November 13th, 2006

Web 2.0, 3.0, X.O should focus on the IT green field

Posted by Larry Dignan @ 9:29 am

Categories: General, Web Technology, web2con

Tags:

Down a few pints of Web 2.0 dispatches, ponder The New York Times’ effort to coin Web 3.0 and you’re left with a nagging question: What parts of these aforementioned revolutions are going to affect the information technology department of the future?

While folks hop on the Web-2.0-3.0-go-round, which was recapped here by Dan Farber, the whole discussion needs to be focused a lot less on the VC dough and a lot more on how this stuff is going to make IT more efficient. How exactly is a mash-up going to help my business?

It’s a tricky question. Pitch a few CIOs on mashups, Web 2.0 concepts and lightweight apps and you’ll get some interest, but it fades pretty quickly—after all these poor folks have to integrate and pare 1,800 applications (if they are lucky), manage legacy systems and fend off bean counters.  To get any real insight you need to rephrase the question: If you had a blank slate what would you do? Would Web 2.0 concepts be a part of it? If you had a green field to build your technology infrastructure what would the building blocks be? Would you start with Microsoft or bypass it? SAP? Oracle? Would you trust Amazon with your computing power and storage needs? How exactly would you utilize the semantic Web (you have plenty of time on this one)?

I don’t have the answers, but do plan on asking these questions repeatedly until the information systems architecture of the future emerges. Rest assured the lightweight applications that are a part of the Web 2.0 vision are going to nibble away at the big software vendors over the next five years—call it revenge of the small. Just like big media is getting thumped by small media, software will follow the same route. Chances are some startup that actually has the luxury of a green field to play with, will show the way. For now the discussion seems to be focused on whether Web 2.0 is bubblicious. (The answer: Yes.) Perhaps the discussion needs to be flipped. What would you do with that green field? Stay tuned. 

November 12th, 2006

Web 2.0 isn't dead, but Web 3.0 is bubbling up

Posted by Dan Farber @ 3:01 pm

Categories: General, Web Technology, web2con

Tags:

Nick Carr was at this best synthesizing the week that was–the Web 2.0 Summit–and he wasn't even there. I agree with other bloggers, commentators, writers, reporters,  pundits, podcasters, vloggers, journalists and user generated content creators who said that the Web 2.0 Summit, which I covered extensively, didn't produce any great revelations or a great step forward for the global Web. It was more about scraping money off the table as the money–VCs, IAC, Fox Interactive, Google, Yahoo, Microsoft and others–were auditioning startups in the hallways and in private rooms.

Nick is having fun calling Web 2.0 mostly over and heading straight to Web 3.0:

I'm claiming the trademarks on Web 3.0 Conference, Web 3.0 Summit, Web 3.0 Camp, Web 3.0 Uncamp, and Web 3.0 Olde Tyme Hoedown. 

Many people I talked to identified a lack of big innovations, little companies with features but not products, too many products crowding the same space, too many startups trying to horn into MySpace  and the lack of sizzle from the hand-picked companies that demoed their products in public and paid their appearance fee. But, it's hard to put down the fact that a lot of energy and creativity is at work, even if most of it won't make the cut.

The alternative is no enthusiasm or creative juices flowing, and we are back in time, reliving the desolation of the bubble burst era. Plenty of room for innovation exists, but the power is concentrating in a few behemoths dominating the landscape, which gradually tend to stifle innovation, slowing it down to the pace of the large and lumbering, and fewer choices.

Back to Web 3.0. There will be one, and it has been associated at this point with concepts of the semantic Web, derived from the primordial soup of Web technologies. It's been a focus of attention for Tim Berners-Lee, who cooked up much of what the Internet is today, for a nearly a decade.

Nova Spivack, founder and CEO of Radar Networks, defined the semantic Web in a long post on his blog that is worth reading:  

The Semantic Web is a set of technologies which are designed to enable a particular vision for the future of the Web – a future in which all knowledge exists on the Web in a format that software applications can understand and reason about. By making knowledge more accessible to software, software will essentially become able to understand knowledge, think about knowledge, and create new knowledge. In other words, software will be able to be more intelligent – not as intelligent as humans perhaps, but more intelligent than say, your word processor is today.

I won't argue with attaching the semantic Web to 3.0, although there may be something unforeseen in between what we have today and a Web in which the machines seem more intelligent, and do more than calculate inbound and outbound links to derive relevant answers to queries.

John Markoff triggered the discussion around Web 3.0 and systems that can reason in a more human way, in a recent New York Times article. He described the differences between Web 2.0 and 3.0:

The classic example of the Web 2.0 era is the “mash-up” — for example, connecting a rental-housing Web site with Google Maps to create a new, more useful service that automatically shows the location of each rental listing.

In contrast, the Holy Grail for developers of the semantic Web is to build a system that can give a reasonable and complete response to a simple question like: “I’m looking for a warm place to vacation and I have a budget of $3,000. Oh, and I have an 11-year-old child.”

While Web 3.0 might now have a concept to hang itself on, we will remain in the midst of the Web 2.0 era for several more years. The semantic Web is still incubating and will take many turns of the crank to become mainstream.

Markoff mentioned Spivack's Radar Networks and Metaweb, two companies developing technology around semantic Web concepts. Both companies are in stealth mode, and only talking in generalities about their product plans. 

In response to the attention the semantic Web has been getting since Markoff's article, Spivack recently posted about misconceptions layered on the concept:

The Semantic Web is not just a single Web. There won't be one Semantic Web, there will be thousands or even millions of them, each in their own area. They will connect together over time, forming a tapestry. But nobody will own this or run this as a single service.   

The Semantic Web is not separate from the existing Web. The Semantic Web won't be a new Web apart from the Web we already have. It simply adds new metadata to the existing Web. It merges right into the existing HTML Web just like XML does, except this new metadata is in RDF.   

The Semantic Web is not just about unstructured data. In fact, the Semantic Web is really about structured data: it provides a means (RDF) to turn any content or data into structured data that other software can make use of. This is really what RDF enables.

The Semantic Web does not require complex ontologies. Even without making use of OWL and more sophisticated ontologies, powerful data-sharing and data-integration can be enabled on the existing Web using even just RDF alone.

The Semantic Web does not only exist on Web pages. RDF works inside of applications and databases, not just on Web pages. Calling it a "Web" is a misnomer of sorts — it's not just about the Web, it's about all information, data and applications.

The Semantic Web is not only about AI, and doesn't require it. There are huge benefits from the Semantic Web without ever using a single line of artificial intelligence code. While the next-generation of AI will certainly be enabled by richer semantics, is not the only benefit of RDF. Making data available in RDF makes it more accessible, integratable, and reusable — regardless of any AI. The long-term future of the Semantic Web is AI for sure — but to get immediate benefits from RDF no AI is necessary.

The Semantic Web is not only about mining, search engines and spidering. Application developers and content providers, and end-users, can benefit from using the Semantic Web (RDF) within their own services, regardless of whether they expose that RDF metadata to outside parties. RDF is useful without doing any data-mining — it can be baked right into content within authoring tools and created transparently when information is published. RDF makes content more manageable and frees developers and content providers from having to look at relational data models. It also gives end-users better ways to collect and manage content they find.

The Semantic Web is not just research. It's already in use and starting to reach the market. The government uses it of course. But also so do companies like Adobe, and more recently Yahoo (Yahoo Food is built on the Semantic Web). And one flavor of RSS is defined with RDF. Oracle has released native RDF support in their products. The list goes on…

Ross Mayfield doesn't put much stock in the grand semantic Web vision or a Web 3.0 in the future.

I'd bet the future is less the Matrix than Soylent Green.  Less semantic fuzz than social discovery.  Less artificial intelligence than human intelligence.  Less automation and more augmentation.  Wandering around the Web 2.0 Summit I saw more presentations using 3.0 than I can enunummerate.  Some were about more immersive platforms, some desire the singularity, but most just wanted to be new and cool.

The truth lies somewhere in the middle–the living and breathing social Web along with semantic Web technologies will help create a new infrastructure for the information Web that is far different and more powerful than what we call Web 2.0…

November 10th, 2006

Web 2.0 Summit, Google speed, offline synch, quad-core and more...

Posted by Dan Farber @ 12:49 pm

Categories: Between the Lines podcast, Entertainment, General, Hardware Infrastructure, IT Management, Office 2.0, Personal Technology, Software Infrastructure, Web Technology, Wired & Wireless, web2con

Tags:

This week on the Dan & David Show, I give my rundown on the Web 2.0 Summit, including presentations by Jeff Bezos, Ray Ozzie, Marissa Mayer and others. David gives his take on Wi-Fi in airports and Microsoft's Zune, and we discuss the impact of quad-cores on datacenters, Zimbra's offline AJAX client, Amazon's revolutionary EC2 and Riya's Like.com visual similarity shopping service.

This podcast can be delivered directly to your desktop or MP3 player if you're subscribed to our podcasts (See ZDNet's podcasts: How to tune in). For more the topics covered during the show, search our blog.

November 9th, 2006

Yahoo planning a 180 on Yahoo 360

Posted by Dan Farber @ 5:36 pm

Categories: General, Personal Technology, Web Technology, web2con

Tags:

Yahoo co-founder David Filo, who rarely gets in front of a conference crowd, chatted with John Battelle at the Web 2.0 Summit, aided by Bradley Horowitz, Yahoo vice president of product strategy. Filo is the product focused half of the founding duo. The other founder, Jerry Yang, focuses on the business issues.

You would think after 12 years in the saddle and with billions of dollars in their accounts, the motivation to keep at it would diminish for the founders, but Filo said that the opportunity that lies ahead keeps him motivated, as well as being part of the "revolution."  

Battelle asked about Microsoft buying Yahoo. "Over 12 years of our history rumors come and go," he said.  On further probing by Battelle, Filo added, "It's pretty safe to say nothing will be announced tomorrow."

Dave McClure, from the audience, asked why Yahoo hasn't acquired Six Apart's blogging tools or Facebook to gain some faster traction in blogging and social networking. Horowitz responded, "We put Yahoo 360 out there and learned a lot from it. 360 may be doing a 180, and change and adapt to address the opportunities." He didn't offer any specifics as to what the 180 degree turn looks like.

Filo added, "We are absolutely interested in blogging, and we hope five years from now to be a major player in that space. We will look at acquisitions."

 Yahoo co-founder David Filo and strategy VP Bradley Horowitz

Horowitz dealt with the Google question, that Yahoo is slowing down as Google ascends.  He admitted that Yahoo has not done a good job on the monetization side. On the other hand, products are growing (Flickr has grown 15x) and are being knit together. 

Filo commented on Google, saying Yahoo has never had a shortage of competitors in its history. "The next MySpace or YouTube is just as much a threat as Google," Filo said. "It's more important to think about how we evolve as a company. One constant is we know things are going to change and the leader in search five years from now might be google us or some startup."

Battelle asked about how Google or Microsoft can outspend Yahoo for acquisitions and monetization deals. "Microsoft could outspend us on anything. It's not going to define our success or failure. In the environment we operate in, we have to build better products and services. We have looked at thousands of companies over years and generally we get the companies we want," Filo said.

On the new ad monetization engine, Horowitz said that there are a few hundred advertisers are on the system. "We are taking it slowly coming up on the Christmas season and want to make sure we get it right. We are cautious but plan to get advertisers converted over in Q4 and Q1 [2007]."

November 9th, 2006

Google's Marissa Mayer: Speed wins

Posted by Dan Farber @ 3:01 pm

Categories: General, Hardware Infrastructure, Personal Technology, Software Infrastructure, Web Technology, Wired & Wireless, web2con

Tags:

Marissa Mayer of Google gave a testimonial to speed. Her key insight for the crowd at the Web 2.0 Summit is that "slow and steady doesn't win the race." Speed is a huge component and big market driver of Web 2.0, she said.

In testing out the user interface for Google search, Mayer said that with more results for a query, users were spending less time on the site. It turned out that the cause wasn't just the paradox of choice–paralyzed by too many choices–but the fact that a page with 10 results was half a second faster that the page with 30 results. So, Google set about making the page with more results faster, and the rest is history. 

Part of Google's secret sauce is that a round trip for a search query that returns a result in .05 seconds touches 300 to 700 Google machines across the country, Mayer said. "Users really respond to speed," she said.

She gave an example of how two of Google's AJAX applications, Gmail and Google Maps, benefited from speed tweaks. The choice to use Javascript for the Gmail interface was key, despite the reliability flaws of the client-side scripting language, resulting in instantaneous action when performing basic email tasks, such as moving or deleting messages.

When the Google Maps home page was put on a diet, shrunk from 100K to about 70K to 80K, traffic was up 10 percent the first week and in the following three weeks, 25 percent more, she said.

"You might say that it's obvious, but it is something bigger," Mayer said. Instant feedback enables a faster learning curve. She mentioned digital photography and Wikipedia as examples of how instant gratification helps drive people to become expert users faster. "If you have each transaction take less time, you have expert users more satisfied. You want lots of small and fast interactions if speed is important," she added.

What speed ultimately means, she said, is that the network will take care of it, but changes in the browser support will be important, such as richer support for client-side applications in Internet Explorer and Firefox. Technologies like Google Web Accelerator to help speed up the Web will also play a role. Mobile is also an area that needs to be sped up. 

For Google, or any other Web site, speed means people viewing more pages in same visit and spending more time overall online…and more money.

November 9th, 2006

Zimbra demos offline client

Posted by Dan Farber @ 11:37 am

Categories: General, Software Infrastructure, Web Technology, web2con

Tags:

Zimbra showed an offline AJAX client, which has the same look and feel as the online client. Mail, calendar, contacts, and documents are cached and then synched with the server when the user gets back online. "Technically the Zimbra Offline client is the same AJAX client UI, but now you are connecting to a local sync'd cache of the data and more importantly the ability to search, tag, organize, etc without network access," said Zimbra CEO Satish Dharmaraj. In addition, Zimbra is selling its collaboration suite to hosting providers and white labelers, and introduced new skinning capability. "You can make it look like anything," Dharmaraj said as he showed Zimbra looking like Gmail. The offline synchronization capability will be available before the end of the year. In addition, Zimbra is planning to split out the mail, calendaring and wiki components of the suite so they can be sold as separate products, Dharmaraj said during a panel at the Web 2.0 Summit.

November 9th, 2006

Try Microsoft's Photosynth

Posted by Dan Farber @ 10:47 am

Categories: General, Personal Technology, Web Technology, web2con

Tags:

After seven months of development, Microsoft is making Photosynth tech demo available. You can't upload photos but it's a cool demo of how to navigate a photo site. In keep with Microsoft's merging of the Web and client, Live Labs head Gary Flake called Photsynth an "elegant marriage between the client and server." Photosynth allows:

• Walk or fly through a scene to see photos from any angle.
• Seamlessly zoom in or out of a photo whether it's megapixels or gigapixels in size.See where pictures were taken in relation to one another.
• Find similar photos to the one you're currently viewing.
• Send a collection or a view of one to a friend.

November 9th, 2006

Yahoo embeds IM in mail

Posted by Dan Farber @ 10:23 am

Categories: General, Personal Technology, Web Technology, web2con

Tags:

Ethan Diamond, director of product management for Yahoo Mail, demonstrated some cool new features that integrate instant messaging chat into mail client. Users can see if contacts are logged on to Yahoo Mail and initiate a chat, and then have the contents of the chat automatically inserted in an email, or email contents inserted a chat with the Mail application. It can also show everybody in an address book who on line and and available to chat. Yahoo has about half a billion people who use its communications products, Diamond said. 

The embedded chat will be available in a few months to Yahoo's 250 million mail users, said Brad Garlinghouse, vice president of communications, communities and front doors at Yahoo.

Prior to the demo, Garlinghouse went on about consumers acting on emotions, the Machine Internet versus the Social Internet, and how Yahoo has a strong role to play in making new kinds of user experiences that will appeal to mass audiences. He suggested that Web 2.0 is moving from being a geek-driven experience to a consumer-driven experience. "Technically, tactically and emotionally users yearn for something better," Garlinghouse said. "We demonstrated tight integration and breaking down the silos of email and instant messaging so a mass audience and making it more social change."

I'll buy that–the presence awareness and flipping back and forth between email and chat/IM improves the user experience, makes it more accessible.

Then, Garlinghouse went overboard, beyond the yearning of users for something better, proclaiming, "We are changing the way the Earth communicates." I guess if 250 million users can chat within Yahoo Mail, it is a change, but not exactly on a planetary or profoundly transformative scale.

  Yahoo's Ethan Diamond and Brad Garlinghouse

November 9th, 2006

The great [structured] database in the sky

Posted by Dan Farber @ 9:57 am

Categories: General, Open Source, Software Infrastructure, Web Technology, web2con

Tags:

MySQL ECO Marten Mickos opened the morning session of the Web 2.0 Summit with his dream of the great database in the sky. He envisioned a database in which all structured data could be shared, open sourced so that people could instantly know what is happening in the world. For example, if all the weather data were made available, users could tap into pool of information and apply it real estate, travel, sailing, surfing, events and many other activities.  

Mickos described the database in the sky as the other side of the coin for what Google does for unstructured data, although it must part of Google’s vision to tackle the dark Web of structured data. Mickos wouldn't mind if MySQL become the underlying database storing the structured data or index for the great database in the sky.

 MySQL CEO Marten Mickos look toward the great database in the sky

"Google gives access to all data in entire world. We could do the same for structured data,” Mickos said. "We could open up all the world’s structured data to all the world’s developers, entrepreneurs and mashup artists. It’s the great database in the sky.”

The biggest database is out there, it’s not just connected, he added. “We need to build a Skype for database connections, which allow sharing, connecting and aggregation of data in real time, so you instantly know what is happening in the world. Today it’s too complex, with different formats, ways to present data and different ways to get there.”

The technology pieces to make it happen exist, but haven’t been fully applied to the problem. Technologies include RSS, ATOM, Jabber, HTML, HTTP, XML, SQL and SMS, Mickos said, and they have to be scalable and work in a peer to peer model without lots of complicated constructs.

He said that a DNS server that knows all the SQL databases in the world would be required. Some of the problems to overcome include routing, making data understandable and accessible to other; and payment systems.

The fact that latency is going down and bandwidth is increasing gets us closer to synchronous flow of data, Mickos said. “It’s less of an issue if one of databases is a bit latent—the value of combining the data is much bigger,” he said.

To make the vision work, data owners need to share, Mickos said. “The simple value of sharing is enough of an incentive.” In addition, a brokerage keeping track of who has what data to offer the world in what format is also necessary.

If the great structured database in the sky materializes, then the data would be the platform, Mickos concluded. It's a great dream, and it will come to be in fits and starts as data sources are opened up and technological frictions are removed.

November 9th, 2006

IBM's Jeff Jonas: Reducing corporate amnesia

Posted by Dan Farber @ 8:00 am

Categories: General, IT Management, Security, Software Infrastructure, web2con

Tags:

Jeff Jonas lives in Las Vegas and used to make his living helping casinos reduce fraud and three letter agencies find the bad guys with unique technology as founder and chief scientist of Systems Research & Development. His software can sift through information captured in multiple database and correlate data, such as identifying individuals listed under different names in separate databases. In January 2005 he sold his small company to IBM and is now he is responsible for shaping the company's overall technical strategy of next generation identity analytics and the use of the new capabilities in the overall IBM technology strategy.

Jonas also spends about 40 percent of his time on privacy and civil liberties issues. For example, he is developing techniques for enabling advanced data correlation without transferring any content that violates privacy.

Jonas presented some of his ideas at the Web 2.0 Summit, and I caught up with him after his talk for an interview. In the video he discusses enterprise awareness, corporate amnesia and how he believes the problem of correlating data from disparate databases can be solved. The solution, he said, is to figure out when two things are the same despite being described differently in real time on streaming data. "You want the data to find the data and the relevance to find the user," he said. 

November 8th, 2006

Wikinomics and mass collaboration

Posted by Dan Farber @ 6:30 pm

Categories: General, IT Management, Personal Technology, Software Infrastructure, Web Technology, web2con

Tags:

During a presentation at the Web 2.0 Summit, author and consultant Don Tapscott previewed some the highlights from his forthcoming book, Wikinomics: How Mass Collaboration Changes Everything.

He views the next phase of the Web as about mass collaboration and harnessing the power of self organization. It will have a profound change in way companies innovate, orchestrate to create value and compete in marketplace, he said. With billions of smart devices on the Net and billions of people on it, for those users it will be like programming a giant computer. “Social networking is becoming the new mode of production,� Tapscott said. He gave an example of Procter & Gamble, which has developed an productive ecosystem of external collaborators that he said creates value more effectively than a hierarchically organized business.

I caught up with Tapscott after his presentation and asked how he thinks the move to mass collaboration will impact the enterprises, which will be slow to adapt to new work paradigms. I also posted on his research on the effect of the Net generation in the workplace.

November 8th, 2006

The new age of platforms and ecosystems

Posted by Dan Farber @ 6:23 pm

Categories: General, Hardware Infrastructure, IT Management, Software Infrastructure, Web Technology, web2con

Tags:

I caught up with Salesforce.com Marc Benioff at the Web 2.0 Summit, where he gave a presentation of enterprise mashups. In the video I asked Benioff about the shift from focusing on individual applications to platforms with diverse ecosystems and utility-like infrastructure services, such as his company’s Apex platform and AppExchange marketplace and Amazon’s Elastic Compute Cloud (EC2). Benioff calls it moving from killer apps to killer platforms, of which there will be many.

November 8th, 2006

Ray Ozzie's very interesting juncture

Posted by Dan Farber @ 6:00 pm

Categories: General, IT Management, Office 2.0, Personal Technology, Security, Software Infrastructure, Web Technology, web2con

Tags:

"We are at a very interesting juncture," said Ray Ozzie speaking with John Battelle at the Web 2.0 Summit. He was partly referring to Office and Vista being released to manufacturing. But is also a time in which the Web is becoming a better application platform, Microsoft is transitioning into subscription services, offering a music player and service, and Ozzie in growing into his role replacing Bill Gates as the chief software architect at Microsoft, charting his company's future in a new era.

On the subject of Vista, Ozzie stated the obvious, "It's not perfect. It's software and software has flaws."

"The biggest single thing that Vista can do is to provide a safe environment that you do interaction on the Web and load code in a way that doesn’t leave it as open as in the past. Vista is secure by design. We have done lots of static analysis, given the surface area of product, to remove vulnerabilities before it ships." He noted that there will be some missing drivers, but said he was feeling really good based on the internal deployment.

On the subject of delivering applications, like Office, in the cloud, Ozzie said, "We are looking at what kind of experience in this era is the best way to deliver value, and we are examining what is the best delivery model is and then to reshape and re-architect the products based on that." It's consistent with what he said when I interviewed him a few weeks ago.

"From my perspective and looking at what Microsoft has in the product space, we have about half a billion users who use Office actively. This addressable market–some are paid and others are prospects we could upsell to. I don't have to buy companies to acquire a new audience and they understand the value proposition. We just have to deliver value to them that is relevant in this era."

The "in this era" refers to different usage models that range from the rich client to the cloud and servers and services. Whether application have a client component or are pure Web depends on the scenario usage, Ozzie said. Porting PC functionality to the Web isn't a obvious solution. Online brings universal access, sharing scenario and quickly going in and out, and the PC is really good at providing a flexible, fast UI regardless of connection speed and it is reliable, Ozzie said. "We are going from world in 80s where the PC was for document editing to embedding different document types," Ozzie continued. Low-cost terabyte hard disks, the power needed to edit and constrained upload bandwidth favor the PC in a video editing scenario. 

Related to the Windows operating system, Ozzie said the we are on the cusp of moving from multi-core to many-core processor environments. "The system needs to help application programmers consume that in a reasonable way without teaching everyone on earth how to factor their code that way,"  he said. Power management is also another area to innovate, as well as state separation for applications that would bring deployment models up to date. "Everything should deploy from the Web–you shouldn't need a DVD to install software. Search is also an area of focus, helping satisfy a user's intent more than has been done thus far, Ozzie said. 

He was also asked about the big shoes he is filling. "There is a certainly mythology around any leader, particularly Bill, he very broad and deep. The way Microsoft works is it is very, very rare that Blll actually gave an order and said go do this. Bill will always have an amazing level of soft power, and people want to follow. The organization reveres him. I was given a free pass. I have known Steve and Bill a long time and came in with benefit of the doubt. I have to earn the followership. I am glad we have a two-year transition so people can see us interacting together. My interaction style is significantly different.

He was asked about any culture changes as he becomes the head software man. "You have to ask people progressively over time who work there. It has to happen locally, you can't mandate it. The Office and Windows group or Xbox group have different ways of working. Thing have changed in certain ways, but it's not one person who makes this happen, it's a conversation.

November 8th, 2006

Ask and Microsoft talk about challenging Google

Posted by Dan Farber @ 4:19 pm

Categories: General, Web Technology, web2con

Tags:

Ask CEO Jim Lanzone and Microsoft search head Steve Berkowitz were asked about competing with Google during a session at the Web 2.0 Summit. Neither came up with a convincing plan for how they might knock Google off its game.

Ask has about a 6 percent share and Microsoft is around 12 percent of the search market (although Microsoft has 460 million users worldwide), which is dominated by Google, with Yahoo following up. At this point there are legitimate competitors, Lanzone said. He identified Google’s soft underbelly as being distracted from search by other development projects, such as deploying Web services to enlarge its footprint. “It requires Google to innovate in areas that don’t have much to do with search,” he said. "Google is the model T of search. Over time peoples' needs evolve."

I don't think that Google will be asleep at the wheel and fail to evolve search with more customizations than a Model T. 

 
  Steve Berkowitz and Jim Lanzone

Berkowitz is focused on creating different search experiences for Microsoft search. "The idea is that at end of day you create a great experience for consumer….marketing in this business is the product itself. I don’t think we have understood it, and if we embrace that we can change the dynamics of market place," he said. For Microsoft, the ideas seems to be creating different search experiences based on where users come from. If users enter search from a community site, there is more context, such as pages visited, which will be different from a searcher coming from instant messaging, where the user could be assumed to younger and might want to tap into results from friends in their network.

Regarding Google, Berkowitz said that Google could have the problem in the future of maintaining backward compatibility as it expands its footprint, as will as cultural issues and the problems associated with being a public company, which Microsoft knows well.  

None of the criticisms add up to much of challenge if Google continues to execute, but it isn't a zero sum game, as Lanzone concluded. The portal players will try to come up with reason that make it less of a habit for people to jump to Google for search.

November 8th, 2006

Searching for human-like search

Posted by Dan Farber @ 10:56 am

Categories: General, Personal Technology, Software Infrastructure, Web Technology, web2con

Tags:

Search is the motherload of the Internet today, as has been proven by Google. But it is still clunky and you end up with 149,000,000 results for "ajax" delivered in .05 seconds.

At the Web 2.0 Summit, hakia showed a beta of its "meaning-based" search engine. According to the company, "the basic promise is to bring search results by meaning match - similar to the human brain's cognitive skills - rather than by the mere occurrence (or popularity) of search terms." 

hakia's Query Detection and Extraction (QDEX) uses semantic analysis, dissecting pages into knowledge bits and storing them as gateways to all possible queries. According to the company, QDEX has superior scalability properties because data segments are independent of each other, and hakia's SemanticRank algorithm includes bits of ontological semantics, fuzzy logic, computational linguistics, and mathematics. Whatever 

Powerset CEO and co-founder Barney Pell told that his search engine would make an appearance around May 2007. In his post on natural language search, Pell spells out some of the ideas behind his development effort:

Instead of keywordese or even advanced keywordese (which few people can remember how to use), true natural language queries have linguistic structure. This includes queries where the function words matter, where word order means something, and where relationships that should be explicitly stated easily are stated. Instead of ignoring the function words, a natural language search engine respects their meaning and uses it to give better results. Instead of being a waste of time for a user to add stopwords in a query, each little word added has a profound effect on the search quality. 

Pell also expects that there will be several more entrants entering the natural language search space. Jack Ma, founder and CEO of Alibaba in China, is also applying his resources to making search more human-like, athough he didn't offer my any details when I spoke with him.

Pell's point of view, which is shared by the entire industry attacking the search problem, is that  there is enormous room for fundamental innovation and that the game has only just begun. 

For visual search, Riya's Like.com engine brings a new dimension to finding objects via similarity.

Later today at the Web 2.0 Summit Jim Lanzone of Ask and Steve Berkowitz of Microsoft are going to talk about beating Google at its own game. Stay tuned…

November 8th, 2006

More from the Web 2.0 Summit

Posted by Dan Farber @ 8:43 am

Categories: General, IT Management, Personal Technology, Software Infrastructure, Web Technology, Wired & Wireless, web2con

Tags:

Yesterday's Web 2.0 Summit included a few highlights worth mentioning. Elinor Mills covered Skype co-founder Niklas Zennstrom's chat with John Battelle, Skype plans to offer bloggers and others the ability to hold audio chats in the next version of it service.

The next version of Skype will enable people to post a link on a blog or Web site that will take people to a public chat room when clicked on, he said during a question-and-answer session during dinner.The live chats would be "Skypecasts," which Zennstrom described as public conversations or audio conferences that people can moderate. He would not provide a timeline for the features except to say it would be "soon." Meanwhile, Skype has had conversations with many social-networking sites about offering services that would allow users to "share content with each other in a conversation," he said.

Zennstrom also briefly touched on two Skype-backed projects, FON and the The Venice Project. "FON could become the largest public Wi-Fi network," he said, and the Venice Project will take the best of TV and puts it on the Net, not just the short, low-quality video clips. He also said that Skype revenues will be close to $200 million for this year, up from $60 million the previous year.

Martin Lamonica posted about John Battelle's conversation with Barry Diller of IAC and Arthur Sulzberger, chairman of the New York Times Co. On the subject of user-generated or amateur content, Sulzberger said that the NYT plans to embrace content produced by the masses, but with caution. "We are looking at information gathering by amateurs, if you will, that we trust. Because at the end of the day, we're putting our name on that work, and finding the right balance is hard work."

"It doesn’t matter how you present the journalism–paper or electronic. It is the quality of editing and reporting that is important," Sulzberger said (pictured below).

Diller came up with his usual memorable comments. He agreed with Sulzberger that "editorship" is not going aways and that "everyone would like to believe that their entrails are of great interest to everybody, but it's just probably not so."

Martin also covered a panel that discussed the hurdles standing in the way social networking applications. For businesses, many of the social networking tools are not industrial grade, and for consumers Web users are stuck in silos.

Despite ongoing attempts to establish single sign-on standards, Web surfers typically are not able to log on to several sites at once, such as booking an airline ticket from United Airlines and then renting a car from Hertz.

Marc Canter, the CEO of Broadband Mechanics, wants end users to have the ability to control their profiles.

End users are asking for the ability to have a master profile so that they can keep data in one place and can update it. They want to move their data and feel like they are in charge of their data.

November 8th, 2006

Riya launches Like.com visual similarity shopping

Posted by Dan Farber @ 12:45 am

Categories: General, Personal Technology, Web Technology, web2con

Tags:

Riya’s facial search service was one of the first darlings of the Web 2.0 world. The company garnered more than $15 million in venture capital to make facial and object search as easy as text search. It turned out that searching by facial similarity was a nice bit of computer science but not what turned users on.

Riya loaded 50 million faces from MySpace into its search engine and tested out the service with 100 users, mostly from MySpace. According to Riya CEO Munjal Shah, the testers didn’t find the facial search very useful, but visual similarity shopping for items such as shoes and jewelry was a hit. Today, Riya is launching an alpha version of Like.com, a new visual search shopping engine based on its visual recognition technology. Click on a photo of an item and Like.com compares the shape, color, and texture to other items in the database and displays the most similar results from merchants such as Amazon, eLUXURY, ICE.com, Lands’ End, ShoeBuy, and Zappos. 

"We see a gap. There is little innovation from Google, Yahoo and other in image search,” Shah said. "Thirty billion items are sold on line with no great cross site search tool." As a result of the opportunity, Riya is not putting a lot of effort into its face recognition service, which has over 10 million uploads since it launched in March, Shah said. 

Like.com is starting out by indexing images from jewelry, handbag, shoe and watch merchants, and will add clothing soon, Shah said. Jewelry, handbags and clothing are about a $15 billion business online. The target audience is women from 20 to 30. Furniture and home and garden (such as search for flowers) are being considered as additional categories.

Like.com offers several search capabilities, including the ability to search by image instead of text; finds items that have specific features, such as a watch bezel; find color variants of the item via a color picker; find clothing, shoes and accessories similar to those worn by your celebrities (Like.com includes 100,000 celebrity images); and in the near future the ability to upload photos. Like.com will also have a browser extension to initiate likeness searches from any site as well as pages to save searches and a recommendation engine. After launch Like.com will also have a cross-matching feature. "If you have a hat and want shirt to with it, you drag a slider and search on new category," Shah said.

For example, users searching for a particular watch style can draw a box over dials on a watch face. If it has three dials, Like will return results with three dials. If the color of the band is chosen as a search criteria, Like will find all watches with the same color band or in slightly different shade. Users can interactively filter the results to find the the product that they desire. A slider can filter  by shape, color and pattern for any of the products.

The core technology is even more complex than face recognition technology, Shah said. Like.com crawls target merchant sites and retrieves the highest quality images. It takes about 20 seconds per image to preprocess, creating a visual signature and indexing the image.

Search results are returned in under a second–the server farm consists of 250 quad-core servers, each loaded with 16 to 32 gigabytes of memory. Like.com converts every picture into a visual signature, a 10-kilobyte vector image consisting of about 5,000 numbers. The “likeness” algorithm determines the order of results based on shape, color and pattern.

“We are extracting and computing the visual signatures and pulling out pieces for comparison,” Shah said. “The results will never be worse than a text search. We index all the metadata and even normalized some of it.”  Currently, Like.com only indexes the merchant sites.The soft goods vector images are more detailed than faces, which are encoded as 3-kilobyte vectors, and include about 40 elements, including shape densities, color histograms broken into quadrants and other properties, such as glossiness and sheen (analyzing color changes in the middle of objects).

 Riya and Like.com CEO Munjal Shah 

The technology performs best when items are 2D or symmetrical, Shah explained. “Jewelry is the hardest because is has lot of artifacts. The way light bounces across the images sometimes get confused with holes in items, like rings. Other elements work on 3D items if they are symmetrical like a vase. It works with clothing like shirts but not well with skirts and bulky items—we can get the color and pattern but not shapes as easily, but the metadata is helpful with results.”  

Like.com is open for business with an alpha version, and is adding about 30,000 items a day to the database. The first 10,000 Like.com shoppers will receive free shipping, up to $8, on their first purchase from any of the featured retailers.  

Like.com gets paid via cost per click and cost per action. The company is  looking to advertising as a revenue source, but is focused on building the audience first. "We are also having conversations with celebrity bloggers with images. We could create a new ad unit that goes into the picture. If someone clicks and buys, we share the revenue. However, our first goal get the search engine right and then add ad units for each category," Shah said.

Shah acquired the Like.com URL for about $100,000, which would be a bargain if the site is successful. From my test drive, just after the site launched. Riya seems to have figured out how to extract value from its visual search technology.

See also: TechCrunch and Robert Scoble for additional coverage of Like.com

November 7th, 2006

Google CEO Eric Schmidt: We would never trap user data

Posted by Dan Farber @ 5:23 pm

Categories: General, Software Infrastructure, Web Technology, web2con

Tags:

John Battelle and Tim O’Reilly kicked off the industry heavyweight portion of the Web 2.0 Summit with a few remarks about Web 2.0 and the theme of the event, disruption and opportunity.  

“Web 2.0 is about harnessing the network effect, which gets better the more people use them,” O’Reilly said. “It could also be called user self-service. This is the start of the real disruption, with asymmetrical competition from user self service.” As an example, Craigslist can be a big player with a few dozen employees.  

“We are starting to see how this is becoming a platform game. A platform beats and application every time….Even though we have the wonderful bubbling up of user generated content, it is the beginning of Web 2.0, not the end. Web 2.0 of next year will be very different than this year.

Hopefully it won’t be called Web 3.0. We need at least another five years to play out Web 2.0.

Google CEO Eric Schmidt was the first interview of the afternoon. Battelle started by asking why did Google bought YouTube. "YouTube was growing faster [than Google Video] and video became fundamental data type on the Internet, so that’s why we bought them.," Schmidt said. He expects to maintain YouTube as a separate property, and pointed to its ideas around community as a reason. "The underlying draw is to see what users are doing and have computers suggest related or adjacent content. It is a whole new paradigm and important to users."

Schmidt was asked about copyright issues related to YouTube and paying off media companies. "We have visited with as many media companies as we can," Schmidt said. "We have to respect copy right." He went on to say that the rights clearing process and monetiziation are complex issues, and Google is working on models as to how to monetize copyrighted content.

Battelle suggested that Google has unfair advantage in pricing deals, such as with MySpace and YouTube. "We build a very good targeting engine and a lot of business success has come from that. We run the company around the users–so as long as we are respecting the rights of end users and make sure we don’t do anything against their interest, we are fine," Schmidt said. He noted that history has shown that the downfall of companies can be doing things for their own self interest. "We would never trap user data," he said.

Schmidt was asked if users could get all of their search history and export it to Yahoo. "We would like to do that, as long as it is authenticated….If users can switch it keeps us honest."

Schmidt described the transition in platforms as a major shift. Software running on a datacenter that is running 24×7 is more reliable, he said, and the architecture transition to LAMP and REST will enable an era of huge massive server farms that are fundamentally more reliable than what they replace. "It’s a big change and means users can use other devices, such as mobile phones. It's a real change in the paradigm." He allowed that there will be many winners. 

On the subject of office suites, Schmidt played the semantic game, claiming that Google is developing applications for just "casual" use. "We don’t call it an office suite. It's not targeted at the [Microsoft] Office–we never made that claim. For many people, it would be just as easy to have computers in the cloud store information you use every day." Data in the cloud should be more searchable and enable more rapid sharing, he added. "We embarked on a strategy to build apps that are search centric and very sharable….as something use in normal life. We are not arguing it is an [Microsoft] Office replacement, but a different way of manage information. I don’t think it replaces Office. The focus we have is casual sharing and usage, and it has one benefit–it is free. 

I don't know how Schmidt defines casual, but it could be products that lack the industrial strength of Microsoft Office, but are evolving in that direction…quickly.

Schmidt said that one of projects Google is engage in is figuring out the underlying architecture for building Web applications. Perhaps that is why Google bought JotSpot. 

One of the limits of the current Web application architecture is the lack of offline support. "It's possible that the new network architecture will always be presumed to be around and a disk is optional. It is going to happen, at least in the Western world. It looks like ubiquitous connectivity is near. For the near term, cache proxy on a PC is the obvious technical solution, Schmidt said.

November 7th, 2006

Inside the mind of the Net generation

Posted by Dan Farber @ 12:37 pm

Categories: General, IT Management, Web Technology, web2con

Tags:

During a session at the Web 2.0 Summit, author and consultant Don Tapscott shared results from a research project on the Net generation, the first humans to grow up digitally. An estimated 80 million people in the U.S. alone are coming into the workplace and marketplace with a far different set of experiences and skill than previous generations. “Kids are lapping their parents on the info track,” Tapscott said.

In the survey of 2,000 people between the ages and 12 and 29, the results showed that freedom of choice was a critical factor. Unlike older generations, the Net generation has many options. In addition, freedom to customize–cars (Pimp My Ride), computers or anything that makes it fit with who they are—is also important. 

The Net Generation also wants the freedom to schedule. In the survey, 42 percent of those survey watch TV asynchronously, time-shifted viewing. Television took 24 hours of the week per baby boomer, Tapscott said, but the Net generation watches less TV and the whole way they process information and think is different. “This is not just a life stage difference, this is a generational difference because cognition and information process different because of way they have grown up.”

The Net generation is also the new “scrutinizer,” and the majority do research prior to purchasing, 42 percent of the respondents at the top of the adoption pyramid write reviews. This means that those who are selling to the scrutinzers can’t slide by with substandard products or services.The search for integrity is another aspect of the Net generation. “They have very deep values and believe very strongly,” Tapscott said. “If you are Sony and give out 50 cameras to actors who pretend they are tourists and take pictures of kids, that  is an integrity violation and you are going to get nailed.”

Relationships and collaboration are part of the Net generation psyche. In the survey 65 percent of respondents preferred a two-way relationship with the brands they select, and the majority are willing to share information. Hence, the rise of social networks, or ‘n-fluence’ networks, in which the Net generation influences each other and other generations.

Experiences, entertainment, fun and playfulness are all an expected part of the relationship with products. Speed and immediacy are another trait of the Net generation. They have grown up multitasking and “The general trend is because they have grown up multitasking

The brand becomes a more complicated construct, and requires messaging that resonates with the most influential consumer generation in history. “This is a generation that will set a higher bar.”

As employees, the 80 million young adults coming into the workplace, the Net generation has more experience in the digital world and personal connections to the job, and will want to be part of an engage and collaborate model rather than command and control. “They want to learn, have fun and do something significant,” Tapscott said. "We have to creating working/learning environments, with learning built into the work system.”

He concluded by saying that Net generation coming into the workforce and marketplace will require a paradigm shift that “involves dislocation, conflict, confusion and uncertainty. New paradigms are nearly always received with coolness, even mockery and hostility. Those with vested interests will fight change, and the shift demands such a different view to things that established leaders are often the last to be won over.”

November 7th, 2006

The invasion of Enterprise 2.0 software

Posted by Dan Farber @ 10:49 am

Categories: General, IT Management, Software Infrastructure, Web Technology, web2con

Tags:

Blogger and stock analyst Paul Kedrosky introduced the topic of Enterprise 2.0 at a morning workshop at the Web 2.0 Summit by trying not to define the concept, calling the definitional discussion a masturbatory exercise. Nonetheless, he offered Harvard Business School Associate Professor Andrew McAfee's definition as a starting point:

Enterprise 2.0 is the use of emergent social software platforms within companies, or between companies and their partners or customers.

Note: Dion Hinchcliffe has a post on Web 2.0 transitioning to Enterprise 2.0, citing a new report from O'Reilly, Web 2.0 Principles and Best Practices. Dion writes about McAfee's explanation of Enterprise 2.0:

SLATES describes the combined use of effective enterprise search and discovery, using links to connect information together into a meaningful information ecosystem using the model of the Web, providing low-barrier social tools for public authorship of enterprise content, tags to let users created emergent organizational structure, extensions to spontaneously provide intelligent content suggestions similar to Amazon's recommendation system, and signals to let users know when enterprise information they care about has been published or updated, such as when a corporate RSS feed of interest changes.

While SLATES forms the basic framework of Enterprise 2.0, it does not negate all of the higher level Web 2.0 design patterns and business models.  And in this way, the new Web 2.0 report from O'Reilly is quite effective and diligent in interweaving the story of Web 2.0 with the specific aspects of Enterprise 2.0.

Kedrosky went on to say that collaborative software, with the exception of email, has not worked particularly well. "Why will it work when they didn't before. I'm not sitting around thinking it's a good time to use emergent social software," Kedrosky said. "What is it going to replace? In the consumer market, social technology replaced other ways of interacting, but in the workplace what do they substitute for?"

Avi Bryant from Dabble DB said that his product tries to do a better job of subverting IT, trying to give users tools that let them do a better job of building a proper data model or in-depth reporting themselves. Also the traditional top down enterprise sales model is subverted, targeting the users rather than IT or the CIO, and charging a small monthly fee on a credit card.

Socialtext CEO Ross Mayfield identified the trend as "enterprise target with consumer approach."

Jeff Nolan of Teqlo, and formerly SAP, noted that Enterprise 2.0 could be applied to any business context, order to cash, human resources, merchandizing. "There are 1500 processes that can benefit from the technology," Nolan said. "The fundamental problem with traditional software is that developers are disconnected from users. Salesforce.com was successful by selling to end users and the company can see at any given moment what is being used in their software."

Bottom up, software-as-a-service, do-it-yourself, monthly subscription, collaborative, modifications of applications by mere mortals are are part of a movement that is seeping into corporations, just as PCs did in the early 1980s. For the more maverick or adventuresome users, if IT doesn't provide the tools they will bring them in, and charge them to their credit card. "You can't stop any hosted services," Nolan said. 

Speaking as a hosted services provider, Nolan hopes that is true. But the reality is that corporations try to lock out software that is rogue or fails to pass the compliance test. They can't be totally successful is keeping unwanted software out, but the IT executives aren't going to turn their back when they need to cover their ass. Peter Rip of Crosslink Capital chimed int that companies will have apps in the could and in the sandbox that can be firewalled by IT.

Mayfield's experience is that the champions who bring in Web 2.0, Enterprise 2.0 or whatever you call it into an enterprise from the bottom up often become the ones who work with IT to legitimize the software used by popular demand. In addition, as new class of bottom up, user centered software matures, it will add the kinds of features that make it fit into corporate environments. Intel put together SuiteTwo, a collection of Web 2.0 content development and distribution products that could make the software more palatable for IT departments. 

There will be a tug of war, but in the end, users will prevail. Bottom line, users unhappy with their tools won't be the most productive workers…