Archive for January, 2007

Dapper Goes to Mashup Camp

January 18, 2007

MIT, Cambridge, MA

Mashup camp is being held yesterday and today at the hotel@MIT in Cambridge, MA, and I must tell you, it’s a really cool event. First off, kudos to Doug and David. They managed to pull off a great, well organized, event that at the same time remains informal and energetic. Especially given the registration price (you guessed it: 0). Getting here was quite an experience, where as usual the good folks at Newark airport immigration managed to work slowly enough for us to miss the connection, but once we got here it was one great ride. We’ve met some really smart and creative people, with really clever mashup ideas and a very clear understanding where the web is going and how services like Dapper can help the community get there. We decided to host two sessions.

The first session discussed the notion of authentication based APIs. How do we create APIs for content that is behind a login and password? As someone said, most of the world’s interesting content is behind a login wall. So how do we go about building APIs for these content sources? We naturally featured the Dapper solution, but in a broader context, tried to open up for the discussions the challenging issues which arise in such scenarios. One of the interesting questions that came up pertains to the issue of trust. If I’m a man-in-the-middle content aggregator/proxy between the users and the content source, how do I build my system so that my users will trust their login credentials with me? A solution that may work for my user’s Facebook data may not be acceptable for creating an RSS feed from his bank statement. So obviously there are (as always in security questions) different layers of security. I think the discussion was very interesting and stimulating, and we got some good responses when we described our solution to this problem, with our credential encryption web service. We also learned that recent attempts to solve the issues of trust and security by the bigger players (like Yahoo and Google) come laden with problems that make it hard to automate and program many desirable tasks. It’s clear that trust will always be an issue for users, and we’ll continue to address that and work to gain the trust of those who use Dapper.

After lunch there was a session called “speed geeking”. Yes, it’s what you think it is. There were about twenty guys sitting and presenting their mashup for like 5 minutes at a time, until the buzzer rang on and everyone moved to the next mashup. There were some very cool ideas floating around, from full fledged applications and infrastructures to cool impromptu mashups. A few that stand out of the crowd:

Proto, with their desktop mashup builder, have presented a really compelling case for desktop apps, in a world where “desktop app” is such a a no no. I recommend you check them out.

MyBlogLogTagger is a cool little widget/mashup thingy from Kent Brewster that allows you basically to tag… people. Now that’s a simple yet very capturing idea. And the use of Mybloglog and as the jumpstarters of this game actually make it feasible.

Hype Machine and Tour Filter are two music related mashups that provide you with an immediate value if you’re an avid music enthusiast. One of the nice things about them that though both are mashups, Tour Filter is actually a mashup that is party based on Hype Machine, so it’s a second generation mashup. And so the chain begins. Both clearly involved a substantial amount of work and their creators should be proud.

Later we had another session that was focused mainly on the eco-system of content providers / content consumers and the chicken and the egg problem Dapper is trying to solve – namely, how do we get both in the game? How do we empower mashup and widget creators while at the same time remain respectful of content providers’ rights and transform this no-man’s land of cease and desist letters into a sustainable market? We’d recommend you keep a watchful eye for Dapper releases soon to hear about our take on the matter. What’s for sure is that we need to pursue a direction as a community. To build a market is larger than just one service, company or solution. It requires approaching content providers with clear offerings that they can relate to, while at the same time listening carefully to what they regard as important. All in all, let’s be respectful as we’re overhauling their world 🙂

We’re looking forward to another full day of camp. For those who are at the conference who we haven’t had a chance to talk to, let’s try to connect today and share ideas.

Finally, here is Jon leading one of the discussions:

jon in mashupcamp

The Dapper Developer Community

January 10, 2007

Today we’re happy to launch the Dapper Developer Community (DDC). The Dapper Developer Community is all about empowering you and enabling you to share with and inspire other users. We hope to foster a rich and vibrant community that you can leverage to make your life easier as a developer. We’re jump-starting the DDC with three important elements: the Dapper SDKs, the Dapper web services and the Dapper Transformer Library.

The Dapper SDKs are already being used by many of you to easily integrate with Dapper, and we’d like to encourage you to go ahead and give them a try, or even go and port the SDKs to a language of your choice. In the future, using the SDKs will allow you to enjoy improved functionality such as higher rate limits, faster performance etc.

The Dapper web services are sort of a must, if we don’t want to be considered “the shoemaker that walks barefoot.” But it also already provides important functionality. The first web service, already released, is the Dapp Search Web Service. It allows you to programmatically find Dapps, just as you would’ve done through the Dapper website user interface. With it, you can allow your users to search for Dapps without ever losing your site. The Dapp Search Web Service is the first in a series of web services that will be launched in the future.

Lastly, and most importantly, we’re making available the Dapper Transformer Library. If you’ve ever used a Dapp, you’ve seen on the Dapp’s page the list of ways by which you can use it, which we call transformers, such as RSS, Image Loop, Netvibes module, Google map and gadget, and several more. But not enough. Today we’re opening up the code behind these transformers and provide you with an API by which you can build your own transformers, and share them with fellow Dappers. If you’re into PHP, you’ll find the task of building a new transformer quite painless, and since we’re also releasing the code of our existing ones, you can go ahead and build on them. So if you always wanted to create an embeddable flash module for weather updates or get your bank quotes directly into a SQL database, you can build new transformers that will cater to your needs. We are, of course, available 24/7 to support you with any question you may have.

CSV Transformer

January 4, 2007

A long standing request from many users has been the ability to get data from all over the web in a format easily imported into spreadsheets. We’re happy to announce that we’ve released a CSV (comma separated value) transformer that allows you to take the output of any Dapp and import it into any spreadsheet, like Excel or Swivel. Let the data flow, and as always, let us know if you have any trouble or additional feature requests.