Wednesday, October 05, 2005

Do we need specs? Linus thinks not ...

Linus Torvalds Outburst Sparks Fierce Debate: Does Open Source Software Need Specs?
"There's Two MAJOR reasons to Avoid Specs," Torvalds Wrote

The Linux Kernel Mailing List (LKML) came alive on Thursday when Linus Torvalds chose to write, in a discussion about the risks of following specs without being flexible enough to take reality into account: "A 'spec' is close to useless. I have _never_ seen a spec that was both big enough to be useful _and_ accurate." The web is currently abuzz with the merits and demerits of this point of view.

As we always try and do, let's first look at exactly what Torvalds wrote:

DateThu, 29 Sep 2005 12:57:05 -0700 (PDT)
FromLinus Torvalds

"A 'spec' is close to useless. I have _never_ seen a spec that was both big enough to be useful _and_ accurate.

And I have seen _lots_ of total crap work that was based on specs. It's _the_ single worst way to write software, because it by definition means that the software was written to match theory, not reality.

So there's two MAJOR reasons to avoid specs:

- they're dangerously wrong. Reality is different, and anybody who thinks specs matter over reality should get out of kernel programming NOW. When reality and specs clash, the spec has zero meaning. Zilch. Nada. None.

It's like real science: if you have a theory that doesn't match experiments, it doesn't matter _how_ much you like that theory. It's wrong. You can use it as an approximation, but you MUST keep in mind that it's an approximation.

- specs have an inevitably tendency to try to introduce abstractions levels and wording and documentation policies that make sense for a written spec. Trying to implement actual code off the spec leads to the code looking and working like CRAP.

The classic example of this is the OSI network model protocols. Classic spec-design, which had absolutely _zero_ relevance for the real world. We still talk about the seven layers model, because it's a convenient model for _discussion_, but that has absolutely zero to do with any real-life software engineering. In other words, it's a way to _talk_ about things, not to implement them.

And that's important. Specs are a basis for _talking_about_ things. But they are _not_ a basis for implementing software.

So please don't bother talking about specs. Real standards grow up _despite_ specs, not thanks to them.

Linus

1 Comments:

Blogger bjnmary said...

Agreed - so the question about specs is: Are specs required to provide coding direction, or are specs needed to document what did take place and for future reference. This leads to the second problem, specs are usually rendered inaccurate before they are needed. If I could wish for a future development tool, it would be a programming tool that could create specs directly from the code. Then the specs are always up to date. Not sure if they are up to date? Then just regenerate!

2:15 PM  

Post a Comment

<< Home