Re: OWASP Top Ten - My Case For Updating It

["Dean H. Saxe" <dean () fullfrontalnerdity com>]

Having worked with Jeff in the past (he's one of the brightest guys  
I've had the pleasure of working with), I'm familiar the problems  
that he's making mention of.  The problem is inertia.  If you've  
always done things a certain way and been successful at it --  
measured by length of time in business, customer base, lack of  
discovered hacks, whatever -- there is no driver for change.  And  
when there is a driver for change -- customers demanding less buggy/ 
more secure software -- the focus is going to be on fixing the  
obvious flaws!

Until we fix the process of building software in shops that are  
without process controls, we'll never change the outcome.

-dhs

Dean H. Saxe, CEH
dean () fullfrontalnerdity com
Here in America everything is bought and sold, you can get anything  
for little bits of gold.
We'll rape the earth and ruin the air, cut down every tree from here  
to there.
    -- Donna The Buffalo "America"


On Jul 11, 2005, at 8:11 AM, Mark Curphey wrote:

> Hallelujah brother !
>
> -----Original Message-----
> From: Jeff Robertson [mailto:Jeff.Robertson () DigitalInsight com]
>
> > -----Original Message-----
> > From: Mark Curphey [mailto:mark () curphey com]
> >
> > If the problem of web application security is poor software quality,
> > it is a natural conclusion that the solution is to build better
> > software. Not once in the top ten does the list address the fact that
> > the majority of software is built without a design, security
> > requirements or a repeatable software security development process.
>
> I would go so far as to say that unless a development shop is already
> following a process (I don't want to start waterfall vs. RUP vs. XP  
> wars
> here) to keep plain old functionality bugs down to a minimum, they  
> have no
> hope of producing secure software.
>
> If a software company haven't even figured out that their  
> developers need to
> be doing unit tests, then the idea that they could successfully  
> implement
> any sort of security testing is just putting the cart before the  
> horse.