I was talking to Abhishek and Amandeep of loremate few weeks back when they were in Bangalore about the issues which they faced (and still facing) because of wrong choice of platform (and the wrong approach their developer had), Rabi/Ashish of idubba also had similar experience with a freelance developer which made me think how important it is to chose the right development platform for a startup as a wrong choice will not only make you lose precious 2 to 4 months, will also void all the time and resources invested in earlier development.
WorkMonk is built purely on PHP/Mysql and though I did thought of considering Django, CodeIgniter (a huge $100M+ site in our domain uses CI) and Yii (everyone was talking about it), eventually I decided to stick with php/mysql.
So to Summarize it from my experience, if you are looking to chose a development platform for your product/service startup, ask yourself these 6 questions -
1. How fast you can develop on it ? – you can’t spend months understanding and learning the language and platform. On web things move fast and things keep on getting outdated in every 18 months.
2. How comfortable you are (and your current core developers) with the development platform ? – if you and your developers can’t code it, don’t expect to hire someone to do it for you. It has never worked and will never work.
3. How mature the development platform is ? – Majority of the Open source project and platform offer basic features for you to build on but after that you are pretty much on your own.
4. How much code/examples are publicly available ? for rapid development, you need lot of code/modules/libraries written by others already
5. How easily you can find 5 developer on that platform in India ? – considering the best developers are already running their own startups, already working for a startup or are havng 12+ lakh jobs. Also remember, no good developer wants to work in a startup which has nothing to show.
6. How easily you can find the help on web/forums/IRC ? – again, you don’t want to reinvent the wheel and solve platform problems which should anyway be solved by core platform developers.
also it is a good idea to discuss this with as many people as possible, specially people who are using the platforms which you are looking to use. One 10 minute conversation can save months later.