Security pros gathering at a Stanford University Law School conference on responsible vulnerability disclosure Saturday harmonized on the principle that vendors should be privately notified of holes in their products, and given at least some time to produce a patch before any public disclosure is made. But there was pronounced disagreement on the question of whether or not researchers should publicly release proof-of-concept code to demonstrate a vulnerability.. . .
Security pros gathering at a Stanford University Law School conference on responsible vulnerability disclosure Saturday harmonized on the principle that vendors should be privately notified of holes in their products, and given at least some time to produce a patch before any public disclosure is made. But there was pronounced disagreement on the question of whether or not researchers should publicly release proof-of-concept code to demonstrate a vulnerability.

UK-based security researcher David Litchfield, of NGS Software, said he publicly swore off the practice after an exploit he released to demonstrate a hole in Microsoft's SQL Server became the template for January's grotesquely virulent Slammer worm. At Saturday's conference, held by the university's Center for Internet and Society, Litchfield said he wrestled with the moral issues for some time. "At the end of the day, part of my stuff, which was intended to educate, did something nefarious, and I don't want to be a part of that," said Litchfield, a prolific bug-finder.

That kind of soul-searching is music to Microsoft's ears. The disclosure standards promulgated by the Organization for Internet Safety, an industry effort founded by Microsoft and handful of large security companies, require researchers to withhold any exploits from the public for at least 30 days following the first public advisory on a bug. But Redmond would like to see researchers abstain entirely, said Steve Lipner, the software-maker's director of security engineering strategy. "We prefer that finders wait before releasing exploit code, or, better, don't release exploit code," he said. "It's something where ... we're trying to ask for cooperation, instead of something that we're trying to mandate or dictate."

California-based security vendor eEye and the Polish white hat hacker group LSD -- both prodigious exploit publishers in the past -- have taken to withholding proof-of-concept code when disclosing serious security holes.

The link for this article located at SecurityFocus.com is no longer available.