Last week, after reading Corvida’s RWW post on image bookmarking services, I started to wonder why I’d never felt the need for such a service. I’m obviously a fan of social bookmarking, and my preferred service, Delicious, has no image functionality. Even before Corvida’s post, I’d been aware that Ffffound! and Vi.sualize.us could fill the gap; I just hadn’t signed up. I've just been sorta “meh” about image bookmarking.
Then I realized why: I’m already using an image bookmarking service. It’s called Tumblr. I use Tumblr’s amazing bookmarklet to save images all the time to my tumblelog, Lifeisathrill.com. It’s true no one thinks of Tumblr as a bookmarking site. It’s certainly a popular microblogging service, but it doesn’t have the obvious social components inherent in a bookmarking site. Which is to say it’s not an unfathomable zoo of OPC (other people’s content).
Tumblr’s image bookmarking bona fides
But let’s talk about what it does. Its bookmarklet saves a new copy of any photo to the hosted Tumblr site; you’re not hot-linking to some other site. And the process could not be easier:
I can save photos, add tags, and add unlimited text comments, including HTML. And it shows up perfectly formatted on my own Web site.
But what if I want to find it later? Tumblr doesn’t yet have built-in search, and its tagging features are admittedly embryonic. But, check it out! Tumblr has visual archive that’s much more useful if I’m looking for a specific photo. Note that I don’t have to remember what tag or text I used to describe it; I can search over months of archives in mere seconds and spot the thumbnail I need. For good measure, each thumbnail does include a good-size portion of text, which is searchable using the browser (ctrl-F).
What Tumblr doesn’t have
Tumblr is, first and foremost, a powerful blogging platform and content aggregator. I don’t just use it to collect images. I can post videos, quotes, conversations, and mp3s, all using the same flexible interface. I can also import feeds from Google Reader, Delicious, Twitter, or any other service that makes RSS feeds available. But as much as I love Tumblr, it’s NOT a true substitute for a social bookmarking service.
Essentially, Tumblr =
social bookmarking. The strength of any social bookmarking service is the recommendation engine—quickly discovering what others have found interesting. Or beautiful. Or scandalous. Or sinfully, horribly cute. Tumblr doesn’t have a recommendation engine.
Not yet, anyway.
But the potential is there. Tumblr already has the guts of a great social bookmarking service. Its got tons of active users submitting content in a fairly standardized manner, either through the bookmarklet or by importing feeds. It also has a full-featured, read/write API. There’s really no reason all that data shouldn’t be leveraged into a Tumblrmeme or Tumblr-icious.
Tumblr has taken baby steps in that direction with a number of features. For example, the Tumblr Radar showcases content from around the Tumblr-sphere. Like real-life radar, stuff tends to fall of without a trace, and there’s no RSS feed, but you can browse recently featured Radar items here.
Every user’s dashboard features a Livejournal-like stream of activity from the Tumblr blogs you’re following. You can choose to re-blog items that your friends have blogged, and popular items feature a handy counter showing how many times an item has been re-blogged. Unfortunately, I don’t keep up with my dashboard as much as I’d like to. That’s because (wait for it…) there’s no RSS feed.
So yeah, unlike Delicious or Stumbleupon or FriendFeed, Tumblr isn’t exactly optimized for content discovery. There’s no way to search for items or sort by tag (although Yahoo! Pipes ninja Joe Lazarus provides a nifty tracking option). And you can’t easily identify popular items, because Tumblr doesn’t have a robust way to measure the zeitgeist. How cool would it be to see Tumblr fed into popurls? The Radar and Dashboard features prove it’s not unthinkable. Indeed, with Tumblr promising a major upgrade soon, I’m hoping to see some progress on this front.
Even if Tumblr doesn’t create an enhanced portal for exploring all that content, the API seems like it would allow another developer to do so. But Tumblr hasn’t really caught fire amongst the tech bloggers and developers. Tumblr has been criticized for its notoriously long dev cycles and the stingy pace at which they add features. Rather than adding features every few weeks, it seems they’d rather unveil a complete overhaul after a long period of dormancy. That’s a great strategy for an occasional big bang, but it’s not a good way to keep bloggers and developers interested in your product.
What could be…
Tumblr may not ever generate the breathless kind of buzz FriendFeed and Twitter have enjoyed. For me personally, Tumblr has been (by far) the more useful and rewarding product. It gives me a little place the Web that doesn’t look like a hodge podge of voices; it’s my voice. But underneath this flexible and intensely personal blogging platform is a sturdy framework that could support a world-class bookmarking site and recommendation engine. It’s these dual possibilities that make Tumblr so fascinating.
Tumblr founder David Karp is aiming for a million users by the end of 2008. Imagine peeking inside a million separate hearts. At once.
UPDATE: I should point out that the aforementioned Joe Lazarus has built a few good Tumblr aggregators including one for frequently Tumblr'd photos, popular quotes, and one for songs. None is the all-in-one portal I'm thinking of, but he gets bonus points for actually using Tumblr to create his mashups.