For the average web developer, I find there is quite a bit of overhead and Amazon specific infrastructure for implementing a personal Alexa Skill. A much simpler approach might be to allow a user to enter a personal phrase associated with a custom URL in the Alexa App or something like IFTTT. When the phrase is spoken in one's Echo, the URL page would be just read, text to speech. It would read SSML. (If it's just plain HTML, it would ignore the unrecognized tags, but take into account items like <p>.) Different people can have the own personal trigger phrase to read from the same URL so there is no global namespace phrase collision. This approach would open Alexa development to a vast majority of people who already know CGI, HTML, ... and can learn to output SSML. It would address a large majority of application spaces where one wants to get a quick hit on information, e.g., latest organizational news, status of an ongoing IT operation, etc. Perhaps someday web sites can start having links like "bookmark to Alexa", and developing for Alexa would be as ubiquitous and easy and natural as ordinary web development. Thoughts?