Comments on: The Long Tail of the SDL and a Product Review http://securitybuddha.com/2007/08/15/the-long-tail-of-the-sdl-and-a-product-review/ Security Enlightenment - Mark Curphey Thu, 08 Jan 2009 13:18:50 +0000 http://wordpress.org/?v=MU hourly 1 By: MikeA http://securitybuddha.com/2007/08/15/the-long-tail-of-the-sdl-and-a-product-review/#comment-5382 MikeA Wed, 15 Aug 2007 16:20:27 +0000 http://securitybuddha.com/2007/08/15/the-long-tail-of-the-sdl-and-new-product/#comment-5382 I know that you gave a caveat up-front, but here's where the long tail doesn’t work, especially in product design. Coming out of the gate you have to design for the largest *common* audience. Sure, the largest audience overall may be in the long tail, but they also have wide and varied requirements, settings, dependencies, etc. When writing a v1 product (even more so in a start up environment), you need as much "bang for your buck" as you can. Adding lots of features to attract the long tail makes the product more complex to design, code, test, deliver and maintain. So here's my caveat as well :) Although a product has to focus its requirements/features on the short tail initially in order to limit scope to a reasonable and actionable level (no, "boiling the ocean" - one of your favorite sayings if I remember!), the product *must* be designed/architected so that it can be easily modified/upgraded/support plug-ins/etc to suit the long tail. Not necessarily straight away, but if it's not at least on the roadmap then the product is not scalable and "brittle" in such that a change in market conditions (or technology as in your example) can mean the end of the company. I know that you gave a caveat up-front, but here’s where the long tail doesn’t work, especially in product design.

Coming out of the gate you have to design for the largest *common* audience. Sure, the largest audience overall may be in the long tail, but they also have wide and varied requirements, settings, dependencies, etc. When writing a v1 product (even more so in a start up environment), you need as much “bang for your buck” as you can. Adding lots of features to attract the long tail makes the product more complex to design, code, test, deliver and maintain.

So here’s my caveat as well :) Although a product has to focus its requirements/features on the short tail initially in order to limit scope to a reasonable and actionable level (no, “boiling the ocean” - one of your favorite sayings if I remember!), the product *must* be designed/architected so that it can be easily modified/upgraded/support plug-ins/etc to suit the long tail. Not necessarily straight away, but if it’s not at least on the roadmap then the product is not scalable and “brittle” in such that a change in market conditions (or technology as in your example) can mean the end of the company.

]]>