October 29, 2007


For the past three months I've been thinking a lot about user interfaces & user experience. The thing that started me on that, I think, is Flickr. Flickr as you all know «is almost certainly the best online photo management and sharing application in the world.» I admit, I am a big fan.

However, as good as Flickr may be, there's room for improvement. Considering that Flickr has such a rich data model underneath — photos, tags, people, groups — the ability to interact with it are frankly speaking quite crippled. I think that the page-based interaction metaphor we're used to from the web is not a good match for something like Flickr. The following image provides you with a remarkable insight into Flickr's complex data model & interactions:

The main problem I see is context. Considering myself a heavy user, many times I browse Flickr, in a sense I get «lost.» Not literally but nonetheless I wish the platform would provide me with more feedback on where I am in respect to time, location, people, groups etc. While the underlying model is very rich, its visualization and interactions on the surface are not. At least not as much as they could be. During the recent months I asked myself the following question, which in itself are probably not very relevant but illustrate the problem I want to start a discussion on.

In Flickr…

  • …when I am looking at a photo, why can I only go to the next or the previous? Why can't I easily skip twelve?
  • …why is it possible to comment on a photo, not even seeing it because I have to scroll down to the comment field?
  • …why is it not possible to view the photos in a different order? In order of popularity for example.
  • …why can't I easily rearrange, filter or group them so that they are easier for me to explore? For example, by events, people or location.

Many open questions. Although each single problem they point out is probably easy to solve or already solved (although not easily accessible), doesn't it question the whole structure itself? Some of these problems arise because Flickr is trying to push its rich data model into the rigid structure of pages. A metaphor that works well with textual information. Flat information. However, it was never meant to provide means to handle a model on which for example Flickr is based on. Flickr is not flat. If you don't believe me, check out the illustration above.

It is apparent that Flickr is not a web page but rather a web application. Now the question is: Why do we interact with it as if it was a web page?

I honestly don't know. But I know we can do better.

Having followed the uprise of «Rich Internet Applications» very closely myself, I sat down, did a lot of brainstorming and started to come up with something that tries to improve the situation at least a little bit. Ever since I wrote my final paper in high school called «Architecting Rich Internet Applications», times have changed quite a bit. The technology for this form of applications has improved drastically but more importantly, penetration & user acceptance have increased dramatically due to their ease of use and the (still ongoing) shift in perception within big companies towards better user experience.

For today, I suggest I jump straight ahead and give a glimpse on an early version of what I came up with. In future posts I will try to dive deeper into the discussion of which I barely scratched the surface today.

Enter tandem. Tandem is work in progress and a place where I experiment with concepts I will discuss on this blog. Check it out, let me know what you think and try to form an opinion on the areas where it successfully addresses the problems I mentioned above but more importantly also on those where it fails.

tandem — exploring the Flickr universe

Enjoy & stay tuned.

P.S. Thanks to Bryce Glass for his wonderful diagram.