Subscribe to Rahul's Blog        RSS Feed
-----

The devil's alternative

Icon 1 Comments
One of the great benefits of being and working in the software/IT world, is the amount of choices you have. Be it any field, from databases to search engines to programming languages to real time strategy games, the choices are always there. While having alternatives may seem to some like a twin edged sword, I personally believe its better to have them, based on the philosophy of competition driving quality. But the point I want to discuss is - what do you go for?

A nice example is - suppose you work for a large IT company and you're doing a database related project. What would you go for - MySQL or Oracle/SQL Server? Keeping in mind all the advantages and disadvantages of the respective packages like the cost of ownership, ease of use, scalability, sheer features, stress condition performance etc. Take another example, for the front end of the same, would you go in for a language like Ruby/Python which are widely regarded as top-quality by all experts but have smaller communities and lesser documentation OR something like ASP.NET/JSP which have stood the test of time? (Take out any bias you have about the company who makes these technologies) Remember that the project is your job, not your average experiment of setting up a personal blog for yourself.

Or you can take this further on another level of non-geek, would you rather use Google since its widely regarded as the best or a meta-search which would include Google and give you the best of breed results? Or even support a new kid on the block like Snap or Hakia?

I'll tell you my take on this, I choose what really works in the situation (like yeah, that's common sense punk, tell me something new!), but being a slightly geeky sorta guy I don't mind toying around with slightly newer technologies. I *do* use Ruby a lot even though I admit its a pain to get some good documentation on it in the form of a locally available book and I use MySQL a lot. I would consider Sybase or Oracle if I were to build a very heavily loaded database server, but till then MySQL will rule for me.

So the question boils down to - (apart from some common sense situations) do you go for the alternative or the tried and tested thing? Feel free to extend the example into any situation from email providers to ISP's to tech news sites.

1 Comments On This Entry

Page 1 of 1

paddy3118 Icon

28 May 2007 - 11:43 AM
Well,
first know your problem. If there still exists a tried and tested methodology that would work to time/budget/quality constraints then its a 'no-brainer' USE IT!
If the problem doesn't fit your previously used toolset, then I would document the deficiences as well as what would be covered by existing methods then seek out alternative solutions having this current problems solution in mind. - You may still choose to use the old methods, but only after evaluating them against the competition.

- Paddy.
0
Page 1 of 1

Trackbacks for this entry [ Trackback URL ]

There are no Trackbacks for this entry

Search My Blog

0 user(s) viewing

0 Guests
0 member(s)
0 anonymous member(s)