What are the reasons behind?

What are the reasons behind creating Hack? What are its perspectives?

• What are the reasons behind creating Hack, a programming language based on PHP (http://venturebeat.com/2014/03/20/facebook-unveils-hack-a-faster-programming-language-to-power-the-social-network/)? Why not just improve PHP? What do PHP developers and PHP creators think about Hack's perspectives beyond the Facebook platform, considering PHP's dominant role in powering today's Internet web sites? How Hack compares with the "new wave" dynamic programming languages, such as http://julialang.org/ (beyond Julia's focus on technical computing vs. general computing)?

• Answer:

  Facebook was originally written in PHP because it was one of the most popular languages at the time for throwing together web applications quickly and it was easy to pick up. It is surely many many times larger than that original implementation but once you write a whole bunch of code it's pretty hard to go back. Languages well suited for throwing things together quickly by small teams generally aren't the same languages that work well for large codebases that have to be fast, high quality, and maintained by large groups of people. Just as most Linux software is written in C, so will Facebook always be written in PHP. As a result, they've had to make do with what they've got. They wrote a native compiler to speed things up and now they're created a new superset of the language which brings in some type safety, to reduce the number of bugs that hit customers (at the cost of productivity). Facebook doesn't control PHP development and no doubt there are many users still out there for it who would not benefit from these changes. Thus, they create a new language and OS it. Who would be interested in using this? Someone else who has a large bundle of PHP but wants to make their application slightly more robust. I don't know if PHP creators lurk on Quora but I'd guess that they're pleased to hear about it. It helps adoption of PHP, as it gives a path to migrate away from it. I don't know if there's too many people who are still starting projects with PHP anymore but it's one less thing for them to worry about should they consider going down that route. Dominate is probably not the most accurate term to use for PHP but I guess it depends on how you count. The TIOBE index places it at #6 right now, behind Java and C#. I don't know of any indices that track only new projects but I would guess that in terms of number of starts, I would guess that Java, Python, Ruby, and maybe even Groovy probably all beat out PHP at this point. But, it'll be a long slow death much like other once very popular languages (COBOL, Pascal, FORTRAN, etc).