O'Reilly Network Weblogs: What the hell is Web 2.0? The great web mash-up begins.

Published on The O'Reilly Network (http://www.oreillynet.com/)
http://www.oreillynet.com/pub/wlg/7825

What the hell is Web 2.0? The great web mash-up begins.

by Mark Sigal
Sep. 16, 2005

People simply hate overloaded terms, and the term "Web 2.0" clearly fits the overloaded bill. Then again, how else do you introduce a discussion on something that is real, emergent, exciting and -- open. And do it in as few words as possible?

So what is Web 2.0? At the core, it is an applied web service model that blurs the line between software and service. It can do this because: 1) it is optimized for the 60 million broadband connections in place; 2) it can count upon an installed base of 300 million video-ready mobile and PC devices; and 3) Thanks to the AJAX meme, it can reliably assume the ubiquity of a really good browser experience.

So that answers the "HOW" side of the equation, but "WHY" now? Simply put, the emergence of the blogosphere has changed the equation. First off, its footprint has become really meaningful. At last count, according to Technorati, there are 16 million blogs in existence, growing 100K new blogs a day and generating a jaw-dropping 1.2 million new posts a day.

Second, the mainstreaming of RSS is enabling syndication and subscription systems that can intelligently process context-aware messages. Over time, these systems will become adept at handling rich content "payloads," enabling further innovation. Slide is an early example of one such application, and I covered the RSS side of the Web 2.0 equation in an earlier post.

But the real shocker in all of this is that people really seem to enjoy generating lots of custom content and then sharing it on a broad scale, proving that the open source movement is not just some techie phenomena.

As a new class of photo and video cameras gain support for wi-fi based upload capabilities, this will remove a perceived hurdle to getting multimedia content online (a challenge that mobile devices happily avoid), fostering creation of some really compelling digital media services.

The Long Tail provides an excellent frame for understanding how all of this plays out in market terms, but let me say that knowing that a ready supply of rich content is going to be available to provide the "kindling wood" for igniting new application services removes a key risk item for would-be entrepreneurs.

Ponder this. Over the next 18 months, not only will the process of creating all sorts of application functions that run seamlessly between PC and mobile environments become more straightforward and require less heavy lifting on the part of the developer, but the tools for managing and sharing the output of these environments will improve markedly.

In my mind's eye, I see a kind of Mad Libs style applet running in my browser. It supports visual, argument-based modeling of procedural workflows. These workflows are essentially IF-THEN-ELSE sequences, and their primary job is to automate the function of syndicating, finding, subscribing to, uploading, downloading, rendering and presenting rich information flows wherever and whenever I need them.

The mash-up part of this equation (as referenced in this post's title), is the offspring of an environment where application developers see it in the own selfish interest to facilitate the creation of integrated, yet highly derivative application hybrids by third parties, something they do by providing rich public APIs to their user base.

Along those lines, the good news is that the "gorillas" of Web 1.0 -- Google, Microsoft, Yahoo, Amazon and eBay (having just captured Skype -- see "A different spin on eBay-Skype") -- all see it in their self-interest to expose broad swaths of their capabilities to developers for the creation of composite web services.

The reason they are doing this is simple. By de-coupling their secret sauce from the web site and enabling it to be propagated throughout the web they can cast a bigger aggregate footprint and more importantly, capture a greater percentage of consumer mindshare during the day.

As a result, beyond unleashing ever-richer API availability, expect to see growing libraries of widgets made available by these same vendors to simplify and expedite the process of instrumenting APIs (see "Yahoo's Konfabulator buy and Yahoo APIs"). Once again, more examples of kindling wood to ensure that the Web 2.0 fire burns bright.

With a nod towards embracing open source as a good business strategy (in addition to a good human one), the goal of the remainder of this post is to try to figure out all of the obvious "jobs" that this new application services model can support.

As a good blog post called, "Stealth Start-ups Suck ," compellingly argues, with web service creation, the bar is low enough to just start building out the idea, learn as you go, and test out your thesis about why your creation matters in real-time.

My hope in writing this is that the straw man that follows can inspire you to have fun, test out some ideas you may be percolating and iterate from there. I have compiled a list of manageable content items that, while I attempted to make it exhaustive, realistically you have ideas in your own head that I haven't contemplated and/or captured from others.

Think simple, systematic, and programmatically actionable:

1. Calendar items: include time based events, containing descriptions, locations of the event, mandatory and optional participants, needed resources, beginning and ending event times, reminder flags and any confirmation ID's, as in the case of reservations.

2. Product listings: these include things like product name, identification number (e.g., SKU) and a product photo, the manufacturer, product description, any categories that it is a member of, pricing, where to buy, etc. A logical container of multiple products is a wish list.

3. Business listings: include the name, address and phone number of the business, the type of business it is and key information about the business, such as hours of operation, menu (in case of a restaurant) and a photo/map. A logical container of multiple listings is a rolodex.

4. Classified listings: include products or services that are available for sale, including a description and photo of the offering, its condition, price, data and contact information on the seller, and whether it is available on an open ended basis, limited to supplies on hand or to the highest bidder (as in case of auction).

5. Property listings: include a description and photo of the property (e.g., number of bedrooms, baths, parking and noteworthy attributes), whether it is available for rent or purchase, the price and geographic area of the property, a map and contact information, as well as if/when an open house is scheduled.

6. Travel listings: typically these will include, but not be limited to, car rentals, hotel reservations and airplane flight booking functions. Parameters such as date(s) of service required, class of service desired and limiting attributes, like no smoking or non-stop are integral. Also, relevant is whether the individual is entitled to a discount based on coupon or group membership.

7. Job listings: includes a description of the position, the company, where the company is located, requirements of the position, the compensation package, how best to contact the company and whether communications with recruiters are encouraged.

8. People listings: whether oriented towards personal communications, such as dating, or professional networking, these listings can include data such as hobbies, skills and professional backround, education, age, marital status, geographic area, people in a trusted network, favorites (books, movies, music) and personal references.

9. Reviews and summaries: include written overviews of products, TV programs, movies, restaurants, etc. Summaries tend to be more objective in capturing the key qualitative and quantitative details of what is being written about, whereas reviews include a subjective component, ratings on one or more parameters, and in a community setting, aggregate (averaged) ratings.

10. Destinations: include maps and itineraries. Such items might logically include a list of key destinations, when they are to be visited, directions for getting from point A to point B, transportation options, such as mass transit, businesses and landmarks adjacent to a given destination and geo-coding tags that indicate the presence of members of a person's trusted network in a given area.

11. News items: include full articles, news summaries, sports scores, schedules of TV programs, movie times, weather conditions, and stock prices. A logical container for a group of stocks is a portfolio.

12. Coupons: these can be instant coupons in the sense that you can print them out and walk into a physical store or go to a web site and take advantage of them immediately. They can also be cumulative coupons that kick in after multiple conditions are satisfied or triggered by a single event. In all cases, coupons should have source identifiers so that it is clear where the consumer found the coupon, and definitions of fulfillment options and expiration dates.

13. Advertisements: in the Web 2.0 world, ads are designed to offer more signal than noise. As a result, they are more descriptive in terms of what the offering is, who it is targeted at, what the specific benefits are to the customer and what discrete actions are available to them (e.g., request more information, view data sheet, buy now).

14. Polls: whether user defined and published or systematically managed and broadcast to an entire community of users, polls track, aggregate and demographically segment opinions on a given topic.

15. Content libraries: include pictures, music and video clips, captured web pages, saved blog posts and archived records, like receipts. Logical containers include photo albums, slide shows, specialized multimedia players and file folders.

16. Business-to-consumer communications: include structured, systematic interfaces for vendors to communicate with their customers, and vice-versa. One example of a logical realm where such communications are necessary is technical support, as simplifying the process of finding the latest versions of software updates associated with a given product, known bugs, workarounds and how-tos reduces support costs for both customer and vendors.

In closing, the goal of this post was to define the atmosphere that Web 2.0 applications will operate within, make some assertions about what the moving parts are and leave the reader to contemplate ways to create more than the sum of the parts value around such a model.

Some quickie takeaways for me personally are:

1. The blog creation process will evolve to enable consumers to create "structured" posts along the lines referenced above and systematically share them with like minds. Further, the elegant simplicity and openness of Craigslist suggests that this evolution need not place a heavy "usability tax" on the consumer.

2. To the extent that the consumer is now at the center of the equation, the process by which consumers write reviews or summaries to be published on a vendor's site will change. The consumer, not the site, will own the review, and he/she will be able to centrally manage its message. This suggests support for either a "write once, publish anywhere" model or a structured carbon copy function.

3. Core comparison, customization, communication and spend/transaction functions, including "handoffs" between PC and mobile spaces will become embedded and invisible to the point that the notion of an on-demand computing model is a given.

4. Security is a REALLY big deal in this model, inasmuch as when information flows unbounded between applications and users, the potential for serious mayhem is huge.

So what the hell is Web 2.0? Let's figure it out together.

Mark Sigal is a seven-time entrepreneur whose focus these days is on digital media services; namely the re-envisioning of traditional video, audio, and advertising channels in a mobile broadband-enabled world.