence Wooten [08.08.05 06:36 PM]

Very interesting comments. What is Web 2.0 and why does it make sense to call it 2.0? In my opinion, 2.0 makes sense because it denotes renewed post-bubble enthusiasm about the Web. Let’s face it, the Web lost its luster in the eyes of most during the bubble years. Using 2.0 to denote this renewed excitement makes more sense than attempting to be technically accurate by calling it Web 1.5 or Web 3.0. The Web is now back in the hands of the people! This is a fact!

  Savanna [08.09.05 02:04 AM]

hrmmm, tim!

well, putting it that way.... =)

actually, maybe you're right. maybe 'web 2.0' is a concept, but i hate the whole idea of the 'meme', even if people think it may not be. it just kinda *sounds* like one. don't you think so?

and i think the comparison is kinda silly too. 'version numbers' are so... 'web 1.0' =)

but don't you think things are happening so fast compared to like a few years ago? we're talking about doing stuff online which a couple years ago they all thought would be 'the killer app'! i mean, remember when people thought that ordering a pizza online would revolutionize the world? it hasn't, but people *do* order pizzas online today. =)

i guess i hear 'web 2.0' and i think 'ugh! 90's version of internet talk'. it's so...old school sounding. =)

hehe...sorry for going off. i still love your stuff. =) i might poke fun at techies sometimes but i do love you guys. =) just don't use 'paradigm' when you start talking about concepts! it sounds =) ugh! =)

*muahs* sav.

p.s. it's 'savanna', and not 'savannah' =)

  Steven Livingstone [08.09.05 09:35 AM]

I agree with most of this.

I think Web 2.0 (the versioning comment above made me chuckle) is a nice way to tell people indirectly that we learned from the stuff that never worked in the past and this is the future. Sure, it's marketing, but seems to have worked for Microsoft's OS's.

To be honest, the things i've seen associated with Web 2.0 are much in line with the way Xml has been thinking for a while (since '99 - outside of huge monolithic schemas). Simple, distributed, collaborative and can be integrated to produce a more powerful whole.

I see much of the very early browser based Internet thoughts coming through in Web 2 - we are back to simplicity and imagination. It's an exciting time again - what i do notice though is that people are actually creating things that DO stuff. Maybe we will see more of us creating simple, but effective things that are either bought or integrated by the big guys (and the small guys), hence distributing the wealth (rather than the huge random investments of the late 90's).

  James Walford [08.10.05 01:02 AM]

An interesting discussion.

While I tend to shy away from excessive labelling and pigeonholeing, especially of a medium that is so clearly transient and evolutionary in its nature, there are some clear pointers as to qualitative differences between early and contemporay web approaches.

It might be useful though to consider that the term Web 2.0 has echoes of "release version" about it. In this context it isn't a useful or desirable term. In the long run the versioning history of the web will be written not by the protagonists but by historians - an epochal shift isn't necessarily best viewed or defined from within.

In support of this view I think back to the William Gibson inspired forecasts for what the web would eventually end up like, with terms like Virtual Reality the order of the day. Though much of that could be put down to the excited naivety of a young technology the differences in the projected future of the web might be equally important to defining the "version" as any current stock-taking. As such it would seem to point to a possibly fundamental inability to make any strong assesment of where we might be regarding versioning.

If where we are headed is considered an important part of where we are now then where we actually arrive, which we don't yet know, must be considered a valuable measure of our assessment. Andthat, clearly, is something we can only tell the story of later.

Right now we have a story of the web in it's early phases. In twenty years time we may well consider our particular current position to be merely a sub-phase of a larger early phase, rather than as a radical break.

But what the heck, Web 2.0 is part of the current Zeitgeist, which in itself is currently largely informed by the web itself. One day the Zeitgeist may be to look back at 2005 and laugh at Web 2.0, which by then we'll be calling Web 0.7 beta :-)

*ramble over*

  James Walford [08.10.05 01:07 AM]


Tim, your preview function doesn't accurately represent the finished styling - I used double br tags to get a double line break, the preview certainly didn't warn me I'd muck up your layout quite so badly! Incidentally, it also semmed to ignore closed br tags. Very 1.0 :-)

  Thomas [08.10.05 03:30 AM]

Is web 2.0 reserved for web applications or can web sites also be web 2.0? Or are all sites little applications (like i.e. a weblog is) in web 2.0 and will the distinction between them vanish?

  Tim O'Reilly [08.11.05 08:48 AM]

Thomas -- I tend to think that an important part of web 2.0 is the blurring of the boundary between web site and web application. A site is an application, and its data is grist for other applications.

  Tim O'Reilly [08.11.05 09:16 AM]

Trackback roundup:

Since we had to turn off trackback display on the blog due to trackback spam (a Web 2.0 mental disorder, to which the immune system hasn't yet adequately responded), I wanted to point to a couple of the most significant trackbacks.

Tim Bray wrote a great response. We agree far more than we disagree. (And since Tim's blog doesn't allow comments, I'll need to continue the conversation here, which I will do in just a moment.)

Sam Ruby also wrote a great post on the debate. Sam, you're right that it's about the data, and composite applications built from the data. I've been arguing that the "internet operating system" will have subsystems that are about access to data. The equivalent to device drivers in the PC will be "drivers" for data service subsystems such as identity, location. payment, search, and so on.

I was pleased to see Business Week pick up on the debate, and to see a Texas VC say that he was polling his colleagues about their Web 2.0 investment thesis. He says none of the people he polled has one, but I'd bet most of the VCs who blog (like this one who explicitly expounds on the need for a Web 2.0 investment thesis) wouldn't agree.

For more, follow Tim Bray's advice, and "if you want to track the conversation, a really good way would be to scan the Web2.0 tagspace at Technorati or"

  Tim O'Reilly [08.11.05 10:12 AM]

Continuing the conversation with Tim Bray:

I'm glad to see that you recognize that we agree more than not. You're absolutely right that it's essential to try to shape the meme while it's still young, so it means the right things, to the extent that we know what those are. This debate has been extremely useful in helping to bring out nuances and ideas that should shape the definition, and I'm so happy it's happening.

I want to respond to a few of your niggles and other points.

1. Yes, most of the people who set out to monetize the long tail will fail, but if that were a criterion for deciding whether a new business idea were correct or not, we'd have to throw out the web itself, not to mention the PC, the handheld device, and most of the new technologies that have launched an industry.

Let me expand on what I mean by this point. Google AdSense is the paradigmatic example of a Web 2.0 business transformation. DoubleClick was Web 1.0 because it requires heavyweight contracts between participating players. Google AdSense is Web 2.0 because it leverages the twinned trends of customer-self service (you can easily set up the account yourself on the web, lowering the threshhold for how far down the tail you can build a monetary relationship) and powerful enough computing resources that it's cost effective to place, track and pay for advertising on sites that would be below the threshold for DoubleClick's model. Their site proudly claims 2000 customers: quite likely the "Fortune 2000" of the Web. Google AdSense not only serves the tail, it serves the head as well, and will eventually be positioned to take away DoubleClick's business.

(Of course, illustrating the complexity of the evolution of new business patterns, DoubleClick and its peers were one of the first signs of another aspect of Web 2.0, services delivered by cooperating web sites. After all, almost every ad banner served is a "mashup", with DoubleClick or equivalent site cooperating seamlessly with the target site to deliver content from yet another site.) EBay is a long tail company because it set up a system, again with customer self service and powerful algorithms, enabling a thriving marketplace among players who were formerly marginal in the high touch economic systems that eBay replaced. Believe me, there will be others, and understanding these dynamics is essential for business going forward. Understand them, or plan to be disrupted by someone who does. Yes, you'll most likely get some or all of it wrong, but if you don't try, you'll certainly get it all wrong.

2. Re "it's just data", I couldn't agree more. I've given talks for years with the tag line that "data is the Intel Inside of the next generation of computer applications." See for example my keynote at the MySQL User Conference.

3. You say, "if there really is a Web, The Next Generation, it’ll be the Web of mobile devices, with not many “computers” in the loop." I couldn't agree more. Again, one of my Web 2.0 design patterns is "software above the level of a single device." But I wouldn't agree that there aren't many "computers" in the loop, just that our definition of computer is changing. Computers 2.0 anyone :-) ? The computer of today is both larger and smaller than we ever dreamed. Commentators like to point out how our wristwatches today have more computing power than the computers that sent the Apollo to the moon in 1969; the trend accelerates as we go forward. On the other end of the spectrum, I love Clay Shirky's riff in his keynote at our P2P and Web Services Conference in 2001: "Thomas J. Watson famously said that he saw no need for more than five computers in the world. We now know that he was wrong. [pause, while people laugh, and think of the billions of computing devices in use, and then, deadpan, Clay delivers the devastating punch line.] ...We now know that he overestimated the number by four."

4. While I agree that "open source" was indeed better defined than Web 2.0 is, it's been just as subject to the hypemeisters. Every good idea is subject to hucksters! It is definitely true that we need to put "Web 2.0" through the rigorous process that I describe (both for "open source" and "P2P") in my article "Remaking the P2P Meme". As mentioned above, this discussion is a great help in that process.

But speaking of P2P, I think both you and Sam are wrong that P2P was a failure as a meme. It was one of the foundational ideas of Web 2.0, and continues to resonate today. I can't tell you how many startups have come into my office carrying Andy Oram's brilliant anthology Peer to Peer, telling me that it was their bible. (Also interesting to see that Citeseer has 105 academic citations to the book.) What's more, P2P still means something. A meme doesn't have to be at the top of the au courant list to have staying power.

5. I love your analysis of how to tell when something is hot. But there are degrees of hot. Linux still doesn't meet your test, but no one would say that it isn't having major impact. And it's also a matter of timing. Neither the PC nor the web would have met your test in the first few years of their existence. You describe "hot" as "post tipping point." But what's really interesting, if you want to get ahead of the curve, is knowing what's ascending towards the tipping point before it's obvious to everyone. (See Paul Kedrosky's great post on this subject, The Source of Superior Investment Returns.

Lots more to say, but I'm running out of time. Let's continue face to face!

P.S. Apologies if you thought I believed you were "completely wrong" and "accused" you of being a "language purist." I did indeed say you were completely wrong about the big picture, but you left out the fact that I said I agreed with you on many of your detailed comments. And while I did refer to the debate between language pragmatists and language purists, I was thinking of this as a general issue, not specifically putting you into either camp.

  Davanum Srinivas [08.11.05 10:41 AM]

More fodder, Is Semantic Web = Web 2.0? How is it different?

  Sam Ruby [08.11.05 11:41 AM]

re: both you and Sam are wrong that P2P was a failure as a meme

What I actually said was the term Web 2.0 will likely be as relevant in 2009 as the term P2P is today.

While I grant that the term "Web 2.0" clearly is helping today to shape what the web will become by 2009, and furthermore I will grant that the term itself will still be recognized in 2009 (as P2P is today); somehow, I doubt that the term will remain on the au courant list in 2009. In fact, it will likely seem somewhat dated by then. But that's OK, it will have done its job.

Offtopic: the title page on the "Previewing your Comment" page contains the characters >. The fact that so much of the web is operating in what Tim Bray refers to as shaky early beta mode, and yet not only survives but thrives, continues to fascinate me.

  Tim O'Reilly [08.11.05 12:17 PM]

Apologies, Sam. I was responding to Tim's piece, and his quick reference. As soon as I posted the comment, I was looking again at your original piece, noted my error, and was going to go fix it, but the press of real business (not blogging :-) intruded, and I was called away to a meeting. I completely agree that web 2.0 is not a "keeper" as a meme, probably not even as long for this world as P2P.

But while I'm on this theme, one more comment. P2P is a great analogy. When we did our first P2P Conference in 2001, everyone was puzzled because we featured web services and distributed computation on the program. "We thought P2P was about internet file sharing," everyone said. But we cast a broader net, because we thought P2P was really about the next generation of internet technologies, the applications and approaches that would become prevalent once we realized that the internet was the platform.

My point: pushing the boundaries of a meme can be a fruitful way of helping people to see with fresh eyes. It's a kind of poetic exercise is which you say something unexpected as a way of helping people see more deeply into what is.

  Edward Vielmetti [08.11.05 12:39 PM]

Tim, from the perspective of someone who was selling people access to the net back in 1993 and who worked on Gopher and WAIS and the various goo that held things back together then, the big application at that time was email. All the web or pre-web stuff was of only a little hard core interest to people who were dabbling with hard to understand things, but email was clearly better than fax machines and was starting to show up on business cards (even here in Michigan).

The 1993-era web servers were often really funky mashups of data warehoused on some mainframe and icky code to get it out to look pretty on screen. Fast forward 12 years and most of the data still gets formatted to look nice on the screen and only a tiny sliver is actively being served up with the intent that it can be repurposed by some other application.

What's more, a lot of the standards writing and consensus development that went into 1993-era development has been overtaken by open but proprietary APIs and poorly documented extensions to consensus standards. (though perhaps it ever was)

  Jared [08.11.05 01:04 PM]

Personally I think the term "Web 2.0" is a valid one that organizes a lot of what is going on today. I would also argue that the version number is correct, that Google actually represented the pinnacle of "Web 1.0" and the seeds of "Web 2.0" in their user based algorithms for determining the quality of a given web page (the long tail).

I've posted some more thoughts on this over at my blog:

  Danny [08.12.05 01:28 AM]

Re. Computers 2.0 - I think you've already got this one covered by "Internet Operating System". What I find disappointing about the current Web is how much of it is merely a reimplementation (albeit a highly convenient one) of pre-computer communication modes. Web pages as electronic books, blogs as electronic journals, connected devices such as PCs as glorified telegraph. A characteristic of Web 2.0 I would hope to see is more *computing*, primarily done in the shared virtual machine.

It's interesting that you talk of the current discussion as a "conversation". Conversation is speech mode. Following the telephone analogy, maybe it can stretch to apply to an online debate between two individuals. But it's not very Web 2.0, whatever the Cluetrain PR agency might say. (And while I'm about it, "worse is better" should be left to the medical profession ;-)

  james governor [08.12.05 09:53 AM]

web 1.0 - content produced by someone else
web 2.0 - content produced by the user

web 1.0 - architecture of consumption
web 2.0 - architecture of partipation

web 1.0 - attempts to create walled gardens
web 2.0 - building value through open fields

web 1.0 - download culture
web 2.0 - remix culture

P2P as "hype" was an inflexion point here

  Peter Nixey [08.12.05 10:45 AM]

When I first heard of the web in ’95, I thought a lot about the consequences of networking the computers sat on every desk. I heard talk of how that virtual-proximity would remove barriers to information-flow and leverage many hands to make light work of large problems, how new communities would form and ideas be exchanged. All of this was true but not nearly as much I expected.

For most people, the true potential of online communities is something they read about only in the Sunday newspapers. Dave Winer may be sought out by readers in every village he visits but for most people it takes years to reach out in even a city and retirees and young mothers find themselves sometimes unbearably lonely once isolated from those physical networks.

For the majority of people, the human-networking potential of Web1.0 never emerged. Everyone uses the web but ask most people what they do with it and they’ll tell you that they use it for shopping, email, research and news. For the man in the street, the internet revolution was arguably less of a revolution than an evolution.

Web1.0 explored what was possible when we took out the human middle-man and allowed our computers to connect direct to computers and data in other networks. It was about what became financially and technically possible when we no longer needed a remote operator in order to interface with a remote machine or a physical network to deliver remote information.

That’s not to say that machines couldn’t communicate before ‘95. Be it counting handle-turns or pulses of light, machines have always had a language of interaction. Nonetheless, the creation of one common language (or triumvirate of languages) lowered the barriers sufficiently to mean more people and machines chose to connect. At the same time, servers allowed those machines to safely expose useful data from their vulnerable interiors to the world at large. More information and lower barriers created more entrants, more entrants created more still and Web1.0 cascaded into existence.

Humans have always been able to communicate and network but when we look to our virtual networks, the lockin of email and the hassle of publishing and reading has kept most of them ultra-private and defensive. Those defence mechanisms are so strong we will sooner reject an email from a friend than let risk spam and even when we want to, we will sooner sit in alone and in silence than chance an interaction with a stranger.

Humans are the network that Web2.0 connects and RSS is the common language. RSS is to conversation what TCP/IP was to data and blogs are to humans what apache was to computers. We have always been able to communicate but RSS makes it that little bit easier and more reliable. RSS lowers our threshold to reading and hence raises our audience for writing whilst blogs allow us to put a little bit of our sensitive data outside our personal firewalls and to give others something to interact with. We all start to realise that giving a little of ourselves can bring far more in return. More interaction breeds more ideas and RSS and aggregators give those ideas a network they’ve never before known.

TCP/IP was DNA for data on a network, it guaranteed its delivery and integrity. XML and RSS are fast becoming the DNA for ideas on a network. Words become nucleotides, items become genes and just as with real genes, we are the vehicle to take the best of them to the next generation. Some survive, some multiply, some will be lost and some will never get read but one way or another, RSS lets them evolve faster than fruit flies.

The blogosphere is the detailed transcription of an entire population’s intellectual DNA, an exhaustive textual, audio and video description of the ideas and emotions of a community. The blogosphere and what it will evolve into become not just the zeitgeist but the very definition of the human state. If the handbook for the revolution is Surowiecki’s excellent “The Wisdom of Crowds” then note that the blogosphere is the crowd - ignore its wisdom at your peril.

Where Web1.0 allowed us to connect to remote computers without an analogue call centre worker, Web2.0 allows us to connect to remote people without an analogue journalist. Despite what’s bandied around this don’t mean we’ll dispose of all journalists and publications any more than online news meant we disposed of newspapers. There will be change for sure but good journalists add value far beyond the information they transcribe and those who think it’s all change are more naïve than those who believe it’s none.

Some argue that Web2.0 really isn’t an event - it’s incremental and undeserving of a name. In some ways they’re right. 2005 and “Web2.0” isn’t where rich human networking started any more than 1995 and the “Information Super-Highway” were where computer networking started.

“Web 2.0” applications were around long before today. EBay was 2.0 and so were Citeseer, Friends Reunited and P2P. Spam email may have been 1.0 but the “I love you virus” is 2.0. Amazon isn’t a 2.0 company but its reader reviews were. Page rank wasn’t a Web2.0 application though auctioning Adwords certainly was.

So what is it that makes Web2.0 a big deal? It’s a big deal because people say it’s a big deal. Lots of people. Lots of people are blogging, lots of people are tagging and the number grows almost exponentially. When a lot of people do something one month and twice as many people do it the next month, it makes it a big deal. The participation in Web2.0 has grown well beyond the founders-circle and shows no signs of slowing down.

And why give it a name? Well when lots of incremental things start happening in a short space of time and when those incremental things build into a qualitative change of state we need a way to distinguish the new one from the old one. Web2.0 seems a very apt descriptor for where we are about to arrive.

Tagging and blogging are the poster children of 2.0 and it’s interesting that many still dismiss them as they dismissed the directories and homepages of ’94. A few years ago they would have been credible. Not today though. Today there is too much noise from too many directions and too many demographics.

The bloggers of today aren’t scouts looking for water and flatlands, those pioneers came and settled back in the late nineties. Those that you see now are the forward party of a full blown caravan and land is being settled. There’s a city being built and I imagine that anyone who doesn’t settle right now will soon be unable to afford the land they stand on.

Web 1.0 was about connecting machines. Web 2.0 is about connecting humans. For my money, when it comes to interaction, human beats machine every time and my time, my emotions and my livelihood are all staked on Web 2.0 being a very Big Deal.

  Eric [08.12.05 02:50 PM]

O'Reilly is in the biz of organizing conferences and bringing people together, and I think in this context, putting a term out there is valuable, precisely because we don't have a clear definition for it. The big picture isn't clear to us, but we do need to talk about it. It's hard to keep momentum going on the conversation without a label or meme to refer to it as. The Web 2.0 meme provides a space for discourse, and for this reason it's valuable.

  Josh Owens [08.13.05 03:38 AM]

We are starting a podcast conversation about web2.0 and the various people who are creating tools and/or standards for web2.0. Our first interview will be Matt Mullenweg of wordpress fame. If anyone is reading this and would like to request an interview, we are looking for more people.

On another note, I wanted to mention microformats and point out that I think it falls squarely under the heading of web2.0. I think the ability to markup data such as calendar events in a special way will lead to many slick web2.0 applications in the future!

  raj [08.14.05 05:08 PM]

All this "web 1.0" and "web 2.0" terminology is pure marketing hype, as has been stated several times, and thus purely ridicle-licle-lous. It's as stupid as using the term "human 2.0" (or whatever). As humans, we grow and adapt, without versioning ourselves (besides the term homo sapiens). Why bother doing the same for the web? It IS going to grow and "improve", and we'll all have to adapt to it - regardless of what silly version number is attached to it. It's not like we wouldn't know that, hey, there's something different about the web today, as opposed to not necessarily knowing the difference between two versions of, say, Perl or PHP.

raj kumar dash
cto, chameleon integration systems
homo programmis :)

  Alex Barnett [08.15.05 04:29 AM]

Since Trackbacks are switched off...I'vr listed the differentiating characteristics between the Web 1.0 and Web 2.0 concepts as I see them here:

  Andrewa [08.15.05 07:39 AM]

If by Web 2.0 we mean the web as platform, then Web 2.0 is 10 years old this month.

  Tim O'Reilly [08.15.05 10:41 AM]

Andrewa - of course the web as platform is as old as the web, just as each of us has within us so much of our adult potential when we were born. But no one would question that there's a difference between a baby and a child, a child and a teenager, a teenager and an adult.

What's different about the "web as platform" aspect of web 2.0 is that Web 1.0 saw the web as platform as centered on the browser (Netscape's "webtop vs. desktop" vision), or else centered on individual sites. Even eBay, Google and Amazon web services are still centered on the originating site. The reason why I call the first true web 2.0 application (in terms of the web as platform axis) is that it's an application built out of two cooperating web sites, by a third party, using the first two as components.

(Ad servers like doubleclick have built cooperating "applications" since they first came out. But all that illustrates is that there's no hard and fast dividing line between web 1.0 and web 2.0. It's more that at some point the predominance of the evidence tells us that something new is afoot, new enough that a new term helps us to crystallize the ideas out of solution.)

  Arnaud Leene [08.17.05 03:04 AM]

I order to get more grips on the Web 2.0 meme, I created a Web 2.0 checklist ( I am applying this checklist to various services on my MicroContent blog. Maybe this sheds some more light on Web 2.0.

  Douglas Reay [08.21.05 01:55 PM]

The key is scalable trust.

First order unmediated trust is where person A decides how much or little they trust a statement by person B based upon their own assessment of person B's previous statements in that area.

Second order human mediated trust is where person A decides how much to trust a statement by person B based upon person C's assessments of person B's previous statements in that area.

Third order computer mediated trust is where person A decides how much to trust a statement by person B based upon a calculation their computer makes based upon various criteria person A has previously given the computer and upon assessments of B's previous statements made by persons D, E and F none of whom person A may know anything about.

When that gets applied not just to shopping recommendations but the wiki/blogosphere, then we'll see Web 2.0

  Alan Kelley [09.05.05 06:48 PM]

I continue to think about this whole idea of the Web as a platform, what some are calling Web 2.0.

In the late 90s, the strategy of many companies was to develop network effects that would allow the companies to dominate a particular vertical and then extend into other verticals. For example, AOL hoped that greater quantity and quality of content would attract more
viewers, which in turn would attract yet greater
quantity and quality of content, and so forth. Once AOL established a critical mass of viewers, it could leverage its brand and financial resources to move into new markets / verticals.

Metcalf's law was the underlying principle for
Internet business strategies of the late 90s -- and in some cases the strategy of the late 90s has worked out well.

I think today that highly effective strategies will involve moving into new markets AND providing tools for others to build their own things - businesses, social networks, etc.

So far, the Internet has provided tools primarily for creating, searching and retrieving content. And many individuals have used the basic tools that are sourced from the Internet to build interesting companies, etc. But I think Web 2.0 will be about providing an EXPANDED set of tools that can be sourced from the web and used to build things. Google's AdSense is a great
example. It's a new kind of web-sourced tool, and it helps a company build an entire advertising campaign.

By the way, there's a book called "Fab" that talks
about distributed tools for building tangible objects. Think of creating a coffee cup, lamp or a running shoe from a device hooked up to your computer. Apparently, the possibilities of this happening are greater than people think. I've skimmed parts of the book. In reviewing the book, Jeff Bezos says the ideas discussed in Fab are going to dwarf what we've seen from the Internet so far. Anyway, the ideas in
Fab would be like the Web 2.0 on super, super steroids - distributing over the Internet capabilities to build things.

The significance of all of this? It's like going from providing tools to build and connect villages to providing tools to build and connect

My thoughts haven't crystallized, but I find this a fascinating topic.


  John Hagel [09.27.05 09:59 PM]

Tim - Your great posting prompted me to make my own stab at capturing the essence of Web 2.0 available here
Any comments or reactions would be greatly appreciated.

  Karl Hungus [10.05.05 11:02 AM]

Bottom Line: You use the term Web 2.0, you sound like a douche bag. Honestly you sound like a really big douche.

  Vic [10.06.05 05:36 AM]

How do you explain a spreadsheet to somone who does not use it? It's great callculator?

No. You can SEE web2 demo at

  matt [10.13.05 05:06 PM]

Web 2.0 is a moniker placed on one of many paradigm shifts technologies go through. It will be used as a buzzword/catchphrase until the next moniker comes along. It will lead to headlines of how Great, New, Cutting Edge, Innovative, ad nauseum "Web 2.0" is and will be in our daily lives. It will also lead to headlines indicating that "Web 2.0 - still in beta!" and "Web 2.0 - emperor without clothes?" in days/weeks/months/years to come.

It's human nature to categorize, "Web 2.0" is the category of now.

  Pierre [10.18.05 09:11 AM]

I'm sorry but I've seen too many web "fads" and titles come and go including web 2.0. The fact is the web is like any other commodity or product. If you make something useful that people need and it gives them results they will use it or buy it.

Everything that people talk about which makes up web 2.0 are products or services that just work. The fact that there's more of them simply means that developers are begining to realise what the public wants out of the internet and we're getting better at filling everyday needs and not focusing on what's cool or hip.

This whole "web 2.0" just smells of the dot com bubble. If we don't want to implode again like we did in 2001 we have to let go of this attitude that the web is somehow different than any other industry. Or that we need to define ourselves with labeles.

The web is a service based and product based industry, just like cars, books, restaurants etc... You make a good product and people will want it, period. There is no self importance, there is no "movement". There is no "new era". The reason why the web has such a hard time being accepted by big industries is because we inist on putting on these airs of self importance. By doing so we prove that we still don't get it.

The web is product based. Without a good product you won't sell. The lables we give ourselves just makes us less credible and alienates the public.

  Davide [11.09.05 09:14 AM]

I discover really amazing news about web frontiers looking for "WEB2.0".

Now, after reading a lot, I know it can be not a good name for what it's going to happen but it was my first footstep to follow.

I hope it can help other people like me =)


(this is my first life post on a blog)

  Paco Marin [12.14.05 03:14 AM]

I understand that we could use a meme like Web 2.0 to communicate this is evolving, we are not at the same stage than five years ago, there is a new usage of what we have and it is useful to make money.

However, in my humble oppinion, it could lead to misunderstanding. The Web has suffer a seamless evolution where specific initiatives lead to implement new ideas like social/community services (what is really new in Blogging but a new successful service concept?). Are we going to claim there is a new Web wave/'versio' everytime anyone has a new idea to make money with it?

Web 2.0, such a meme would intuitively mean something disruptive, a new revolution. Tim, I know that you claimed revolutions are known once they are mature and started, and, How do you know, then, that is a real revolution?

I don't refuse using Web 2.0 to indicate a wave a successful business concepts. I wouldn't consider we are in a new stage of the Web evolution.

I enjoyed reanding you. Best Wishes.


  Luis Morais - Web 2.0 allergic [12.29.05 09:25 PM]

Oh, please.

You guys take the name of a global means of media and communications, label it with a version number and stablish it as a rule, oh sorry, as an attitude of what the web is for the whole world without even taking a look around the globe?

Since you guys started, what about TV 3.5, Democracy 5.0 and Radio 9.0 for the next discussions? Have fun kids, label whatever you want, the world is a little bigger and more diverse than the little imaginary bubble you live.

P.S.: Congratulations for the marketing stunt!

  Anonymous [04.11.06 09:11 PM]

i understand very well your point of view but it is interesting to distinguish an html page from more evolved content such as script,on the technical point of view

Some problems are more easy to resolve with scripting language,the counterpart is that the website is less acessible,for example to mobile phone or old computers

A great example of this is this page
it has a problematic and resolve it with scripting languages,the counterpart is huge because it can only be acedded by a gecho browser,and it is slow on old computers...

As you said theses scripting languages have been around since a very long time,a great example is forums...

But in a technical point of view(non commercial), this "Web 2.0" also make people realise the potential of such languages...that have been around since a very long time(java for example)

  Jennifer Hershey [04.16.06 02:14 PM]

Ok, what can see here - bloggers think they invented internet. I do agree with "The bloggers of today aren't scouts looking for water and flatlands, those pioneers came and settled back in the late nineties. "

  Craig Hubley [04.28.06 11:40 AM]

I'd avoid all this talk about "platforms" or the role of the user, and talk more strictly in terms of semantic webs as a form of database. The word "web" has a meaning in the context of information and knowledge management. Of course we can talk about today's World Wide Web or "public web" as "web 1.0", take the grandest idea of "service oriented enterprise" or "service oriented architecture" as "web 3.0" (as Jeff Schneider calls it), and then define "web 2.0" as something in between. Fine, that'll happen no matter what we agree on. But what's the essential thing that makes a "web" different from non-webs? Without that, you really have no business trying to state that there is some sequence from 1.0 to 3.0 etc..

I argue it's the links, the fact that expectations of responses like "what happens when I press" link names, or (to a growing degree), buttons like "post" or "preview", are growing among users in common. It's those expectations that define a web. I was designing user interfaces which put the REST verbs (put, post, delete) on buttons in 1996. So yes web 2.0 is ten years old by any of the "platform" definitions, and that platform is REST.

Meanwhile, the "GFDL corpus" of text that is contributed via all sorts of projects (but mostly via Wikipedia) is a "web", because it has a lot of links and a very large number of associated words (like "discuss" and "version" and "revert" and even "troll" and "vandal" and "neutrality") that also have operational meanings (though often disputed). THE TECHNOLOGY DOESN'T MAKE IT A WEB.

The links do. The words on those links do. The same words on the buttons do. The same words in the URIs (web addresses) do. A web is a coherent body of links named (or semantically "typed") with a coherent type of behaviour recognizable by users as being "the same" for things of "the same type".
THE USER "COMMUNITY" IS IRRELEVANT. IT'S THE DATA STRUCTURE (wiki) AND THE AFFORDANCE PROVIDED BY THE VERBS (REST). It's the infrastructure that the users leave behind, like the many generations of a termite mound. Go read Lewis Thomas on that!

So, yes, web 2.0 goes "beyond page metaphor" and has all those technical advantages (uniform semantics at least for agile computing and loosely coupling a lot of web services together, lets software be delivered as services, applications on-demand etc) but let's learn from Microsoft:

- version 1.0 is so bad you just have to change it anyway and it attracts all the complaints in the world, and eventually crashes out in a stock market bubble because absolute crap got funded and it was distracting too much talent from real problems

- version 2.0 actually defines the concept that you are trying to fix, and mixes all metaphors so badly that its unusable, attracting competitors and more specific complaints - but the absurd technology-driven definitions waste everyone's time and discourage investment as everyone tries to claim that something UNDER the data structure and verb is central or critical - when it can't be - you get to "mainstream" intranet/extranet with this technology but you can't do governance or even management through that interface since there is no unifying central core of this federated web

- version 3.0 limits the metaphors to those that are compatible, and rely on a very few verbs (HTTP or REST principle) and only one or two basic types of process (updating webs, browsing webs) if that - it becomes impossible to fund anything that is not obeying these industrial standards that evolve from the most difficult applications (government and politics, in my opinion, where decision making is most constrained) and the most successful of the federated webs (GFDL corpus access providers, cooperating green movement entities, those working to standardize a single tag cloud, specialized use of terms to support profession-specific searches)

- version 3.1 actually works and takes over since it's so focused on only those very few principles and is proven in those applications - it begins to be trusted to make actual decisions at least in those domains - but eventually in serious domains

Personally, I'm spending my time on that elusive single interfacing language. But not to knit the applications together. To knit groups together, and to characterize their common goals in terms of a common issue language (issue/position/argument), that has a hope of neutrally representing issues for decision, as the current "blogosphere" can't, and the current "neutral point of view" pretends to (but also can't).

What differentiates the economics of this "boom" from the previous ones, is that there is actually a pre-existing basis in economic analysis that explains the shift and the motives for it: green economics, which is classical economics as if the natural capital mattered, characterizes economies ONLY in terms of services (there is no "commodity" nor "product" because these imply services at the very least from nature, which provide the inputs and clean up the wastes and keep air and water and pollination flowing to the fields, to the workers). This "service economy" happens to already have a "service-oriented architecture" underneath it, that being, the ecological structure of nature's services (read Robert Costanza on that). What you are doing by building good "web 2.0" and "web 3.0" applications is minimizing use of energy and waste by moving decisions into a realm which is easier to support without humans moving around, and making resource flows and allocations easier. In economic terms, you are solving the information problem that prevented socialism from working. If corporations are just socialist entities, and they are INTERNALLY, then, you are just providing more feedback so that the larger organism can function with something less than 2.5 hours per day wasted "looking for information". My own experiments show that can be cut in half by prudent quality management on your taxonomy or ontology, and in half again by disciplined use of even short words. If you need less than half an hour per day spent on "looking for information", or need to do your organizational governance or operate on a 24x7 basis globally, you'll need to basically go to a zero base solution and redesign the protocols of your organization.

But in any case, expect the big web2 and web3 breakthroughs to come from inside organizations and in nonprofit consortia that wish to outflank their political opponents. Not from the public web.

  Anonymous [05.12.06 06:46 PM]

So, Dale Dougherty is the moron I can go shoot for inventing this over hyped "web 2.0" crap?

There IS NO WEB 2.0! When will people stop using this stoopid and meaningless term? Show me a W3C draft! Show me a IETF draft! THEN maybe "web 2.0" would actually be some sort of tangible technology. Right now, well, it's a buzzword that marketing types throw out so they can go after VC. Same marketing types who burned all those investors so bad in the first .dot com bubble.

Way to go cheese!

  Tim O'Reilly [05.12.06 08:59 PM]

Anonymous, I'm sorry that you're so literal minded. There are many things that are "real" that don't have W3C specs. (There's no such thing as a "scripting language," for instance, by your logic, or an "operating system," only specific instances of such things.) Names are pointers, and if they point to something meaningful, they are useful. They can also be misused.

Dale is not responsible for those who misuse the Web 2.0 concept, nor am I. I believe that my "What is Web 2.0?" article lays out a set of key principles that distinguish web applications that follow them from those that don't.

And yes, there's been a lot of hype. There was a lot of hype about the web before, too, but it didn't stop those who understood what was really important about it from outperforming those who merely embraced the hype.

  Tony McNamara [05.28.06 07:15 PM]

What is Web 2.0? Why it's a marketing slogan. Just as I thought all along. The functions and possibilities emerging currently on the net are evolutionary. The Web was and remains revolutionar.

  Andrew [06.01.06 09:53 PM]

This is all interesting stuff, and I'm glad to see such thorough debate and analysis. However, I see two conflicting claims about Web 2.0 at its heart:

1. that it is something exciting and new, and is revolutionary compared to what happened in the dot-com bubble

2. that it is essentially the best practices of those companies that survived the bursting of the dot-com bubble

How do you resolve these competing views of what Web 2.0 means?

  Joe [07.04.06 09:52 AM]

In reflection to Tim O'Reilly's comment posted at August 15, 2005 10:41 AM:
But no one would question that there's a difference between a baby and a child, a child and a teenager, a teenager and an adult.
But no one would give a new version number to a baby just because she/he walks already! Imagine Tim 2.0 who walks Tim 3.0 who reads and writes etc!
(I am really missing the opinion of Tim Berners-Lee from this site, and from the W3C site too.)

  Frank [08.08.06 07:26 AM]

Concerning all the Web2.0 stuff here: it�s just a marketing-slogan! Nobody can really explain what it really means. From my point of view it�s just an addition of new aspects with some possible interaction based on the Internet....

  Allen Razdow [09.28.06 07:51 AM]

Is it the "2.0" part that offends those partial to technical taxonomies? Many seem to equate naming a trend with "marketing"; and marketing with "hype". It is neither. The web is unique because it is by definition collectively developed, and there is no question that developers are in a new phase beyond presentation and forms. Search is on the boundary because searching was part of web-1 first. But semweb, blogs, rolls, mashups, RSS/Atom, wiki and on and on are hard to see as anything but "second generation" riffing on the original web concept, and calling that Web 2.0 makes a lot of sense. I think you have to be pretty grumpy not to see and go with that ;-)

  rajat [03.19.07 10:55 PM]


From the above discussion I have following observations:

WEB 2.0 is NOT a technology
WEB 2.0 is NOT a tool
WEB 2.0 is a Name given to identify the beginig of a new era of Internet/web when it has started moving in a direction of enhancing User experience with the use of latest concepts (not technologies e.g. wikis, ajax, blogs etc...) to evantually make the user experience as good as (if not better) using a desktop application.

Let me know if any one disagrees to that.. and help me improve my understanding..

Thanks & Regards

  Adam [04.13.07 08:47 PM]

"I guess it's the old debate between language purists, and language pragmatists..." Here a language purist would point out that is rather the debate between prescriptivists and descriptivists.

  Lewis W. [06.13.07 11:14 PM]

I really like the concepts of Web 2.0 and have enjoyed living through them. However, I work for a web company that runs around all day saying "we need to create something Web 2.0". I'm not mad at Dale or Tim, rather the people who use it without making any attempt to understand it.

Who am I to talk though, even reading each of these blog posts I still wouldn't say I confidently understand it myself!

  Timothy Fish [07.05.07 05:19 AM]

It is interesting that Tim O'Reilly wrote the article "What is Web 2.0" nearly two years ago and we still have questions about what it is and whether we need the term. I would like to offer the following as a concise statement of what I believe Web 2.0 to be:

Web 2.0 is the subset of the web that consists of the web-based sites and services that facilitate broad collaborative efforts by users and machines with web access in producing, recording and disseminating collective knowledge.

A few years ago, with Web 1.0, so to speak, the internet consisted primarily of websites that had information. The user was free to access the information, or not, but there was very little notion that they user might provide feedback. User feedback was usually limited to a GIF image of a mailbox that linked back to someone's email address. Under Web 2.0, websites handle user input differently. In collaborative effort, users now built whole websites. Websites allow users to post comments and other users may respond to those comments. Multiple machines rather than one device may distribute large files. Websites are now much more than a place to gain information. They are now a place to facilitate communication among users. Take as an example. Their goal is to sell books and other things, but they are doing it by allowing their customers and suppliers to use the site to discuss the products and other things. I would suggest that they are a prime example of Web 2.0.

  Tim O'Reilly [07.05.07 10:26 AM]


I agree. I call this aspect of web 2.0 "harnessing collective intelligence," and you've got a good description of how sites are different in this model.

But it's more than just comments. Google is a great example of a collective intelligence application where no one thinks they are explicitly contributing, but are doing so nonetheless. Implicit "architectures of participation" are often more powerful than explicit ones.

  Greg [08.29.07 06:49 AM]

"Don't believe the hype"... This comes out of one side of my mouth while "This changes everything" comes out the other side. My job is to help CMO's reach their target to influence people to buy the stuff they make. A model that worked for 50 some-odd years. It's breaking down. Marketers are scared. They don't understand. They don't know what to do. In this vacuum comes the shrillers and fortune seekers and they convince Coke to create a community site on their global home page. They convince Bud to create Bud TV. Massive multi-million dollar investments with little to show for it. The cover and lead story on AdAge is about community almost every week.

My advice to CMO's - be wary of people who claim to have the answer. Tim's reference to the blind man and the elephant applies to marketing as well. Experiment, make a bunch of small investments and see what works. Let go of your brand, let down your hair, talk conversationally, and admit mistakes openly. Hire people that know digital. Web 2.0 is changing everything - but (aging myself here) as they said on Hill Street Blues "Be careful out there."

  Martha [09.11.07 10:56 AM]

Great article, thanks for all.

  Sagar [09.21.07 12:33 AM]

I remember the story of elephant I heard in my childhood. The story is about how four blind people accedently bump upon an elephant and then describe it in their own way as they touch different body parts of the elephant. Actually, they all were talking about the same object!

In my opinion, same if the story of Web 2.0. We all are talking differently, but essentially the same thing! To me, Web 2.0 is nothing like a rocket-science. It is simply a very intuitive, fast and customizable user interface, intertwined with high performance (yet easy to use) software tools.

  saravana moorthy [07.11.08 05:11 AM]

I thought its a new technology and later came to know that its not a new technology, new version, guidelines etc...

You mean that that a web2.0 design should have a pleasant banner, glossy buttons, white space, good typo,(dont know if there anything should be added) all together a neat appearence...but some of the designers developed websites based on these concept before the term web2.0 came itself.

Hope you agree with me?

~saravana moorthy

  Mich [08.07.08 09:30 AM]

I have now read through every comment on this list going back over 2 years and I have come to the conclusion that, while we are still pondering what web 2.0 is, it is no more than a website design trend to include more features (tools) that allow visitors to interact with the site and sometimes adjust the site feel and function to their individual needs. Not every site has the same audience. Not every site has the same purpose. So before the rest of you get carried away with the need to design a web 2.0 website and spend the $$ to incorporate the features that claim to be 2.0, remember what the main service your website is to provide and ask yourself if there will there be a ROI for all the fun, flashy, intuitive features you want to add that will make you feel like a trendy designer. Is that what your audience really wants or needs? In some cases it may actually be a turn off to using a website.

  Kishor Singh [11.20.08 03:43 AM]

My concept is same when you say that web 1.0 was the Netscape but I goes in different direction when you say that web 2.0 is the leadership of computer. Here I want to say that web 2.0 is the evaluation of Web collaboration with the participation of people.

On the second place you say that Google Page Rank was the important part of web 2.0 but I say it was the blogging and social networking in web 2.0.


Post A Comment:

 (please be patient, comments may take awhile to post)

Type the characters you see in the picture above.