Lone Star PHP 2017 Wrap-Up

This past weekend I had the opportunity to attend Lone Star PHP in Addison, Tx. Even though this was the last running year of the conference, the quality of speakers and sessions was not lessened.

Here’s a wrap-up on my top 5 favorite talks of the weekend:

Your API is Bad And You Should Feel Bad

In this talk, Ben Edmunds (great speaker!) discussed best practices when building a RESTful API. Edmunds discussed when designing an API, you should keep it simple, expressive, stable, and consistent. These guidelines are especially important when you are working to build up your user adoption. A stable API is one that continues to work the same even with updates. A consistent API should have a similar POST for every GET endpoint. Keep your api intuitive by naming endpoints relevantly and knowing when to use singularity vs collections (ie a single user vs users).

Discovering Dark Knowledge in the Social Web

This talk was unique in that it started off especially relevant to the advertising world but quickly changed to expose the *creepy* amount of public data that is available via social media. The speaker, Keith Casey, introduced us to the term ‘affinity groups’ which are concentric circles of people who share a common interest. Social media creates a vast world of affinity groups ranging in granularity. Meetups provide very tight-knit affinity groups because these members usually have strong similar interests and advocates. Facebook on the other hand, usually cultivates less cohesive groups that have less-involved members. These affinity groups are key factors when identifying market segmentations and audiences. However, since most of this data is publicly accessible via social platform API’s it can be easily misused…

Projects are not Products

This was another session by Keith Casey where he discussed the difference between projects and products and what we should care about when building them out. As developers, it’s easy for us to focus on shiny new features and having the best-looking code base. But, this isn’t what sells the product. We need to make sure we spend enough time actually solving the problem and who the users of the product will be. Long-term goals are awesome and good selling points, but this should come *after* an initial, working version is released. Long story short- the outstanding code isn’t very important, solving the problem is the most valuable.

How I Learned to Stop Caring and Made Better Software

This talk was similar to the previous one and focused on a common development issue: spending too much time on processes and best practices while sacrificing quick feature turn-around. In many startups and startup-like cultures, functionality is valued over perfection. Fail hard and fast so that mistakes can be cleaned up while they’re young. Deal with problems as they come, don’t spend too much time trying to predict them prematurely.

Groupies, Roadies, and Rockstars

This was the keynote session of day 2, given by Cal Evans. He opened up a world of thoughts for me as he discussed the many ways you can contribute to open source software. There are 3 big roles in open source: groupies, roadies, and rockstars. Groupies are the people that don’t contribute much code-wise but get others excited about the project. Roadies are similar but often have an end-goal to be on the main stage so they pour a little more effort into code production. Rockstars are those who have never been an overnight success but continue to perform (code) because it’s their passion. Moral of the story: be all 3 roles and contribute to open source!

–

All in all, this conference taught me a lot about the PHP world- everything from best practices to communication. As an unbiased developer at PMG, I can honestly say we do an amazing job at following best practices while having quick turnaround time. Although Lone Star has said goodbye for good, I look forward to other related conferences in the future!