Skip to Content.
Sympa Menu

cacert-policy - Re: SP going to WIP or DRAFT?

Subject: Policy-Discussion

List archive

Re: SP going to WIP or DRAFT?

Chronological Thread 
  • From: Ian G <iang AT>
  • To: cacert-policy AT
  • Subject: Re: SP going to WIP or DRAFT?
  • Date: Wed, 31 Mar 2010 14:31:54 +1100
  • Authentication-results:; dkim=pass (1024-bit key) header.i= AT; dkim-asp=none

On 31/03/2010 11:45, Mark Lipscombe wrote:
On 3/31/2010 10:46 AM, Ian G wrote:

Reading the policy, it is pretty clear that the veto is entire, over an
entire document (strike 2.). [b].

This is an argument that logically has quite some merit, but lacks the
policy substance within PoP to make it clear.

PoP lacks direct words because the default meaning of VETO is quite clear. Also, the history, the practice and other documents quite clearly support the singular, complete intent.

However, I would like to
put forward some alternate readings of the policies (and lack thereof):

"Policy" is not the same as "policy document"

A policy document, such as the Security Policy, is a collection of
policies, able to be distilled into "units" of policy. Under this
theory, the board may therefore use its veto to strike down any policy
that "effect[sic] the running of CAcert Inc.".

Well, it's a theory, but it has no place in CAcert.

A policy is a well defined thing, in practice, in PoP, and in the words of each individual Policy. A policy is a single cohesive document that is identified as a special thing under audit, documented in the Document Officer's list of CAcert Official Documents (COD, soon to be CCS Annex 1) and referenced frequently in itself, singularly, and in each of the other policies.

For example, do a grep of COD in CPS. Each of those refers to "a policy" with "a number" and "a status."

If we accept this reading, we then have to decide what a distinct "unit"
of policy is -- is it:
(a) SP bulletpoint 5 ("Board") (the narrowest unit)
(b) SP - Background Check Procedures - Coverage
(c) SP 9.1.4 - Background Check Procedures
(d) SP 9.1 - Staffing
(e) SP 9 - Administrative

Those are headings under a policy. Calling them "units" and then inserting this term into "a reading theory" over PoP is just a lot of smoke and mirrors.

There are substantial advantages to this reading, in that it narrows to
the smallest possible focus the authority of the board veto, to strike
only at individual "policy units" which affect the operation of CAcert,

It also allows CAcert, Inc. to use it's veto with the smallest amount of
fallout necessary to satisfy its obligations.

When we wrote all this, we did not write MagicWandPolicy! If we had done such a thing, we would likely have added this clause:

  1.1 Usage
  The Magic Wand is reserved to the Arbitrator,
  henceforth known as The Good Fairy.

CAcert, Inc's veto applies only to CAcert, Inc.

CAcert, Inc's veto may be reasonably construed to be merely a veto of
it's obligations under the policy.

That's a dangerous argument.  On the one hand,

1.3 The policies so created are generally binding on CAcert, registered users and related parties.

So, CAcert Inc can decline to follow the obligations, but everyone else would be well advised to treat the PoP seriously, coz they aren't going to risk a raised eyebrow before the Arbitrator.

On the other hand, we have nothing: the SP is mostly about people and processes directly inside the CAcert Inc's board's role. It's all about the CA that CAcert Inc's board runs on behalf of the Community.

So there isn't a sensible "limitation of damage" argument there... The only major group outside is the Arbitrators, and they'll not hear any nonsense :)

(Not to mention, the policies are quite clear about when and where they "limit" the scope of any clause. E.g., the very famous ABC is clearly limited in its scope. So crafting your own limited scope is dangerous.)

(And, the more CAcert Inc acts to be different to the Community, the more the Community starts muttering of rebellion and revolution....)

In the absence of any conflicting policies and the distinct lack of
contemplation on what happens next in the PoP, it's not a big leap.

Question for the Board.

The community through its association membership, still retains an
effective method of overriding that veto, through the use of a special
general meeting.

Well, this is true. But I don't see the relevance, as most people are not denying Board's right to veto.

An arbitrator would also be entitled to look behind the
reasons for doing so, on the basis of CAcert, Inc being a party to the
CCA and a member of the community, and rule on the validity of the veto.


Policy group could vote to return to DRAFT upon change to WIP

The policy group could seek to establish rough consensus prior to the
veto motion coming into effect, deciding that the document return to
DRAFT with the section that CAcert, Inc. objects to being struck.

It would be pointless to do it while the veto is not in effect.

However, once done, and the dust settles.... the policy group can, on its WIP document, strike the offending parts, rewrite them, enhance them, or leave them as they are. I think everyone here is well aware of all our options, and at least two suggestions have been made (e.g., strike, and make optional).

But, that's for next week.

would remove the problem of any actual or perceived "vacuum" of policy.

Oh, you mean a period where we have no policy? No I'm afraid that's impossible. We need at least a week for a vote, and for this controversial area, we'd need a couple of weeks, or a month, out of an abundance of caution.

These policies are binding over the critical roles, and they create quite severe risks, liabilities and obligations over the people who are ruled. By their appearance they also protect the people in quite serious ways. The Arbitrator needs a fully-thought out document in which to apply his rulings, and protect those critical roles from R/L/O. Courts need to see that we laid out good documentation and good powers, in order to have respect for our actions.

So we've brought this policy group up to take measures carefully, seriously and with due thought. Lots of thought, because the way these policies interlock into CAcert society is important.

DRAFT-in-VETO quasi-status

Occam's razor isn't irrefutable. It may well be that we are not yet
beyond necessity.

Well, sure. What I meant by that is that nobody's put a good idea on the table for a new status, and in the absence of good clear ideas, it looks unlikely.

Given the lack of direction given for this situation in PoP, it could
reasonably be argued that a new DRAFT-in-VETO "quasi-status" is created
through the use of the veto, allowing the policy group some period of
time in which to "fix" the policy before it would otherwise revert to WIP.

Sure, except VETO is such a solid word, and the policy goes to extreme care to make things simple. If it wanted another state, it would have said so, but actually ... the draughters would simply say "return to WIP." because that's the simple consequence. Simplicity => certainty => efficiency. It's hard enough writing policies as it is without having to throw the sheep bones and consult the gods every time we change status.

The DRAFT policy that is vetoed already meets all the requirements to be
in DRAFT contained in PoP, and it has previously satisfied all the tests
needed to pass through WIP status. How can it make sense to return to a
WIP status that it bounces immediately past anyway under PoP?

Because something serious went wrong. We don't know what it is, but from the policy group's perspective, something went so wrong with the way it was written, that the board hit the big red button.

If it was a minor problem, we would have been told (a long time ago), discussed it, and got it fixed. Ergo, it's a major problem. We need time and thinking. QED.

We're all doomed, it's law of the jungle

Just the same as you could reasonably read the things above into the
vast open space that is the PoP's description of what happens now, you
can also read it to mean, as some already have, that the only way
forward is to throw out the SP, start again, and let the law of the
jungle rein free in the meanwhile.

Well, we in the policy group have done both that and the alternate. CCA, DRP, PoP, SP were all done from scratch. CPS went through a long migratory phase from "what we had". So I think we can make that decision comfortably as to whether we choose the jungle or the highway for SP.

If it's this reading that becomes the consensus, then future policies
should be built with this in mind, and broken down into smaller
policies, so that the baby doesn't get thrown out with the bath water.

Smaller babies ... means smaller mommies, smaller bottles & spoons, and a whole lot more activity for smaller results!

Breaking bits off, to wash them one at a time?  Sounds noisy & messy ;-)

In closing, given that the PoP lacks even the most basic of guidance for
this situation, I would submit we should read it in the light most
favourable to the intent of the instigators of the action at hand.

I would submit, it should be read how an Arbitrator is going to read it.

The draughters and the board at the time of 2007 did not agree with the viewpoint of the "instigators" of today. OK, I am the only one left and maybe my memory is faulty ... but I do know for a fact that they clearly understood the problem, they considered it, and they put in the fix.

The fix was deft, and small, and the absolute minimum.

They certainly didn't intend to give the board what it wants, because that was precisely the problem we were there to solve. In those times, the board had the power, did not know what it wanted, did nothing, and totally ignored the community's audit priority. PoP was one part of the solution to the general problem. It took the policy approval power (one plank of audit) totally out of the hands of the board (PoP). In other words, a wholesale transfer of policy approval from the board to the community.

A clear-as-crystal conclusion to that was that they weren't about to return any of that power back to the board. They only gave the board the absolute minimum necessary *to preserve the legalities*. No selective veto; the board weren't up to it, they already proved that in spades.

We can all see this today. In the opinion of the board, veto. This preserves the legalities. Nothing more is needed to solve the board's claimed problem of legalities.

(It is an exercise for future historians as to how inspired a choice that was.)


does not preclude the drafting of amendments to clarify future action
under PoP, even if those amendments are the polar opposite.


Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Archive powered by MHonArc 2.6.16.

Top of Page