Tag: Hunch

  • ‘Like’ Minded People

    I read an interesting post by Dare Obasanjo titled “There will be many social graphs“. It took me back to the context of my post on Google’s social plans, where I’d mentioned the possibility of creating networks around different contexts – with not all connections being ‘friends’- a Twitter kind of asymmetric relationship, and how Facebook and Google both have an opportunity at that level. The form and kinds of data that we share – blog posts and thoughts, status updates, photos, videos, answers, people and the contexts we share them in, are many, and sometimes I wonder if one service can actually aggregate all this, while still providing user friendly privacy options.

    When i read (and saw) that Facebook is replacing ‘boxes’ with tabs, and also saw that my ‘Like’s (Interests and Likes seem to be undifferentiated now!) were now displayed prominently on my profile, I wondered if Facebook could really aggregate everything. Imagine, if those Likes+ interests were differentiated – i.e. Interest was ‘Music’ and ‘Greenday’ was a Like. Now, the way I’d like it is, if I had Interests displayed on my Profile page (or a tab, if you prefer) – there could be options of ‘how many’, ‘most active’, ‘most recent’ etc, with the existing ‘who can see’ privacy option, but more finely grained for each interest. For each interest, I should be able to build a page – with third party content included – subscription to blogs on the subject (either through FB Notes or say, Networked Blogs), Facebook Questions  and Quora, it could be Facebook groups, Pages that I have liked (so the interest ‘Music’ could have every artist/band/music media brand/ label I’ve liked), Friends who share the interest, people I ‘follow’ in that interest category (will explain in a minute) and when FB plays location, include that too, and sync Events. Goes without saying that I should have micro-level privacy setting options for sharing with others. I should be able to ‘Like+follow’ an interest of a person even if I’m not his friend (assuming he’s kept his interest public),  and even recommend to my friends.  A sort of ‘Twitter list’ for each interest. Yes, of course I need to be able to import Twitter lists too. There would also be a universal ‘Interest’ page that collates data from all the Interest Pages created by individual users, and also gives suggestions on ‘Whom to Follow’ for that interest – an algorithm based not just on mass ‘Like’s, but also basis contexts like Location, sub-genres, and my previous activity. To scale even further, use (mass and personal) data from services where I’ve used some form of Facebook Connect. Of course, Facebook would then have ginormous data on me, but they have it anyway, so I’ll be optimistic and hope that they use it to ‘personalise the internet – like Hunch, than for anything evil. Of course I’m assuming I get data portability too. Then maybe the different Facebook Search options can also really have fun. The entities who want to ‘engage’ users would also find this useful. I realise that I might be being simplistic about this, but what about the direction?

    And though most people are skeptical about Google’s social efforts, perhaps justifiably so (read this at GigaOm and Stowe Boyd’s “Can Google go Social“), and the Wave crash doesn’t really help perception, I don’t want to rule out  the possibility  (like I said in the earlier post) of Google getting over their privacy agony, and surprising us – imagine the ‘Interests’ as a separate service/ something around or integrated with iGoogle/Profiles/Buzz (brrr)/Chrome (browser or OS)/ Search itself.

    While on interests, suggestions and discovery, Twitter’s ‘Who To Follow’ hasn’t excited me much in terms of the people it has suggested. It says that the algorithm is based on people you follow and those they follow, but for now everyone’s busy trashing it, using its own acronym – WTF. So, how about using interests (Why To Follow – work harder on the existing Interests structure?) – either ask me when when i register, you can ask me now too, the lists that include me, keywords from tweets and bio, hashtags and hopefully ‘learn’ my preferences over a period of time. Popularity by itself is really not that great a parameter – if they’re popular, chances are I already know, and there’s a reason I don’t follow, even if its ego 😉

    Its not as though these are the only guys who can build a more nuanced social platforms – perhaps its possible for someone like Quora to start with questions and build more – eg. relevant posts from say Networked Blogs, and more people from Twitter Lists? Foursquare, or any of the LBS could scale too – from places to activities and consumption that happen at those places.

    Meanwhile, interests, context, relevance, building authority and influence, all of these are established on identities,  but there’s a debate on whether an old friend merits a return – anonymity. 🙂 More on that later. 🙂

    until next time, interesting?

  • Re: Search

    There’s this lovely Greasemonkey script that does the job of getting the best of both worlds- Google and Twitter, for me, but then it would be quite interesting to have Google bring out a microblogging search engine. (conjecture, but very much in the realm of possibility) In fact, with the recent pain I’ve been having with Twitter Search, it would actually be a big help. What this would do to Twitter Search and would Google add sense (in terms of ‘authority’/ relevance etc and indexing shared links) and Ad Sense to it, and would they share revenues with Twitter are a few questions that interest me. Let’s wait for the engine before we talk about that.

    Meanwhile, the Facebook vanity url brought out an interesting service from Social Too. According to RWW, Social Too is “extending our existing Facebook profile redirect URLs, which redirect yourusername.socialtoo.com to your Facebook profile, and adding an additional layer of analytics to the whole process. So tonight, you’ll be able to get a Facebook vanity URL and get the SEO benefits, but the URL you’ll want to hand out to all your friends will be your SocialToo vanity URL because we’ll provide statistics around those clicking on the URL, the browsers they’re using, where they’re located, and where they’re clicking from.” This would be a very useful resource for brands and even people, just ask any blogger who obsesses over page views 😉

    So what is Facebook upto? I just read that Facebook has rolled out a beta version of its news search that allows people to search their News Feeds and brings up results chronologically from their streams. It means that I can now easily see what my friends think about a recent event/product/service, and that can be notes, videos, status messages, photos etc. ( I did ask for that 🙂 ) The kind of search results that used to get displayed in the earlier avatar of search (people, groups etc) are now listed on the side. (via Tech Crunch) Wonder if it searches comments on the shared content too? Also, thanks to this beastly tweet from Karthik 😉 ,  I noticed that there is a location tab in my newsfeed now. Nothing much for now, but that is a start, and they could add events, among other things. For instance, since a lot of people are auto updating their Twitter status on to FB, I noticed quite a few with the #cisia tag, which is for an event happening in Bangalore. But both the friends using the tag are from Mumbai, so they don’t show up in the Bangalore feed. But yes, it is a start.

    So it does seem that Bing’s launch and the simultaneous bundling of One Riot with IE8 (One Riot is a real time social search engine) has shaken the search box a bit, especially in the realm of real time, though it could be just a coincidence. I have wondered why Bing didn’t launch with some One Riot like mechanism, that would really have been a differentiator, at least in the short term. A reason why I should use Bing, which seems the problem that MS should be addressing.

    Before we end, while we’re on Search, a couple of services I came across that you might want to look at. Hunch, a new discovery engine, from Caterina Fake, co-founder of Flickr, helps you answer questions, basis its understanding of you from the questions it asks you and your further usage of the site. Yep, this is a decision engine which asks you a few questions first 🙂 Mashable has a full analysis on it. I’ve just started using it a couple of days back, and it takes some time to get used to, and what it can do for you is directly proportional to the time you spend on it. So the task for Hunch would be to keep nudging people to interact with it more.  While I did start with a simple question, for which the sponsored link was an obvious choice, i think the question-answer way of search gives sponsored links a lot of relevance. I like the me+crowd way of answering questions and I think it just be the kind of service that grows on you. The other service I came across is Aardvark (via WATBlog – and they’re giving away free invites) which again uses a community network (via IM) to answer your queries.

    Google has claimed that search is in its infancy and there are many avenues for new and existing avenues to explore. Now while Hunch and Aardvark seem to be great products, if Google puts its mind to it, there’s no reason why it cannot replicate either of the services, especially considering the data it has of me from its various services I am a user of. Also, Wave, when it happens, can build on the power of incremental networks easily.  If the microblogging engine does emerge it will give Google the path to real time too.  The advantage for now, is that once these services reach a certain usage level, a me-too product, even from Google will have to work hard on getting people to jump ship, simply because of the lethargy and the time, content etc already invested.

    until next time, the search continues