Choose back end programming language / programming model

[fabianmuehlboeck]

Requirements:

• Language should be well-supported
• A good number of team members should have basic knowledge of the language
• Language should support web-programming well
• Must work on our server (@fabianmuehlboeck will determine this case-by-case, but the general platform is Linux)

Languages that support web-programming have at least one programming model for web pages. This may have a huge influence on several things, for example how sessions work, or what databases are supported well.

[bozhou94]

So I've looked into some of the commonly used languages and came up with these:

PHP

Pros

• Easy to learn
• Aimed to be a server language to begin with, so a lot of helpful features built in
• Was the standard for a while, so there's a ton of support and libraries

Cons

• Hard to pick up where others left off
• Built in sessions can't store resource variables e.g. database connections
• Easy to create an insecure application need to be aware of pitfalls

Sessions Libraries

• PHP Built in Sessions
• Zebra_Session
• uLogin

Databases

PHP supports a variety of databases, including Mongo, MySQL, PostgreSQL, and SQLite. The full list can be found here.

Java

Pros

• Easier to manage larger projects
• Used to be very popular, meaning a lot of libraries lying around
• Good built-in concurrency management
• Good built-in garbage collection

Cons

• Complicated to navigate through for smaller, more informal projects
• Currently diminishing in popularity

Sessions Libraries/Frameworks

• Apache Shiro
• Play

Databases

• JavaDB
• Any database that provides a JDBC driver. A list can be found here.

Node.js (JavaScript)

Pros

• Trending upwards in popularity, meaning a more active community
• More supportive of end-to-end development, allowing for more flexibility in switching roles

Cons

• Not suited for CPU-intensive tasks - single threaded by default
• Relatively recent in comparison

Sessions Libraries/Frameworks

• Stormpath
• ExpressJS
• PassportJS

Databases

• SequelizeJS
• MongoDB
• Other modules

Python

Pros

• Fast to develop and debug in

Cons

• Lack of concurrency mechanisms

Sessions Libraries/Frameworks

• Django
• Flask

Databases

Django supports MySQL, PostgreSQL, and SQLite, among other databases. The full list can be found here.

Ruby

Pros

• Easier documentation
• Good testing frameworks

Cons

• Larger projects become harder to manage

Sessions Libraries/Frameworks

• Lotus
• Sinatra
• Rails

Databases

• Rails supports MySQL, PostgreSQL, and SQLite out of the box.

Go

Pros

• Good concurrency management and garbage collection
• Fast in compilation and running
• Good documentation support
• Built in testing framework

Cons

• Incomplete testing framework
• Takes some initial time to pick up due to having to understand arrays vs. slices
• No method overloading
• No support for generic types
• Relatively recent in comparison

Sessions Libraries/Frameworks

• Gorilla

Databases

• SQL and SQL-like databases

[fabianmuehlboeck]

All Hail Python and Django