Now, that’s a pivot. I wrote about Sulia a year and a half ago as an example of a smart company building a business off of filtering Twitter’s data for publishers and other site developers who wanted to create topically related tweet streams. Back then, the Sulia consumer-facing product was more of a proof-of-concept than how the company would make its money. But now, post Twitter’s API policy revamp, the company is reinventing itself. I had hoped Twitter might see Sulia as a valuable B2B partner in its ecosystem, but apparently they’ve been on the outs for a year. Sulia makes smart use of humans plus algorithms to curate short-form information topically. I hope there’s a media business buried in there.
I typed that headline while shaking my head. Google+ is iterating – and integrating – quite rapidly, but it isn’t solving some problems its early adopters are encountering. Matthew Ingram agrees with Robert Scoble that it’s too hard to filter signal from noise in Google+, and that content discovery suffers. Google still hasn’t done anything with search, and relies on Circles that are people- rather than topic-centric. Twitter has similar issues. (Facebook’s discovery is at least as bad, but it’s less noisy because Facebook’s feed-ranking algorithm has a heavy hand.) I’ve advocated that Twitter use Sulia’s topic filtering for the mainstream; Google could do something similar. Here are some other ways to build a better feed.
Just as 140Proof, an ad network for real-time feeds, hires a new sales VP from Gawker Media, we’ve released a new report on the market opportunities and challenges. Twitter doesn’t want anyone but itself serving ads within the feed, and lately it’s been de-emphasizing what it calls Promoted Tweets in favor of other formats. That leaves tricky issues around sponsored tweets, or targeting feeds on third-party Twitter clients or other platforms. Last fall, I wrote about how to get into the market early. It’s still relatively early in this immature but potentially lucrative segment of social media advertising.
It’s a stream of Twitter today. The company explains why “WikiLeaks” wasn’t trending: its algorithm favors velocity over popularity, though it uses both. Meanwhile, a Pew Research Center survey suggests fewer than 15 million adults use Twitter in the US. Yes, there’s a margin of error in the survey, and Twitter has more than half its users outside the US. Butt there are a lot of inactive accounts, brands, and bots in Twitter’s self-proclaimed 175 million global registered “users.” (Treat the survey’s activity stats with caution – the sample size of Twitter users is about 100.) Finally, with the loss of its product head, tongues are wagging about a lack of innovation at the company, or even corporate turmoil. Horrors! A fight between features and monetization? Who ever heard of such a thing?
There’s talk of the future of media in the air. So let’s examine how NewNet technologies like social media and real-time feeds are helping to re-invent the business.
As the NewNet cognoscenti await the outcome of Groupon’s reported board of directors teleconference, I notice there’s a lot of activity surrounding another hyped, fast-growing, and still private social media player. That would be Twitter, that may be adding a money guy to its board and may soon be taking in Kleiner Perkins money. Meanwhile, Twitter integrations go both ways: Google added data from Twitter feeds into its News homepage, while Quora enabled question-following on Twitter’s feed. How will Twitter make its breakthrough? A publisher thinks it’s a publishing problem, that if Twitter can migrate its primary follow mechanism away from people towards “interests,” it will be more valuable to advertisers and users. Perhaps. But Twitter is a two-way communications path, and some people are more interesting than their primary Twitter topics. How would natural language interpretation and pattern-matching handle that?
Twitter may be developing a news service for professionals. Or, more likely, it’s not, and that’s just Biz Stone thinking out loud. Here’s some advice on what such a thing might do, even as investigative journalism seems to be thriving alongside of social media. Twitter licenses and syndicates various chunks of its feed to developers already, and many journos and bloggers I know already use Twitter or a third-party app like Tweetdeck to filter news streams by topic. Although higher-powered filters and feeds could be powerful tools for news organizations, those potential buyers don’t necessarily have big budgets. I think that if I were Twitter, I’d leave this opportunity to the ecosystem, and focus on building products for consumers and marketers.
This Guardian interview with the head of Facebook engineering is interesting, and a little scary. If this is an accurate portrayal, it’s definitely a hacker culture at Facebook. Very small teams work on “several dozen to 100” mostly short-duration projects, with little in the way of formal oversight processes. Their project mix is 80/20 focused on features that will increase usage or grow the number of users, versus generate revenue or cut costs. Top priorities include mobile, infrastructure (server, database, development tools), and machine learning. Machine learning is critical to feed the algorithm that powers users’ news feeds, as well as for ad targeting and security – spotting spammers, etc. And Facebook is mostly a Mac shop.
Do we really need another browser, even one that’s optimized for social media and information feeds? RockMelt launched this weekend, and it’s getting a fair amount of attention because it’s funded by Andreessen Horowitz. Some power users remain unimpressed because it’s not mobile, it separates search, and isn’t real-time enough. RockMelt is based on HTML 5 and the same Chromium engine as Google’s browser, and it does a lot of pre-caching for extra speed. It requires a Facebook log-in and puts your favorite friends on a pane right there in your face. The original “social browser,” Flock, never gained traction, but it was early – a little before Facebook became huge – and somewhat geared to blog-posting, a minority activity. RockMelt is less about creating and more about consuming the social and info feeds, so it feels more on-trend to me. But this is a crowded space, and I see no evidence of an innovative distribution strategy.
In the realm of might-have-beens, MySpace shows more promise than Friendster. But unlike some former web leaders, it’s still salvageable. MySpace shouldn’t try to challenge Facebook for social network leadership, but it can remain a valuable consumer media business, if not a technology driver. Here’s how.