Dabble DB

The Dabble Blog

Archives: January 2007

« December 2006 · February 2007 »

Contributed Plugins

Just one day after we announced the Plugin API, Tim Fletcher has posted two new plugins for Dabble users to try. Check it out.

Update: also worth pointing out is Chad Fowler’s post on the somewhat unusual “don’t call us, we’ll call you” approach we’re taking to web services.

Announcing the Dabble DB Plugin API

Have you ever wanted to sort your competitors by their search engine rank? Look up the current retail prices of your Christmas wish list items? Group your contacts by the estimated cost of their homes?

Data loves company, and your private data can be so much more valuable when it’s combined with information pulled from the web. That’s why we’re announcing a new way for anyone to extend Dabble DB to do exactly that.

Seth Godin calls this kind of thing a websheet. We call it the Dabble DB Plugin API, and you can read more about it in Chad Fowler’s excellent tutorial. There’s plenty of info there on how it all works, but the key things I want to emphasize are:

  • It’s open. Anyone can write, host, and share a plugin on their own web server using whatever language and platform they want, without any fancy libraries or XML dialects needed.
  • It’s integrated. Once a user’s installed a plugin, all of that plugin’s functions will show up inside the Dabble UI just like any built-in function, and any new fields or data that the plugin produces will look and act just like anything else within Dabble.
  • It’s simple. We’ve modelled plugins after the tried and true Unix pipe-and-filter model of data manipulation. Dabble DB passes data in as plain text (in the form of a CSV file), and the plugin passes the transformed data back in the same format. That’s it.
  • It’s flexible. Maybe you don’t need to pull in data from the web, but just need to do some custom statistical calculations. Great. Maybe you want to do some crazy conditional formatting or generate Tufte-style sparklines. Perfect. Anything that involves taking one piece of data and returning another, whether it’s text, dates, times, durations, money, or images, is something a plugin can handle.
  • Did I mention it’s open? Seth Godin points out that although there is a tool with some basic websheet functionality, “it needs to be open so others can add to it”. This API is dead simple to implement; if any other vendors want to support these same plugins, we’d be thrilled.

Oh, and for Dabble DB users who just want a sample plugin to play around with, here’s one with pretty universal appeal: http://dabbledb.com/plugins/currency will perform some basic currency conversion for you using the daily rates from the Bank of Canada. But we don’t really want to be writing plugins ourselves. What we’re really excited about is seeing what the community does: since anyone can easily host a plugin at a cheap webhost - with or without distributing the source, with or without charging for access, with or without our approval or knowledge - the door’s wide open for developers. Come on in.


It’s been a couple of months since I last shared customer feedback, so here’s some more. I received the email below from John Felt, the CIO of the Advocacy Center for Louisiana. I’m reproducing it with his permission:

Let me commend you folks for creating a really impressive platform. I have already created several working applications and have passed a couple of more down to one of my employees to manage. I did all of the evaluation I need, and am now ready to take the plunge - not just with a purchase but with a migration of all our data (mostly spreadsheets but some Filemaker, SQL Server, Postgres and MySQL databases) to Dabble.

As a database guy I can’t praise y’all enough. The thing Dabble gives me is TIME. I have no trouble creating and managing SQL databases, that sort of thing is easy and fun for me… what’s NOT easy and fun is UI and basic CRUD functionality rolled out to a non-technical user base. I have found with Dabble that my time to implementation has gone down from a week with a typical Postgres/MySQL database with a web front end to a few hours with Dabble. Having a consistent look and feel for my applications and the freedom of not having to create a new UI for every app gives me the ability to roll out more actual apps to my users so they can get things done. As a sidebar I would also say that while I know you are marketing your application toward the spreadsheet user, Dabble DB is an INCREDIBLE tool for data modeling. If I found something I needed to do in MS SQL or Postgres for some reason I would now model the data in Dabble first and then, once I was comfortable with my tables, export out and back into SQL.

Thanks, John, and best of luck with your migration.

« December 2006 · February 2007 »