Webware

Social news site Digg is experimenting with a new way to give upcoming stories a chance at the limelight with an experiment the company is calling Digg Trends.

Stories that begin to experience a heightened amount of user interest in the form of off-site sharing, user discussion, and of course Diggs, will be presented up at the very top of Digg's home page, as well as being spouted in a special RSS and Twitter news feed. Once at the top of the page, those stories have 10 minutes to get voted onto the front page as a normally dugg story, otherwise they're buried into Digg's dead pool. All the while a giant counter ticks down how much time the story has left.

Along with the countdown timer, Digg is also putting forward some of its outgoing traffic numbers. Users can see how many clicks a trending story has gotten from the front page. Normally, the only other traffic numbers you see on a Digg story is when you're on the source site itself, though users must have the DiggBar enabled.

Trending stories get just 10 minutes to prove their worth like any other front-page story. The company is also making available how many users have clicked to view the source content.

(Credit: Digg)

This new system is a stark difference from the somewhat nebulous promotion algorithm that exists for regular stories. Under the current system, stories have to earn their way onto the front page which involves standing out among an ever-growing pool of other upcoming stories. The company made this process a little more custom-tailored with the introduction of its recommendation engine, but it still requires that users actively visit that part of the site to see what's new. The new trending idea puts some of those stories up for everyone--and right on the front page.

In a company blog post about the new process, Digg's senior software engineer Kurt Wilms called it an "experiment," and said that it could change based on user feedback. Some Digg users have already voiced their opinions in the posting's comment section, citing that "bury brigades" (groups of highly opinionated users) could keep some stories from ever making it past their 10-minute window. The end result being that a story that could have legitimately made the front page on its own gets shut down before ever having a chance under the normal algorithm.

As with other new features, Digg seems to be rolling out Trends slowly, and to a small group of users. I'd expect it to show up for everyone in the next day or two.

Note: Gifsoup, as it turns out, is a violation of YouTube's terms of service, which state that user-submitted videos cannot be downloaded unless the author has allowed it and the download is taking place on YouTube.com, and not via its API.

In my book, animated GIFs are one step above glitter graphics in terms of junk trends of the Internet, but I'm a big fan of any tool that makes creating them easy and fun. Gifsoup is no exception--you just point it towards any YouTube video and it turns it into an animated GIF.

To do this, it first downloads the clip to its servers, and then gives you simple controls to choose when you want it to begin and end. When you've picked out that perfect 10-second (or less) section of the video, you just hit a single button to finish the job. The GIFs are then hosted, and ready to be embedded elsewhere; you're also able to save it to your hard drive in one of three sizes. Either way, it retains a small Gifsoup watermark that sits in the bottom right-hand corner of the image.

Below are two I made in less than a minute using Gifsoup's tools.

See also: Gickr does software-free animated GIF creation (which uses multiple image files instead of video)

Finding the hot conversation keeps getting easier, but predicting what the next big trend will be continues to be a crapshoot. Palm and Federated Media have teamed up to create a new tool called Trend Tracker that does its best to figure out, what in fact the next top trend will be by analyzing items that are gathering buzz.

The system is a mix of tools that can help spot popular URLs and trending topics before they hit it big. But it's more about organizing that data in a simple-to-parse format.

Included are the current top 30 trending topics on Twitter, which can be stacked up against each other to see what's pulling in the highest percentage of tweets. Each trend is represented over a 24-hour time line, where you can see how each particular trend has gone up or down in popularity.

Trend Tracker can give you a visual analysis of when each trend became popular, as well as its decline.

(Credit: CNET)

But 24 hours doesn't tell the full story, which is why the tool will soon expand to keep an archive that covers the last 10 or 30 days.

Along with the top 30 trends, Trend Tracker includes a "Pre Trend Watch" (emphasis mine) which tracks five up-and-coming trends that are about to break into the top 10 based on their velocity--the speed in which tweets on that particular topic are gaining in popularity. These are also marked in the trend archive with a little blue flag.

When I was looking at the tool last week, one of the most interesting things this picked up on was... Read more

(Credit: James Martin/CNET)

SAN FRANCISCO--The claim has been made in the last couple of weeks that cloud computing has reached the top of analyst firms' famous hype cycle and is a top-of-mind issue for most IT organizations.

That's a bit misleading, as the interest in cloud computing is often taken out of context, and when you bring virtualization into the picture, that interest seems to remain exploratory rather than strategic.

Amazing innovation is happening in both public- and private-cloud offerings, and the overwhelmingly positive response to cloud computing--in particular to Amazon's top-notch Elastic Compute Cloud, Simple Storage Service, and related offerings, as well as Google Apps and the first generation of software-as-a-service superstars, such as Salesforce.com.

But the critical truth--that interest in virtualization technologies currently outweighs interest in the cloud-computing model--has been evident at trade shows I've attended over the the last several months targeting subjects ranging from networking to next-generation data centers to cloud computing itself, and it has hit home here at VMworld this week. The bottom line is that virtualization is where the money is this summer; cloud computing isn't.

Technology trends follow the patterns described by the science of complex adaptive systems. There is constant change and mutation, and there is a feedback loop that encourages stronger innovations to survive and grow while killing weaker ones, yet somehow, the system maintains a working balance that doesn't get too chaotic to manage or too ordered to allow innovation.

As with any complex adaptive system, traits that eventually come to dominate the system tend to start small: a single mutation, or the introduction of a small number of invasive foreign entities, for example. In the case of the "invasive" cloud computing model, the "DNA" is strong.

Amazon Web Services proves that you can get your infrastructure over the Internet. Salesforce.com proves you can run your business relationships through a browser. Both public and private clouds introduce flexibility and efficiency into IT services.

Cloud-computing bellwethers
Cloud computing is definitely in your future, in one form or another. It probably already plays a strong role in your day-to-day computing experience. That said, when you measure audiences at technology trade shows such as Cisco Live and Interop, you see where the real interest of the everyday IT professional is. At VMworld, the audiences at virtualization-related sessions have been consistently larger than those at cloud-centric sessions.

Recent cloud-only conferences have remained quite small--typically in the tens or hundreds of participants--in comparison to their brethren, and cloud-focused sessions at larger shows have been attended by fewer people their virtualization peers. Several of my cloud-focused colleagues have even noted that some shows end up with the same vendors pitching to each other over and over again.

Without a doubt, this is simply an indication of the current stage in which we find ourselves in the long evolution from internal data centers to cloud-centric operations. The ratio of interest will change (or, more appropriately, converge). But if you want to get into the head of most IT tech geeks today, you need to address the subject of virtualization first, then acknowledge cloud computing as a future target.

The best evidence I can personally attest to are the breakout sessions and panels in which I've participated. I have been giving two basic talks this summer, one focused on cloud computing's future ("Achieving the Intercloud"), and one about the journey from virtualization to cloud computing. Without a doubt, sessions with the term "virtualization" in the title have seen the best attendance, whether measured by room capacity or interaction after the talk.

Transitions ahead
What does that mean to the average cloud enthusiast? Well, for one thing, it remains important to see cloud computing as a transition--an operations model that requires addressing technology and cultural issues before widespread adoption. The good news here? Current trends in virtualization, automation, and early cloud offerings are forcing most of those issues to be faced head-on.

It also highlights how much work is ahead of us in helping those responsible for application operations see the value in cloud environments. This education will be greatly accelerated this year, thanks to the amazing work that customers large and small are doing, especially in public clouds. However, it will also require technologies that address the concerns that many have about moving virtualized workloads into someone else's infrastructure.

I'm betting that at this time next year (or the following year, at the latest), most of the convergence of virtualization and cloud interest will have happened, with the exception of the continued interest that service providers and enterprise data center operators will have in the physical infrastructure and management systems needed to provide cloud services.

It will be harder and harder to tell the difference between a talk discussing how to manage an application running in a virtual machine and one discussing how to manage a cloud workload. Many management vendors will demonstrate tools that manage virtualization (such as VMware vSphere) and public cloud services (especially Amazon's EC2 and S3) at the same time, with the same interfaces. Long lines will be form for topics that will have little to do with who owns the infrastructure or how it is paid.

At that point, the decoupling of physical infrastructure management and virtual workload management will nearly be complete--and the cloud-computing DNA will really begin to take over.

I always enjoy wild hand-wavey prognostications about the future, so I was pleased to attend the 11th annual Churchill Club Top Tech Trends event last night, moderated by my former co-workers from Red Herring, Tony Perkins (now running Always On) and Jason Pontin (publisher of MIT Technology Review). Of the 12 trends, two really made me take notice. Most of the rest, which you can see at the end of this story, were pretty standard projections from existing market circumstances.

Trend prognosticators, left to right: Tony Perkins, Vinod Khosla, Steve Jurvetson, Ann Winblad, Ram Shriram, Joe Schoendorf, Jason Pontin.

(Credit: Rafe Needleman/CNET)

Interesting trend #1: Centralized search will fall

Venture capital whiz-kid Steve Jurvetson gave an impassioned pitch for this trend, which he called, "The triumph of the distributed Web." He said the aggregate power of distributed human activity will trump centralized control. His main point was that Google, and other search engines that analyze the Web and links, are much less useful than a (theoretical) search engine that knows not what people have linked to (as Google does), but rather what pages are open on people's browsers at the moment that people are searching. "All the problems of search would be solved if search relevance was ranked by what browsers were displaying," he said.

Jurvetson believes that the future is "federated search," in which the Web's users don't just execute search queries, they participate in building the index by the very act of searching, immediately and directly.

What I find most interesting about this concept is that we can see it already happening, although via a different technological vector. Twitter Search is real-time search. It tells you what people are saying right now, and on popular topics, it gives you far more current information than Google. I think Twitter Search also shows us that Jurvetson's vision of search, while compelling, is incomplete. To get the real-time wisdom of the crowds for the purpose of search, you have to know not just what Web pages people are displaying, but exactly what is on those pages, and you probably also want to know what's showing up on users' computers in apps other than the Web browser.

I am not sure the Web's users will want to participate in the creation of this search engine, nor am I convinced that there's a lot of value in the concept for obscure or "long tail" search queries. But the idea is interesting, and I certainly agree that the value of real-time searching, as well as social-network-aware searching, will increase dramatically and quickly. ... Read more

Correction at 4:13 p.m. PDT Thursday: Roni Zeiger's last name was misspelled in an earlier version of this post.

Google is attempting to find out how much of a role Internet searches play in the self-diagnosis process.

The company plans later Wednesday to start rolling out a subtle question at the bottom of pages with search results for a few common ailments, such as "Did you search because you or someone you know may have an ear infection?" That question will only appear for a very small number of users who search for terms such as "ear infection," but it will help Google start to understand how many people are searching on such terms looking for treatment remedies or options as opposed to doing research, said Dr. Roni Zeiger, a product manager for Google Health.

Understanding how many people are searching on Google for help diagnosing their health could improve future search results, the company thinks.

(Credit: Google)

In a way, this is an extension of the work Google has done tracking the flu with Google Flu Trends. The company noticed that search activity related to the flu tends to rise about two weeks before a similar rise is reported to the Centers for Disease Control by doctors, but years of data on flu patterns validates those trends, Zeiger said. Similar data does not exist for more common health issues.

Google is not exactly sure what it wants to do with that data, or how much useful data will be produced by the experiment. Ultimately, however, everything at Google goes back into the search process, so it's possible that the data could be used to offer searchers more options, such as "Did you mean to search for treatment options for X?" at the top of the search page.

This is a temporary project: Google plans to gather data for several weeks, starting Wednesday afternoon.

(Credit: Compete.com)

Unless you have been inhabiting the underground bunker formerly occupied by Dick Cheney, you've probably seen loads of press coverage over a "25 Things About Me" Internet meme that was spreading on Facebook. Basically, members would create a Facebook "note" containing 25 facts about themselves, and then "tag" 25 friends encouraging them to do the same.

Yes, it was a bona fide phenomenon, but I avoided writing about it, because I thought the whole thing was...dumb. Internet memes of that nature have been around since goodness knows when. Breathless press hype over it seemed a tad silly.

But here's something legitimately interesting. Analytics firm Compete.com says that there may actually have been a boost to Facebook traffic as a result of "25 Things," at least in the U.S.: 60 percent more Facebook profiles were created in January than in December. That's not surprising, because Facebook still requires a user account to access all its content--curious newcomers who read about "25 Things" would need to register for accounts in order to explore it.

More noticeably, U.S.-based traffic to Facebook's "notes," normally one of the social network's quieter features, skyrocketed. Four times more visitors than usual hit up the notes feature in January, according to Compete, with 28 percent of Facebook's U.S. users checking them out. (The wildly popular photo-album feature usually draws 60 percent of visitors, for comparison.)

The caveat is that Facebook continues to grow fast and so some of this could be attributed to natural growth rather than "25 Things" momentum. That said, Facebook's U.S. growth has long since started to stabilize--three-quarters of its new users now come from overseas.

Compete has said that its analysts will be posting a blog entry about this later in the week, ideally with some more insight into just how much those annoying "25 Things" lists really did catch on. I've also pinged Facebook to see if they have any internal numbers on the topic.

Here's what'll be interesting to see, at least from my perspective: Will this mean that the newfound popularity of "notes" will last? I post photos, links, and other share-able items to my Facebook profile all the time, but I think I've written a Facebook note a total of once (to alert my friends list that I'd lost all their phone numbers in a personal-electronics mishap). Note-writing always struck me as something that was a little bit too promiscuous for the mainstream Facebook user, the sort of thing that navel-gazing, overshare-prone Twitterers would spring for but which didn't fit in quite as well with the directory-like nature of the social network.

Guess I was wrong. Facebook CEO Mark Zuckerberg, after all, likes to say that Facebook has incrementally made the Web's masses more comfortable with sharing more and more information. The success of "25 Things," consequently, must be one of his great triumphs. And now he knows all these useless facts about so many millions of people.

Heaven forbid: Facebook notes could be like a gateway drug to blogging for everyone.

This post was expanded at 9:51 a.m. PT.

Itzhak Perlman? Isabel Toledo? Simple Gifts? Huh?

During Tuesday's inauguration of President Barack Obama, people curious about unfamiliar references used Google to supply the footnotes for the ceremony. The phenomenon was visible on Google Trends, a service that shows which search terms are rapidly rising in use.

Inauguration-related searches were hot on Tuesday, according to Google Trends.

(Credit: Google)

According to the U.S. results, Toledo, who designed First Lady Michelle Obama's dress, bubbled up to fifth place on the list earlier in the day. Once the ceremony began, up came violinist Perlman (ninth place), cellist Yo-Yo Ma (12th place), composer John Williams (26th place), and the variation on the Simple Gifts melody (14th place) that he wrote and the musicians played. Aretha Franklin rose up to third place for a time, too, and even "My Country, 'Tis of Thee" ranked 21st at one point.

People were curious about politicians, too--Sen. Dianne Feinstein made it as high as eighth place, and "Dick Cheney wheelchair" was 91st place.

More interesting, perhaps, is that in aggregate, every single one of the top 100 Google Trends searches were related to the inauguration on Tuesday. Many had to do with people's evident desire to find news about it or a place to watch a streaming video.

Update 7:39 a.m. PST January 21: See this Google blog post for some more details about the phenomenon. For instance, there was a lull in regular search in the United States while people watched the inauguration, and 12 percent of inauguration-related search queries came from outside the country.

All of the top 100 searches on Google Trends were related to the inauguration. (Click to enlarge.)

(Credit: Google)

Google on Tuesday unveiled a new site to track the progress of the common cold.

Using the same keyword tracking technology found on Google Trends, it keeps an eye on people searching for queries involving the word "flu" and tracks them both by date and location.

What makes the technology so fascinating is that its data set goes back to 2003, and has been cross-referenced with the last several years of survey data from the Centers for Disease Control and Prevention (CDC). Google says that because its own system is based on a constant flow of searches as opposed to surveying techniques it's able to provide results one to two weeks faster than the CDC.

The same trending technique could be used in tandem with other organizations to track contagious viruses or threats besides the common cold, including AIDS, bird flu, and Africanized honey bees.

One limitation of the current system is that it does not track worldwide flu traffic. There is, however, quite a bit to discover from data from years prior--especially when you get several years that stack up on top of each other with similar rises and falls during certain parts of the year. According to Google's chart, we're about three weeks from hitting the heavy season, which goes until early January.

Google Flu Trends tracks flu activity across the United States.

(Credit: CNET Networks)