Posts

Showing posts with the label tech choices

Tech Choices a StartUp Developer has to make

The eco system for programming languages and frameworks is well evolved - and is continously evolving. As a web developer you are faced with many choices. I have tried to capture some of the options I had to evaluate for Apartment Adda. I have left out the esoteric languages and frameworks which have not become mainstream yet. Programming Language - Java, PHP, Ruby, Python This is the first decision a developer has to make and might be the easiest choice - pick the one where you have the most expertise. However if you are well versed in more than one language - pick the one that will suit your startup - based on the funds you have ( cheapest to host will be php to Java hosting being expensive) , speed at which you want to release ( php/rails being fastest to Java being the slowest ), team's expertise ( java developers being easy to find and ruby developers being hard to find ) etc. Framework Choices - Rails, ActiveRecord, CakePHP, Spring, Hibernate, Django.. In most of the cases t...

You Used that Programming Language to Write What??

I stumbled upon this nice post about the tech choices we have. Resonates with my earlier post about PHP - how its good as a prototype bring it out fast language but needs to be re written in Java in the long run. http://www.cio.com/article/193750/You_Used_THAT_Programming_Language_to_Write_What_

Why PHP?

I have decided to build the online portal in PHP, supported by Apache, MySQL hosted on Linux. The most cliched LAMP model.  Did I consider other players? * Ruby on Rails : It was very much on my plate. I have done small utilities ( a server status monitor and a code snippet repository ) in rails and was blown away by the speed and simplicity of doing things. However 2 things went against RoR. ORM : My last project was a .NET project on Hibernate. It was a 2 year old project and already it felt like a legacy application with things tuned for performance to the maximum. They had crossed the return on investment period of Hibernate ( when relations stay simple at the beginning of any project ) and now were paying heavily for having Hibernate - performance, complexity, things moving to stored procedures..the usual pain points. So active Record did not interest me - being a one man army I did not want battles couple of months from now when I have to hack to implement a feature. Server supp...