M. Handley
University College London
INTERNET-DRAFT E. Rescorla
draft-handley-doc-market-00.txt RTFM, Inc.
A Market for RFC Publication and Review
Status of this Memo
This document is an Internet-Draft and is in full conformance with
all provisions of Section 10 of RFC2026. Internet-Drafts are working
documents of the Internet Engineering Task Force (IETF), its areas,
and its working groups. Note that other groups may also distribute
working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference mate-
rial or to cite them other than as ``work in progress.''
To learn the current status of any Internet-Draft, please check the
``1id-abstracts.txt'' listing contained in the Internet-Drafts Shadow
Directories on ftp.is.co.za (Africa), nic.nordu.net (Europe),
munnari.oz.au (Pacific Rim), ftp.ietf.org (US East Coast), or
ftp.isi.edu (US West Coast).
Abstract
We describe a market mechanism for providing incentive for the review
of IETF documents. Reviewers would be "paid" by the IESG to for their
reviews. In turn, document authors would need to "pay" the IESG to
take up their documents. This mechanism rewards reviewers for their
reviews, thus (hopefully) increasing the quantity and quality of
reviews.
1. Introduction
The IETF process depends critically on peer review of documents. How-
ever, it is notoriously difficult to get quality review of documents.
Most documents receive few if any Last Call comments and it's quite
common to see that almost noone has read the documents at IETF meet-
ings. We believe that much of the reason for this lack of review is a
lack of incentives. Since reviewers are not compensated, it is not
surprising that the IETF process lacks sufficiently detailed early
review.
Handley, Rescorla [Page 1]
In this memo, we describe a system for compensating reviewers for
reviews. Although the IETF is not in a position to pay reviewers
actual money, that does not mean that there is nothing of value to
offer them. The most obvious thing to offer is the right to publish
documents. We propose to use this right as the source of our incen-
tives.
2. An IETF Currency
We propose to use the common tactic of introducing an internal IETF
currency, the "Itil", based on the right to publish documents. The
source of the Itil's value is that the IESG charges individuals and
working groups for the right to publish documents. The IETF can then
pay individuals for specific activities judged to be useful.
2.1. Charging for Publication
The central idea is that the IESG charges for admission to the publi-
cation queue. When a document is submitted to the IESG, the authors
must also provide a certain number of Itils. This admission fee is
what gives the Itil its value. Since no document can be published
without paying the fee, it's therefore desirable to be in possession
of Itils. It is not necessary that the authors of the document per-
sonally have the appropriate number of Itils. Rather, they merely
need to convince enough people to "chip in" so that they can collec-
tively pay the fee. This allows people to pay for the publication of
documents they consider valuable.
Note that the idea is not to circumvent the IESG quality review pro-
cess. One cannot simply pay one's fee and publish. Rather, the fee is
simply for admission to the queue. Once the fee has been paid, the
IESG review process proceeds as normal. Similarly, Area Director
approval would be required to submit documents, as per usual. The
purpose of this fee is purely to create the proper incentives.
The publication fee is intentionally left unspecified here. It is the
intent that the IESG could adjust the fee as necessary to keep the
queue of appropriate depth. We envision the IESG publishing a price
list and adjusting it on a periodic basis.
2.2. Issuing Itils
The basic way in which IETF members obtain Itils is that they perform
services judged to be socially useful. These services might include:
* Reviewing documents.
* Chairing working groups.
* Serving on area directorates.
Handley, Rescorla [Page 2]Internet-Draft A Market for RFC Publication and Review
We envision that the IESG would publish a price list of the value of
various services. Thus, IETF members could know what they were going
to be paid for various services.
2.3. Transferability
In the proposed system, the only built-in way to obtain Itils is to
get them from the IESG. The only built-in way to spend them is to pay
the IESG for publication. However, in practice, a system of fungible
tokens will usually end up being used for private transactions as
well. Rather than discouraging this, we think it better to encourage
it. Thus, for instance, members of the WG might pay an external (to
the WG) expert to assist them in developing their system. The exis-
tence of such a secondary market for Itils means that the IESG needs
to set the price on fewer quantities since the market will automati-
cally adjust them to their efficient values.
3. Practical Issues
In this section we discuss a number of practical issues relating to
our proposed system.
3.1. Ensuring Quality
One obvious objection here is that reviewers will submit low quality
reviews in order to accumulate Itils. This is easily solved by
requiring reviews to be accepted by a member of the IESG. If reviews
are not judged to be of sufficient quality then they simply would not
be paid for.
In practice the IESG might simply accept reviews by default, and
withdraw the payment after the fact if a valid complaint about a
review were received. As most reviews would be public, kept online,
and attributed to an individual, we expect social pressures would
prevent people trying to game the system.
3.2. Accounting
The conventional method of providing a currency is to have actual
physical tokens. This is inconvenient in the IETF. However, it's rel-
atively simple just to keep electronic accounts of everyone's Itil
balance on a web site somewhere.
The third-party transfer of Itils poses a minor authentication prob-
lem. However, unlike cash, Itils are not anonymous. Thus if someone
asserts that someone else has transfered one of their Itils without
their authorisation then it is relatively simple to audit what really
happened. Again, we expect that as the IETF is a relatively small
Handley, Rescorla [Page 3]
organization of peers, social pressure will mean that we shouldn't
need strong authentication mechanisms.
3.3. Undesirable Incentives
If the IESG changes a per-document fee, then there is an incentive
for document authors to merge what should be multiple documents into
a single document. However, if the IESG charges a per-page fee, then
there's a risk of the system quickly getting very complex, with pos-
sibilities like paying more for better reviews. A per-page fee also
provides disincentive for properly explaining the background to an
idea. Clearly it's undesirable for documents to be excessively long,
but it's also undesirable for documents to be excessively brief.
We believe the system has to be simple to work well. Thus we propose
a flat fee per document as a starting point. The IESG can always
require a document to be re-submitted if they believe that it would
be better as multiple separate documents. Charging a re-submission
fee would then provide a disincentive to submitting merged documents
in the first place.
3.4. Priming the Pump
In the initial stages of this system, there will be an undersupply of
Itils relative to the demand. We propose to prime the pump by simply
issuing a modest number of Itils initially. A number of criteria
could be used for ths initial grant, including a lottery, a minimal
bar (everyone who's published an RFC), or delaying charging for pub-
lication until a sufficient number of Itils had been earned.
4. Some Initial Prices
Although the IESG needs to periodically adjust prices, it is useful
to have an idea of an initial starting set. We propose the following
starting price list. Negative numbers are paid to the IESG, positive
numbers are paid by the IESG.
Activity Price
-------------------------------------------------
Admission to the IESG queue -5
Resubmission of a revised document -3
Review of a document +1
Chairing a WG +2/yr
Serving on the IESG +5/yr
Serving on the IAB +2/yr
Serving on a directorate +1/yr
Handley, Rescorla [Page 4]Internet-Draft A Market for RFC Publication and Review
Author's Addresses
Mark Handley <m.handley@cs.ucl.ac.uk>
Department of Computer Science
University College London
Gower Street
London
WC1E 6BT
UK.
Eric Rescorla <ekr@rtfm.com>
RTFM, Inc.
2064 Edgewood Drive
Palo Alto, CA 94303
USA
Phone: (650)-320-8549