Nowadays, legacy systems are laughed at if they don't have a REST API. But retrofitting a REST API onto an existing application can be a time (and money) consuming and tearfully-boring project. Surely there are better uses of our code-weaving abilities? Perhaps things aren't as bleak as all that, though. In this talk, I'll explore various ways of using Python to make the job of creating REST APIs for existing systems less painful.

For years you've used the same language for custom applications, and now it's market position is waning. How do you assess a replacement? Wharton Computing developed an assessment framework called DevTAP which raises red flags if a platform is beginning to decline and evaluates whether the next choice meets your organization's needs. The framework was created because Wharton Computing found itself in the position of being heavily invested in a web platform no longer being improved or supported by the vendor. Many institutions face a similar dilemma.

The decisions made in the beginning of a project are the ones you are stuck with for years. A early bad decision for a project can cost organizations time, money, and engineering resources throughout it's lifetime. In this session we’ll attempt to answer the following questions: How do you take scale into consideration during the early stages of a project? How do you do this without using up the critical ramp-up time of a project or organization? Finally, how do you fix critical problems when they arise without bringing everything down?