[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
Re: CVE for hosted services
Breaking this off to a separate discussion.
On Mon, 27 Feb 2017, Art Manion wrote:
: > - CVE supported two prefix schemes for a decade (CVE and CAN).
:
: There was a reason for two schemes, the world changed, and CVE
evolved.
: I recall it being cumbersome at best (although it was probably
: worthwhile in the early years of CVE).
:
: What does CAN/CVE mean in this discussion?
I am truly, and honestly baffled, at this question.
The CNA/CVE abstraction from day one made sense. Historically, it was
the
board voting on if an issue warranted a CVE assignment. It was a
CANdidate
until the board voted, or MITRE made an execute decision. The MITRE/CVE
site actually showed those votes for a decade.
If there were two schemes, for vuln in software (i.e. the context and
purpose of CVE), for a *decade*...
How can you possibly ask what CAN/CVE means in this discussion?
Where we're (starting to) debate tracking site-specific vulns, which
were
absolutely against CVE policy three weeks ago, that I had to clarify on
list as some CNAs were "we're selfish, we want to use CVE to track
site-specific crap".
This on the back of some CNAs voting against logic back during the epic
renumbering scheme, moving past a four-digit identifier. That years
later,
MITRE arbitrarily said they were changing again, without a word to the
board, until the news outlets called them on it.
Seriously Art, there are levels upon levels of history here, about
changing the scope or numbering scheme of CVE. I can't begin to
understand
why anyone would casually dismiss that history and then argue, "lets
mix
in vulns that were against the rules for 17 years" without considering
an
abstraction in prefix.
The CNA/CVE choice, in 1999, made sense. But the board was radically
different. After the board stopped voting on each CVE entry a year or
three later, the CNA/CVE designation lost value. Years after that, it
was
hitorical academic masturbation at best. OSVDB was the first VDB that
publicly and loudly told MITRE "we're not playing that game", and
dropped
the CVE/CNA designation. We started using the numeric identifer only,
because it worked either way. Both schemes took you to the same entry.
I
argued with Christey/Coley on that for years, and ultimately we told
him
we were dropping it because it made no sense. Back when OSVDB had some
measure of industry respect, that said something. Within a year, MITRE
dropped that designation.
So now... we're faced with adding site-specific vulns, that again...
were
against policy for 17 years. And you are really questioning the *idea*
that they get a different designation?
Please. This isn't about CNA/CVE, at all, and it shouldn't be to anyone
involved in this process.
This is about CVE / CME / CWE / CPE / [other C*E] projects. Spin it
off,
let it develop and evolve under a separate project [0]. If a CVE vuln
impacts a site-specific service, they can cross reference. And there is
some failed precedent here, as IBM has issued CVE IDs to site-specific
issues in the past (IBM BlueMix junk, that later became a hybrid
customer
premise / SaaS offering, further convoluting things [1]). It caused
problems back then, and the mix of site-specific vulns still plagues
the
CVE offering to this day. Anyone can request a CVE ID with minimal
information, and MITRE assigns. Then we find out it is a) not a vuln b)
site specific or c) both!
.b
[0] This may be problematic to MITRE to figure out funding, be it in
the
scope of CVE / 2 other projects under that contract, or spin up a new
contract and convince DHS to fund it. Don't care. They are a horrible
orgnaization wasting too many tax dollars as is. They can figure out
how
to con the government out of more money. That is not the CVE board's
concern. If you disagree, cite the threads where you challenged them on
wasteful spending in the past decade. =)
[1] If anyone on the board is surprised by this bit, why? The CVE board
is
about directing CVE in the context of the *industry*. Not just YOUR
organization. I am getting really tired of pointing this out.