written in a shitty language (SQL) I disagree with this. After doing a bunch of work with SQLAlchemy and Hibernate over the years, I've developed a major disli

written in a shitty language (SQL) I disagree with this. After doing a bunch of work with SQLAlchemy and Hibernate over the years, I've developed a major dislike to ORM's as a replacement for SQL and would much rather write my queries in plain SQL. My currently favourite SQL library is HugSQL[1], which lets you write your queries in SQL in .sql files and HugSQL will load them and provide functions which you can call to invoke the query. I find SQL quite solid for querying and updating data. Its what it was designed to do. Much more solid that the query API's I've seen in ORM's. However, more importantly, when performance matters, it is very difficult to properly understand performance characteristics of complex ORM queries (I often find myself making the ORM generate raw SQL, which I then manually review to determine what its doing (eg what indexes it uses, etc)) and I find it much easier to optimise SQL than ORM calls. My point isn't to badmouth ORMs. If you like them and they solve your problems, that is awesome. My point is that calling SQL a "shitty language" is purely subjective and your opinion and other people have different opinions. SQL is not an objectively shitty language. I think its a pretty good data query and manipulation language. Not perfect, but definitely not shitty. And no, I'm not a DBA. I'm a programmer who hates having to mess around with databases. It is for this reason that I want things to just stay out of my way so I can get my work done with minimal touching of the database. I've recently found myself forced to do things I never wanted to do (1. change data capture on a MySQL database; 2. query and table optimisation to make a database perform and scale better) and for #2, my life would have been a lot easier if I could have worked with raw SQL. [1] http://www.hugsql.org/ http://www.hugsql.org/