sorted by: recent | see : popular

Content Tagged with startups + goog

It has been a long time coming, but Powerset, a San Francisco-based contextual-semantic search engine has finally launched. I urge you to try it out, for this is quite an impressive search effort, despite the fact it is currently limited to searching Wikipedia along with some supplementary results from Metaweb’s Freebase. I think it has made Wikipedia much easier to use. I like how you can do more topic-based searches and get a holistic view of the information you’re looking for. Danny Sullivan over on Search Engine Land has an elaborate and fantastic indepth review of Powerset, and that frankly obviates the need for any other review.

That said, Powerset faces an uphill climb, especially when it comes to consumer mindshare. I think Google has become so synonymous with search that it is virtually impossible for a newcomer to establish a toehold. Powerset’s approach is different, and its tactic of applying its technology to specific content repositories such as Wikipedia is smart. But will they (web searchers) come and use Powerset?

At our recent GigaOM PM event, Chad Walters, director of engineering, search and platform at Powerset, gave a talk about how his company was using Hadoop and other clever technologies to meet its immense infrastructure needs. Here are some bits from OStatic’s live blog coverage of the event:

Powerset applies deep natural language processing (based on technology licensed from Xerox PARC), which means the company needs 100 times more processing horsepower than a simple keyword searching and indexing. Powerset uses a distributed database system called HBase in tandem with Coral, its Document Processing System. Coral uses Hadoop as its job control machine. Powerset uses 92 eight-core machines to do processing.

We get all giddy over here at GigaOM when it comes to storage and backup products, so it’s worth noting that today a service called Syncplicity launches in public beta. What’s nice about the service is it offers both storage and backup as well as automatic syncing across PCs. What makes it better than most is its ability to sync offline documents.

Right now, the bridging feature that syncs your offline work when you get online is only available for Google Docs/Word files and Facebook photos. However, Leonard Chung, co-founder of the service, says more offline syncing options will come soon. So will a Mac product. Because my PC is wonky at the moment (hey, it’s four years old,) I didn’t get a chance to try out the software, but if you guys do, please leave us a note in the comments section with your thoughts.

Initially the service is free, but Syncplicity plans to charge somewhere around $20 a month for unlimited storage and access. When compared to a backup service like Carbonite, which charges $50 a year, that seems pricey, but it does offer the syncing, backup and offline access all in one package. The company is using a combination of their own servers hosted by Rackspace and Amazon Web Services to support Syncplicity.

More GigaOM backup and syncing reviews are here:

• SugarSynch and Capzules
• FolderShare vs. Dropbox
• Drop it Like it’s Dropbox
• For Mac Users Some Backup Option

And for reviews from Web Worker Daily see here:

• 9 Free Online Storage Backup Solutions
• Who Protects Your Cloud Data?
• When Services Fail: Relying on Web Email Services for Work

Today’s Wall Street Journal devotes 2,400 words to Mark Zuckerberg’s hiring of former Google exec Sheryl Sandberg as Facebook’s COO, and his attempts to mature into the CEO role at a large company. After reading the article I decided to sum it up in haiku form for our readers.

Growing up is hard
Doing it in public sucks
Facebook needs money

All levity aside, as a young founder Zuckerberg is treading on the worn footsteps of entrepreneurs everywhere. College is one of the best times to start a company because you don’t need to be mature. You have the ability to spend all hours obsessing over your creation, not paying rent and generally avoiding other responsibilities. Bill Gates, Michael Dell, Marc Andreessen, Sergey Brin and Larry Page all began their efforts in or immediately after college.

Zuckerberg is lucky, his collegiate startup has soared. And he now has the unenviable task of growing from a self-absorbed adolescent into a gracious adult in public. Not everyone makes this transition and no one does it without a few awkward moments. He’s also hampered by a point of view that’s unique to entrepreneurs.

Like many entrepreneurs, he’s recognized his flubs and is seeking to address them. Also like a true entrepreneur, he doesn’t spend a lot of time apologizing for his mistakes, but opts to go out and act differently. Those of us in the real world can fault him for that, but that mindset isn’t as much a function of age as it is a hallmark of an entrepreneur. As Zuckerberg matures, he’ll still have that mindset, but he’ll likely hide it a bit better.

An in-depth look at the debate over web analytics over at Computerworld has concluded that page views are dead (although we track them assiduously at GigaOM), and like heirs squabbling over an inheritance, startups are fighting to define the page view’s successor: engagement.

The story mentions Israeli startup Nuconomy, which measures engagement by tracking things such as widget downloads and comments. Nielsen, which abandoned page views to focus on time spent on a site, thinks that’s a better measure of engagement, especially when it comes to online video. But a one-size-fits-all approach may not win out. The article quotes Avinash Kaushik, a web analytics consultant whose clients include Google, who says:

“A lot of people think the page view is dying so we should measure engagement,” he noted. “Just because the page view died, who…gave you the right to move to engagement? The web is becoming more fluid in terms of how people interact with it. The fluidity does not mean the core questions you wanted to answer go away.”

This isn’t just a semantic debate, it’s a struggle to define how advertisers will dole out the dollars for the myriad of online publishing sites and content networks out there. With online advertising reaching $21 billion last year, that’s an estate worth fighting for.

Outspark, a San Francisco-based casual games publisher with offices in Seoul, South Korea, launched its North American games portal yesterday. Like Nexon’s South Korean-developed MapleStory, Outspark games will be free to play — in addition to advertising built into the games and the portal, the company will rely on micro-transactions of virtual goods sales to generate revenue.

Their first game, Fiesta, published by OnsOn Soft in Asia, is an MMO currently in open beta. Outspark, which secured $4 million in funding earlier this spring from Altos Ventures and Doll Capital Management, plans to work with other developers to publish community-oriented multiplayer casual games as well.

I put a few questions to CEO Susan Choe and Chief Studio Officer Nick Foster yesterday to get a better sense of the company’s plans.

The micro-transaction model has been shown to be very successful in South Korea, where Outspark also has experience, but has been slow to take off in North America. Why do you think that is and why do you think it’s time to launch this revenue model here?

SUSAN: The micro-transaction model was slow to gain traction in North America due to a lack of payment solutions like those readily available in Asia. The response of North American gamers, however, to this type of game and item sales model has been tremendous and forms the basis of Outspark’s initial releases. Our expertise in running global portals like Yahoo (YHOO) and leading game product management at companies including EA (ERTS), Nexon, Blizzard and NHN will help us continue to deliver great results.

What demographic do you see as your primary target and how will you reach it?

NICK: Outspark’s initial target demographic is the youth market, specifically those between the ages of 13 and 24. Friendly, socially driven games appeal to all ages, however, and we’re attracting a diverse community of people looking for a different style of play than can be found in conventional console or hardcore games.

Your competition, in my view, is not necessarily World of Warcraft but socially rich Web 2.0 apps like Facebook and YouTube (GOOG). How will your products compete — or integrate — in that space?

SUSAN: Outspark’s goal is to provide a socially active virtual playground for online gamers. By providing games that players genuinely want to spend time in and building a community around that shared experience, Outspark can be a good partner for socially rich Web 2.0 companies by providing their communities with additional engaging activities.

You talked [in the release] about Outspark as a “platform.” Can you tell us more about that?

NICK: Outspark understands online gaming and the human drivers that make game communities successful. We’re combining our expertise in global entertainment with an understanding of virtual item sales and good game design. Outspark’s goal is to find media partners and work with them to apply this holistic “platform” approach to help build additional channels for their IP, around which online communities can grow.

There’s big action in search engines these days: see Web Worker Daily’s list of 11 alternative search engines, Read/Write Web’s top 100, and the GigaOM show’s opening question to Google’s (GOOG) Marissa Mayer about the spate of search engine competitors.

But even as would-be Googles aim at the apparent heart of search — the keyword or natural language query itself — the place of greatest user need could lie a level above, at the iterative and exploratory pathways that users often follow in finding and refinding the information they need.

We know who owns the search engine. Who will own the entire search experience? Right now it’s up for grabs.

Teleporting or orienteering to what you want

In 1993, researchers Vicki O’Day and Robin Jeffries contrasted two types of information-seeking behavior: teleporting and orienteering. With teleporting, you try to jump directly to what you’re looking for — think of the Google “I’m feeling lucky” search. With orienteering, you use local and contextual information to guide you, step-by-step, to your information destination.

In teleporting, all you need is the perfect search engine. With orienteering, you need a bunch of different tools: general and specialized search engines, browsing history, query reformulation and refinement, information refinding support, and so forth.

Here’s an example of orienteering from Jon Udell:

1. Find [a New York Times magazine article about Olin College, a clean-slate redesign of an engineering school]
2. Note the date: Sept. 30, 2007
3. Search the Times archive for Sept. 30, 2007
4. Restrict that search to the magazine section

Microsoft user interaction researcher Jaime Teevan told me that search engines “theoretically support teleporting, but in practice, they’re used for orienteering.” Despite this potential disconnect, she’s found that people are pretty satisfied with search today.

Users do, however, express feelings of being overwhelmed by information, and searches sometimes outright fail, as when Teevan observed a user trying without success to find the customary tip for a hairdresser. Could better orienteering support help with information overload and failed searches?

Why people orienteer

Teleporting doesn’t always work. Search engines don’t know enough about what we want to get us there, and we’re not giving them much help — people tend to use two- or three-word keyword searches. Better personalization will help, but won’t, on its own, remove the need to orienteer.

Orienteering has advantages over teleporting, too. It gives you confidence in the information you’ve found, teaches you the context for future searches, and allows you to backtrack if you happen to go down the wrong path. With teleporting, you either get there or you don’t, and if you don’t, you need to start almost from zero.

Search experience improvements, present and future

Google itself includes features that support iterative, step-by-step search with options such as “refine your search,” which suggests ways to quickly narrow in on just what you want; “similar pages,” which finds pages similar to a result you like; and a “spell checker,” which suggests alternative queries with more common spellings than the ones you’ve already used.

Search engines like Google, however, usually only see certain parts of the search experience. In refinding the article he knew existed, Udell may have used Google as a first step, but then he navigated through the New York Times site. No one search engine sees all of the pathways you take through the information forest.

The browser’s the place

Your browser already supports your orienteering, both with basic history features and by giving you access to all the search engines and web sites you wander through on your way to your destination. But it could do better, and product developers and researchers are aiming to improving it.

Microsoft’s Teevan is experimenting with a browser toolbar plug-in that supports refinding information online, a very common and often frustrating task. She told me other researchers are working on better history displays; for example, linked thumbnail images. Not to be left out, Google already has the start of an orienteering solution: its “web history” feature, which works via the browser-installed Google Toolbar.

Of course, other search engines have a place in orienteering as well, because orienteering-style search relies on multiple tools to gradually step towards the goal. For example, if a general search engine doesn’t get you moving to where you want, a vertical or people-powered search engine might. Still, there’s an opportunity here to improve the entire search experience — and you don’t have to build your own search engine to do that.

It shouldn’t come as a surprise: some of the early Googlers who have cashed in their millions and are looking for new challenges are hitting the escape button. We wrote about this pending exodus a while back. Some of them are leaving to start their own companies, a trend that seems to be gaining momentum.

My venture capital sources are telling me there’s been a sharp increase in the number of business plans crossing their desks with Googlers as founding teams. With Google (GOOG) stock being where it is, and Google continuing to pay top dollar for startups, it makes sense to play Sand Hill Road Roulette.

However, a few Google employees are changing zip codes and defecting to Facebook. My ex-boss, Josh Quittner, who is back to being a reporter, has this piece on yet another recent high-profile defection.

Benjamin Ling, a high-ranking engineer described by Google watchers as one of “Larry and Sergey’s golden boys” has defected—to Facebook.

For folks like Ling, the lure of a pretty young thing like Facebook is too much to resist. Of course, there is little downside if you have Googillions in your bank account. So what if Facebook turns out to be Fakebook!

Predictify, a social prediction startup, launched out of private beta today. It’s based on a belief in the wisdom of the crowd, especially when it comes to predicting the future. After all, if nine out of 10 people think the same thing is going to happen, odds are that outcome will occur, right? Well, so the site’s premise goes. Users are lured in with the opportunity to earn a few bucks here and there by offering up their guesses as to the outcome of future events, from which presidential candidate will win the Democratic ticket to the color of Angelina Jolie’s gown at the Oscars.

The site has an ever-changing list of “free” and “premium” questions. The premium questions, sponsored by companies looking to poll the Predictify audience or advertise a product, pay accurate guessers a percentage of the total money offered by the sponsor for their participation.

Guessers offer up answers to non-paid questions as well, and in turn help to “build a reputation” based on the accuracy of their predictions. The more accurate the user is over time, the higher their payouts are on the premium questions.

In order to be eligible for a premium-question payout, users have to provide certain demographic information, such as age, income, political party, and religion. The question sponsor gets to chose which specific information they wish to gather from the user and is allowed to keep the data gathered through that question private. Free questions, meanwhile, poll 100 users at most and the data is made publicly available.

Predictify competes against other social prediction startups, like ZiiTrend (which doesn’t pay users to predict future events) and traditional online survey companies that pay users to answer a series of questions online.

Company founders Parker Barrile and Mike Agnich, currently students at Stanford’s graduate school of business, started their Menlo Park, Calif.-based company in March. CEO Barrile previously worked at Google (GOOG) in business strategy, while CTO Agnich used to work at Zazzle, where he was VP of product technology.

We predict that unless the team finds some popular partners to pay for premium questions, the site isn’t going to last. Paying people to answer questions is a promising method of convincing users to sign up for the site, but the site doesn’t make it clear exactly how much money users can earn. It does, however, note how much the premium questions will pay out overall. It looks as if, currently, most of the premium questions on the site pay out a total of $500 (the biggest payout is $2000). But so far, according to a list on the site, the top-earning users in the private beta have made a total of just $9 . So we’re not really clear how broad of a user demographic the prospect of earning a few bucks for providing predictions will draw.

It’s a challenge Predictify’s founders acknowledge. “It’s absolutely a concern. We believe the Internet population is getting more and more diverse,” Agnich told us. Their belief in the site’s premise, however, remains undeterred. “We’ve taken a lot of time looking at CNN.com,” Barrile added. “They allow users to post comments after the stories, and a lot of those comments are actually folks predicting the outcome. Prognostication is human nature.”

What if the solution to having too many social web services was another social web service? That’s FriendFeed, a yet-to-be-launched service from some former Google (GOOG) employees that I’ve been trying out this week.

Yup, you’ve got to enter in your friends again, and upload a profile pic. But then you depart FriendFeed to do the things you normally like to do — upload photos, write 140-word broadcasts about your life, listen to music. Your friend, also a FriendFeed member, goes and does the things he likes do to — update his Netflix queue, write full-length blog posts, Digg stories he finds interesting. Through FriendFeed, you’re kept apprised of one another’s activities.

Say he starts writing reviews for Yelp — he doesn’t have to tell you about it, you just start seeing it show up in your FriendFeed. You get an aggregated feed, in one place, of everything all of your friends do, while you can push out an aggregated feed of everything you do onto a blog or social network (click on the thumbnail on the left for a screenshot of the setup, and the one below the jump for the public site feed). It’s a Facebook news feed for the whole web.

FriendFeed was just recently pulled together — it’s currently running a private beta test, incorporating, and closing funding all at the same time, co-founder Bret Taylor said in a phone interview on Wednesday. Taylor said he built the tool on the side while working as an entrepreneur in residence at Benchmark Capital with fellow ex-Googler and FriendFinder co-founder Jim Norris. “We became addicted enough to scrap what [else] we were working on,” is how he described the origins of the company.

The funny thing is, in some ways FriendFeed makes the web less social — stripping away the community features that make specific sites special. But it also makes the web more social, by emphasizing your real-world connections rather than relationships built around common interests or objects, and bridging together the little online islands where we express ourselves.

Taylor said early users tell him that “it’s not really about socializing in the traditional social network sense, it’s really about finding things to watch or read or listen to when you’re bored or waiting for something to happen.”

So what is FriendFeed, a big-picture idea or a side project? The question, said Taylor, is “Can we take the whole web and sort of project it through your social network?” Co-founder Paul Buchheit added, “We’re just trying to glue together the web.”

And if the impetus was improving on the Facebook news feed, what’s to keep Facebook from doing the same thing itself? “We want this to be open in all directions,” said Buchheit, citing various blog widgets and iGoogle modules. “It’s not a destination.”

With what appears to be a few hundred users in the beta, it was hard for me to find 10 people I knew on FriendFeed, and thus hard for me to see how useful this product really is. Still, we can’t be entirely uncritical: these are closely watched ex-Googlers we’re talking about, and they’ve already received a writeup in the New York Times. For me, some potential concerns that come to mind include private and public distinctions, and filtering loads of little alerts and activity records into something that’s not overwhelming. There’s also a risk of becoming too abstract — for instance, a FriendFeed item of a blog post that was Dugg could have at least three separate groups of comments.

Oh, and at some point this little project might have to make money!

Funnily enough, FriendFeed is not the only startup from former Googlers we’ve profiled today. Also check out my post on NewTeeVee about Ooyala, a video syndication and advertising platform.

With every new technology, innovation generates excitement, the excitement turns into buzz and the buzz, more often than not, gives way to a buyout in which a buyer overpays for a hot new startup — and triggers a slow deflation in the valuations of the startup’s competitors.

Cisco Systems (CSCO) spent lavishly for Cerent, eBay (EBAY) opened its checkbook for Skype, and Google (GOOG) ponied up for YouTube. Now EMC Corp. (EMC) has reportedly bought American Fork, Utah-based Berkeley Data Systems, the company behind online backup service provider Mozy, for $76 million.

BDS had raised about $1.9 million in venture backing, and as of April counted some 180,000 users of the Mozy backup service. Reviews of the service — including one in our own Web Worker Daily — have been positive across the board.

The acquisition by EMC is a surprise, though as Jeff Clavier explained on our show earlier this week, it is the non-conventional buyers (in other words, not Google, Microsoft or Yahoo) that are giving web 2.0 mergers and acquisitions a nice nudge.

The rationale behind EMC buying Mozy is still unclear, however. But one suspects that it would be a key component to them going after small- and medium-sized businesses, much as SoonR is doing by shifting its focus to backup-and-restore features. And although the $76 million sticker price does seem rich, EMC can afford to be lavish: it’s sitting on a windfall generated by the VMWare (VMW) IPO.

Valuations of VoIP providers haven’t been the same since eBay’s momentary loss of rationale, and while a shakeout in online video hasn’t exactly come to pass, there is little chance of someone cutting another $1.6 billion check anytime soon. For optical hardware startups, the Cerent buyout was a top, as was the case with another Cisco purchase, Stratacom.

From that perspective, the $76 million buyout of Mozy should be a warning signal for dozens of online storage startups that are trying to get our attention. Perhaps it is time for them to start accepting whatever offers they can get — fast.

PS: If you are aware of academic studies about big buyouts and the after-market impact on competitors, do let me know.

Back in September 2006, Nokia (NOK) CEO Olli-Pekka Kallasvuo declared that phones were no longer phones but multimedia computers that play back music, record videos, snap photos and — oh, yes — make phone calls. Apple’s (AAPL) iPhone has only reinforced that notion. And as the phone morphs into a multimedia marvel, there is a growing realization that the traditional user interface of a phone, the 12-key keypad, may no longer be enough.

The keypad limits how much information we can input into the tiny devices, and acts as a speed bump when we’re trying to navigate through a complex array of features. And what that means is that we need new ways to interact with mobile devices. Apple, for one, has bet on the touch screen and the fluid UI.

And then there are those who believe that voice input is the way to go.

Microsoft (MSFT) bet about $900 million when it bought TellMe Networks. Some startups are voice believers as well, such as Cambridge, Mass.-based Vlingo Corp, which I’ve previously written about. Earlier this week, I got a chance to see a demo by Yap, a Charlotte, N.C.-based company with a similar approach — that is, taking voice and inputting it as text for everything from IM to navigation.

Like Vlingo, you need to download a Yap application on your mobile phone to get going, and then use voice to enter everything from instant messages to TwitterGrams. Yap also does voice processing on the server side, and then sends information back on the mobile data channel.

There are others who are taking speech recognition even further — embedding it right into the chips that go into Bluetooth headsets. Cambridge Silicon Radio, a maker of Bluetooth chips, is now embedding speech recognition technology from Sunnyvale, Calif.-based Sensory into chips that will find their way into the Bluetooth headsets by the first quarter of 2008.

Sensory CEO Todd Mozer believes that everyone wants to do big things with speech recognition and mobiles, but in the end, the simple functions that enhance the hands-free experience are what make the most sense and are the most useful. Bluetooth devices make perfect sense as a starting point for voice commands. I agree with Mozer.

When I saw the Yap demo, I got the feeling that the application was trying to do too much; it needed some focus. After all, Yahoo (YHOO) and Google (GOOG) can take a similar server-centric voice synthesis approach and provide a more enhanced offering. Moreover, they can use their partnerships with large carriers to squeeze out little players such as Yap.

P.S.: All this interest in voice recognition and related technologies could explain the nice bump in the share price of Nuance (NUAN): up 64 percent for the year, even despite a recent pullback that’s mostly because company’s move into the mobile voice recognition arena isn’t sitting well with the Wall Street types.

Related Post: Sit Up and Listen, Future of Software.