full-disclosure - it's all about timing

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <1028150790.30447.82.camel@stantz.corp.sgi.com>
From: florin at sgi.com (Florin Andrei)
Subject: it's all about timing

(i'm going to go a little bit further from the HP/Snosoft case, so don't
be surprised if some of the statements below do not fit 100% in that
case)

All these problems will vanish if people will choose to disclose
vulnerabilities in a responsible way.
Sure, HP's response has been harsh. But every security problem
(especially when it's accompanied by an exploit) should be reported
first to the vendor! There should be no exception from this rule. The
person doing the reporting should give the vendor a reasonable period of
time to fix it; say, a few weeks or so.

Only if the vendor does nothing in these weeks, only then the
report/exploit/whatever should be made public.

If hacker H writes a comment on Slashdot, making public an exploit
against some software made by vendor V, and does not notify V in advance
(say, 2...4 weeks in advance), and then V sues H, then who's right?

H is right, because (s)he disclosed a vulnerability, and disclosing is
good.
V is right, because not being warned in advance, their customers are
left to the mercy of script kiddies.
H is wrong, because (s)he's obviously looking for cheap publicity (i
published a zero-day exploit; mine is bigger), not for improving
security.
V is wrong, because they are filing a lawsuit against open disclosure,
which is not a good thing.

See?

And the solution is so simple: DO NOT publish "zero-day exploits". Give
the damn vendors an early warning. Only if they are lazy and do nothing
within a reasonable time (2...4 weeks), only then you are entitled to go
slashdot-happy.

I'm a big fan of open disclosure, freedom of speech, etc. But people who
look for cheap publicity are not my favourites. If H is going to publish
the exploit without early warning, i'll say V has all the rights in the
world to sue the crap out of H, and put him(her) in jail for one
thousand years, and i'll applaud that.
However, if there was an early warning, within a reasonable time, like
one month or so (unlike some popular security companies did recently),
and the vendor did nothing and didn't provide a good reason for the
delay (because such reasons could exist, if you think of it), then H is
100% entitled to publish whatever exploit he likes.

It's all about timing. It's all about being reasonable.

-- 
Florin Andrei

"Some times are fuzzier than others." - Dan Farmer & Wietse Venema