In the past I’ve never had a say in what was the most appropriate script to use to build something. This is mainly because other people have been far better qualified than I to make the decision but also because it’s never really been in my interest to know. So long as it worked and adhered to the ‘experience’, the features and functionality we specced then like, whatever. Recently though I’ve been developing an idea to form the basis of a service and it has been in my interest to know [because I'm carrying the risk] mainly because it’s near impossible to get an ‘agnostic’ view of the most appropriate language in which to develop. Key factors for me:
- development time
- transferability [i.e. is this stable? how easy is it to maintain? is the language well known or is it like Latin and dying out fast? or even worse like some street vernacular that isn't even documented properly and unknown by all but a handful of very well paid scruffs ]
Yet trying to get a straightforward answer around such criteria without actually doing a full blown tech scope and incurring the cost of that, was and is tough. Developers and technical project managers are steeped in what they know and that can become obsolete or partial very quickly as languages adapt and change. It’s a big issue. I know the BBC had there architecture pinned down early, were heavily dependent on Perl, all for good reasons, but now find it hard to adapt to web2.0 functionality increasingly expected of the younger audiences they covet.
Hence I’ve had my head in lots of primers over the last 4 months. Trying to get up-to-speed with languages like PHP, Python, Ruby etc. It’s helped but it’s not the best use of my time and there has to be a market for people like me looking for a good overview of the foundations of web development both client and server side. Maybe there already is, but I don’t see it. This is one of the first ‘lay’ views of the language of web 2.0 and their ‘facets’:
Dev time seems to be a straightforward trade-off with scalability and transferability which makes languages like Ruby great for start-ups looking at proof of concept and limited scale. Presuming once they’ve proven the business model they can invest in re-developing the application or service, underpinning the house and putting the scaffolding up. I do find it kind of ironic though that in the search for ‘purity’ amongst developers [and they all talk about how 'pure', 'clean' etc. and use of such metaphors is in itself really interesting cf. Mary Douglas' work] the architecture of the web is actually pretty messy despite or perhaps because of the search for such ‘purity’. And to say that much of our future could depend on that architecture I’m surprised there’s so little discussion of it beyond enterprise level as bodies like W3C concern themselves more with accessibility standards than building regs.
Link: PHP vs Java vs Ruby.