« Paul Theroux | Main | Shit Comes In Waves »

November 24, 2007

Dopplr Case Study From Building Social Applications Workshop

At the recent Web 2.0 Expo in Berlin I presented a revamped version of the Building Social Applications workshop that I had previously given at Lift in Geneva and the Web 2.0 Expo in San Francisco. I have touched on parts of the workshop in other posts (here and here), but this post focusses specifically on the case study and group exercise.

[I guess I have been flapping my mouth too much in Europe... starting to be strange when I am used as the example of what a conference is not going to be:

[from European Tech Tour: Web and Communities Event in Montreux by Fred Destin]

ETT is not meant to be leWeb3. Or LIFT. You won't meet Stowe Boyd doing a podcast about lifestreaming :-). It is a venture focused conference designed to put venture capitalists and local entrepreneurs in touch, and it did that extremely well.

The Case Study

So, the case study walks through Dopplr, a social travel web application. Dopplr is the perfect sort of beast for this kind of dissection, because it doesn't do very much -- at least not yet. It's like a microbe with only 27 genes.

We walked through the entire app, and then -- applying the tools that we discussed earlier in the workshop (see Theory And Practice Of Conceptual Design) -- we broke into groups, acting more or less like design consultants, and formulating recommendations for Dopplr.

I love the pixelated picture that Dopplr uses when someone who doesn't have access to your info (or hasn't logged in) looks at your Dopplr page.

Slide21



Once you login to Dopplr, you see your 'trips' page. This is shown in the figure below in the right hand side. Because the page has so much on it, I have zoomed in on just the top part in this slide.

Note that the page doesn't have a title on it, so it is possible get fall into a 'where am I?' mode in the app once you wander around a bit. There is no obvious navigational cue that you are even on the 'trips' page.

The top of the page has a number of major navigational titles, which we will return to. Then there is a list of upcoming trips in the 'Where Next?' section, which has an 'add a trip' controller. Just below that is a search area, that displays its purpose only by the "Type the name of a city or a traveller" text.

Below these areas is the profile area, with photo, home city, and other information. My feeling is that this information would be more senibly displayed at the top of the page, above the 'Where Next?' and search box.

This region of the page seems a bit confused:

  1. Why are upcoming trips not part of the other profile information, like home city or where I am scheduled to be today?
  2. I would make search more obviously a navigational tool, like 'manage connections' or 'your trips'.

Slide22



The middle section of the 'trips' page has the actual list of upcoming trips, displayed in list (here) or map (not shown). The various trips can be edited or deleted. Adding a new trip is (strangely) buried at the bottom of the list, and this is also a duplicate of the 'add a trip' functionality of the "Where's Next' list at the top. On reflection the 'Where's Next?' and the trips list are actually two views on the same thing, which is perhaps how they should be set up, instead of two things on the same page that do the same thing.

There is a 'subscribe to your trips in iCal' lurking at the bottom of the bottom of the list, too.



At the bottom of the 'trips' page they is a mosaic with the avatars of all my friends. This should be on the 'trips' page? There is a 'Where Next for them?' (badly capitalized) controller with associated RSS feed: note that there is no corresponding RSS feed for the 'Where Next?' capability.

The bottom of the page also has a number of navigators to various Dopplr related pages, like FAQ, Terms of Service, and About.



When you add a trip, the 'add a trip' window opens. Pretty straightforward, although the 'add a note' falls short of writing blog entries. And it's not at all evident, here, that the notes are accessible through the 'journal' tab. Shouldn't it be 'add a journal entry'?



When you click on a trip, you see the controllers at the upper right that allow you to 'edit this trip' and 'delete trip' [shouldn't it be 'delete this trip' to be symmetrical?]

What you quickly realize is that a 'trip' is actually a pastiche of your trip and the city where you are going. You see who else will be there (Berlin, in this case, during the Web 2.0 Expo a few weeks ago). I love the term 'coincidences' to represnt who you overlap with. And the little seismograph that indicates 'coincidensity' -- a term that Matt Biddulph of Dopplr coined.

At the top under the name of the city you see the 'notes' that you have associated with the trip. Are they linked to the trip or the city? Both, I guess, but it still seems a bit confusing.

There is a place where -- if you set it up -- pictures you take and post (to Flickr and other sites) will be pulled automatically, which is cool. But they are not part of the journal entries?



The socialistic 'fellow travelers' tab opens the travels (if any) of your various friends.

One of the fuzzy areas in Dopplr is that travels do not have hard start or end times, so you don't know if your buddy will be available for lunch on his first day in town, or if she is leaving too early to get a last breakfast.



I find the map view useless, because (I think) it shows where they live, not where they are, which is contrary to the whole point of Dopplr.



Here's a 'note' (better would be 'entry') in my journal. A lot of areas for improvement here.



If you click on the 'Your Account' navigator at the top of the page, you get this page which lists all sorts of things you can do. Let's touch on some of them.



I dislike the notion that you can only have one home city. I have two. This is just like the same stupid restriction in Facebook, where you can only belong to one local network.



There are a number of notification options.



Adding trips by text message? Maybe, especially if you are on the run. Haven't used it myself. Might be better to set it up to get notified when new coincidences occur, like a friend decided to come to town today.



There is a way to associate an OpenID with your account. Haven't tried it.



Managing the visibility within Dopplr is interesting. There is an interesting asymmetry allowed. If I let you see my trips, that does not necessarily mean that I get access to your trips. You could decide to reciprocate, or not. This asymmetry is just like that we find in streaming applications like Twitter, where you could be followed by many more than those you follow. There is an upstream and downstream asymmetry. However, with the exception of RSS, iCal, and various email and SMS notifications there isn't anything that feels like a stream in Dopplr.

Here we see the list of travelers whose trips I can see. The title might be a bit misleading, because you don't really have control on the visibility of individual trips, as the tab might suggest.

There is a corresponding 'who can see your trips'.



New people are joining Dopplr all the time, and you might know them (or the friend that invited them), so the 'New travellers' tab displayes them, so you can hook up.



I really like the 'Who you might know' tab, which guessed correctly that I might know Jyri Engestrom, Dave Sifry, Tim O'Reilly and Catarina Fake.



I can invite people to join Dopplr; in fact, I have infinite invitations.



Dopplr has a pretty neat integration with Facebook. This shows the main canvas there.



This shows the Friends canvas. Basically these allow me to access all the most critical info from Dopplr without leaving Facebook.



I subscribe to my Dopplr in iCal on my Mac. I had already been using a single iCal (and Google) calendar for the cities I was going to be in, so now I just enter that info in Dopplr, once, and subscribe to it in iCal. (I use Spanning Sync to sync between Google Calendar and iCal.) At any rate, as long as I am willing to open Dopplr to create and edit trips, all works in a reasonable fashion, although it would be nice to be able to make changes in iCal and sync back to Dopplr.



RSS sort of works, although the dates aren't shown, which makes it useless. Why don't they order by date? Shouldn't the 'notes' be included?



Here's the interface to get pictures from Flickr pulled into trips.



And a recent trip -- Tel Aviv -- with some photos from Flickr.



[At that point, I broke the group into something like 8 or 10 groups, and suggested that they try to use some of the techniques we discussed in the earlier part of the workshop (see Theory And Practice Of Conceptual Design), and try to apply those ideas to the next hypothetical version of Dopplr. It was interesting that approximately 30% of group just opted to read email, surf the web, or wander out of the room. A cultural difference?

Each of the groups dubbed someone to be the spokesperson, and to make recommendations. There were a lot of good ones.]



My first observations are based on making Dopplr a preeminent place to get high quality advice and recommendations form heavy-duty travelers. To this they need to beef up and rethink the entire 'notes' and 'journal' angle. In particular it should be more like a blog, and probably should supplant what is currently being displayed on each person's profile. They also need to incorporate tagging, and perhaps some sort of karma system, so that people can determine whose advice is worth taking.

Given a real focus here, there is a path to money: advice and recommendations on restaurants, hotels, and so on would lead to possible revenue from reservations and/or advertising. Remember, the whole point is to have an application that makes money.



The nest observation has to do with the basic purpose of Dopplr: hooking up with friends once you are in the same place. Except that Doppl does not actually allow you to invite people to have dinner: in fact, there is no messaging in the system at all. No events.

The thrust here would be to compete (or integrate) with existing invitation/event services. My suggestion would be to implement something very basic, which might be sufficient to accelerate interest in Dopplr, and then see if Yahoo or Google wants to buy.



The traveler is traveling, and needs to get acommodations. Air travel, train, hotels. Integration or competition with existing services is an obvious need and/or direction.



Where's the streams? I would like to see more of a streaming model, where I would be getting updates on my fellow travelers, their notes, their recommendations, and so on.

I have mentioned the notion of a common service for applications -- a shared stream architecture -- and perhaps Dopplr could get together with folks like Twitter and Facebook on that?



I would really like to see a finer grained geography in Dopplr. I am not just staying in London, I am in Shoreditch; I am not in San Francisco, I am in SOMA.

Also, when I am visiting Geneva, I am interested who is in Lausanne, with is only 30 minutes away by train. So, larger and finer grained notions of 'locale' are needed.

One of the most direct competitors to Dopplr is TripIt (see TripIt), which does a great job of importing email itineraries from the various airlines and travel services, and automatically generating travel portfolios. Dopplr could at least allow me to capture the time of my flights.



I have already mentioned the need for finer grained notions of time: what time someone is landing at Heathrow or Oakland, for example.

There is also the need for larger grained notions of time relative to travel: when I plan a trip to Europe, for example, I might visit three or four cities. In my mind, it's all one big trip, with various segments.



Conclusions

Dopplr has a lot of potential, and many potential paths. It's obvious that they can't stay where they are: they have to do something that makes money, and they need to stay away from services that will become commodities in the near term.

I don't think they have exploited all the touch points surround people's interactions around travel. In particular, the coordination of travel -- what days are good, based on the schedules of the people you are trying to visit -- is a thorny, fuzzy area. At the very least, exploring the many unexplored touchpoints, like inviting people to dinner, would be smart, and most likely necessary for long-term success.

My guess is that the notion of premium intelligence from the business travel elite is a winning plan, and could lead to a clear and defendable niche, supported by advertising and perhaps various premium (for fee) services.

We'll have to see.

I enjoyed the workshop. Now that I have written it up, though, I guess I will have to use a different guinea pig the next time I do it.

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c50ba53ef00e54f9ff6238834

Listed below are links to weblogs that reference Dopplr Case Study From Building Social Applications Workshop:

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Hi Stowe

Thanks for a great post. You really got it down. I'm just wondering, and I've been finding this annoying to the point of not using dopplr all that much: Have you found a way to show travel from home -> x -> y -> home? Quite often I visit two places before returning home and that means there's a whole day where it says on dopplr (and facebook) "Olof is returning home today".

Kind regards.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Comments are moderated, and will not appear until the author has approved them.