Alex Iskold at the ReadWrite web posted his thoughts on Dapper, the web service that lets users create data services from other websites. I’ve tried Dapper and it is pretty cool. In addition to Dapper I’ve looked pretty carefully at a wide entire range of services that, broadly speaking, are mashup and application development platforms including Pipes, DabbleDB, Teqlo, Coghead, and even Salesforce.com’s AppExchange.
With Microsoft’s new Astoria project it’s pretty clear that a lot of people are paying attention to how to leverage the web as a new development platform.
Given the wide range of capabilities and potential uses for these kinds of services I’ve found it useful to organize them by their target user: Are they developer focused, or end-user focused. For some it’s clear, for others its not. Astoria and AppExchange seem to be developer focused, but these others could be either developer or end user focused. Its not obvious to me where the firms are targeting them. Some seem to target end users, but require skills more advanced than 99.99% of actual end users. I’d say that DabbleDB is end user, and maybe Teqlo as well. But I’m not ready to predict where these other services might gain adoption.
It’s also useful to see where each of these solutions lie in the stack. Some clearly sit in a single layer: AppExchange and Coghead are clearly application solutions. SnapLogic and Astoria clearly in the data layer (some might argue network as well, but let’s not split hairs just now). DabbleDB is kinda, sorta in the data layer, but its an app as well. Teqlo seems to traverse a number of layers as well and might even be looking up further into Orchestration, but I’m not entirely sure of that. Dapper and Pipes don’t seem to be firmly in either the data or app layer, which is why people can agree that they’re cool, but might have trouble shaking the ‘toy‘ moniker.
Call me old fashion, but whenever something traverses too many layers I’m skeptical the approach is going really gain widespread adoption. Give me the best of each and I’ll put them together.
I need to think some more about how these services as well as the raft of open source projects that are swirling around this space can be organized so that they’re easier to understand. Let me know if you have any ideas.