Comments on: Is there REALLY any rigor in Waterfall?

By: René Johnsen

René Johnsen — Thu, 02 Feb 2012 09:25:08 +0000

Great post!

By: briefcases

briefcases — Fri, 30 Oct 2009 09:03:12 +0000

I am impressed, you did a great job with this post, thank you and keep coming with fantastic posts like this.

Dave

By: CleverWorkarounds » Why do SharePoint Projects Fail? - Part 8

CleverWorkarounds » Why do SharePoint Projects Fail? - Part 8 — Tue, 10 Jun 2008 13:03:20 +0000

[...] post entitled "Observations on the Rigor of Waterfall", and referred to a post from David Christiansen. Both authors do a great job in systematically pulling apart some of the misconceptions of the [...]

By: Followup: Rigor of Waterfall

Followup: Rigor of Waterfall — Tue, 06 Nov 2007 01:49:28 +0000

[...] now and then I write an article that gets a lot of attention for whatever reason. My post, “Is there REALLY any rigor in waterfall,” was one of those. Sure, it wasn’t as popular as “The Rules of Office [...]

By: David Christiansen

David Christiansen — Tue, 30 Oct 2007 22:33:56 +0000

Thanks Ben. I agree – in an industry marked by so much innovation, we seem to have barely moved at all in this area. That’s ironic.

By: Ben Simo

Ben Simo — Tue, 30 Oct 2007 18:34:57 +0000

Great post!

It is sad that software development philosophies and practices developed in a world of government regulation, punch cards, and very expensive computer time still have such a strong a hold on today’s commercial software development.

Ben Simo
http://QuestioningSoftware.com

By: David Christiansen

David Christiansen — Tue, 16 Oct 2007 18:00:56 +0000

Thanks Chris for a great comment. I learned something new from my own blog!

By: Chris R. Chapman

Chris R. Chapman — Tue, 16 Oct 2007 17:52:16 +0000

Great post – it’s good to see developers actually go back and do some fact-checking to formulate an informed opinion on things like BDUF/waterfall.

I have a couple of points to add to what you’ve mentioned on Royce above with respect to context:

* Royce developed the phased delivery model to cope with regulatory requirements set out in the US DoD STD-2167 document, which was so byzantine and bureaucratic that the waterfall was the only way to cope with it;

* A common misconception is that Royce “invented” the single-pass, phased model – actually, in his paper he recommended that the phased model _be repeated cyclically_

* Even the DoD realized 2167 was a dud and abandoned it in 1994 with MIL-STD-498 which promotes evolutionary requirements design.

* In 2000, the DoD released a further spec, 5000.1 and 5000.2 which emphasized using iterative/incremental software delivery.

It’s a shame that no one bothered to RTFM 30 years ago before entrenching the waterfall into computer science to the extent that it’s difficult to undo.

Not to say we’re going to stop trying, though…