[r6rs-discuss] meta r6rs

From: Thomas Lord <lord>
Date: Fri, 15 Jun 2007 10:54:56 -0700

Paulo J. Matos wrote:

> This is my first post to this ML but I'm following the discussion
> since the beginning and I can't understand your reasons at all.


I've enclosed a guide that might help.


> [...] what you want [....] not only seems anarchic to me but also
> defies the increased portability R6RS is about to bring.


Contents

   Summary of Conclusions
   Summary of Argument
   Final Remarks About Authority


Summary of Conclusions:

  a) the best outcome would be voluntary attrition from the
     process by both the steering committee and editors, with
     the intention that they re-convene around a SRFI

  b) in the alternative, people should boycott the vote
     because it is illegitimate

  c) in the alternative, people should submit a vote
     of "abstain" with a rationale that spells out the
     illegitimacy of the process

  d) in the alternative, people should vote "no", with a
     rationale that public discussion of the largest
     changes in the draft has been inadequate

  e) in the alternative, people should vote "no",
     with a rationale that explains objections to
     some of the large changes in the draft

  and

  f) should the process complete and end without
     ratification, the draft should be submitted as
     a SRFI

  g) should the process complete and end with
     ratification, that will be a damn shame




Summary of Argument:

1. There is no *need* to publish as R6RS

  1a. because the portability outcome will be the same

      (It won't change the plans of current implementers
       much. We don't know what future implementers will
       do either way.)


  1b. because the formality of the document will be the same

      (It's the same words on the page, the same history of
       how it was created, and the same "signatures", either
       way.)


  1c. because the economic value of the document will be
      at least as good

      (A SRFI, like an R6, can be cited in contracts,
       grant applications, text books -- any economically
       significant place where a reference to a formal
       standard is needed, a SRFI will do, for all
       practical and principled purposes. *This* SRFI
       can (and probably should) include "signatures"
       and endorsements from a very large number of
       people, distinguishing it clearly from less important
       SRFIs.)




2. There are *reasons to not* publish as R6RS

  2a. because the report process is too slow

      (A process which has taken this long since
       conception to nearing ratification is too
       slow to operate in meaningful feedback with
       the Scheme community.)


  2b. because this report process is too divisive

      (Most of the major decisions in the draft
       were taken in private, among the implementers.
       The formal comments have succeeded in collecting
       "bug fixes" to the draft and at opening up
       some "what color shall we paint the bike shed"
       issues (like library version numbers), however,
       notably, all formal comments attempting to
       revisit some of the major decisions have been
       rejected and with very little meaningful two-way
       discussion or debate.)


  2c. because the authority to publish revised reports
      no longer exists, with anyone

      (The draft demonstrates that, if nothing else,
       the historic continuity of the revised reports
       -- the continuity of a line of thought that
       is rooted back the "lambda the ultimate" papers --
       has broken. People have moved away from the
       process and new agendas have moved in.
       People who have been with the report process
       all along are in very, very different places
       than they were in revisions past. From
       a literary perspective, if nothing else,
       there would be some kindness in allowing R5
       to be the last revised report, ever.)


  2d. because the R6 process is anti-democratic

      (Formally, the R6 process is tyrannical.
       A sui generis oligarchy has granted itself
       sovereignty over community membership (the
       150 word essay issue), monopoly to publish
       draft "laws" for possible adoption,
       and the authority to override any vote,
       even of the "approved" citizens. No
       rationale has been offered for this arrangement
       other than the definite economic goals of
       a few and the alleged economic benefits to
       others. The R6 process is basically of a
       tyrannical oligarchy offering a (pretty thin,
       pretty theoretical) bribe (of a world rich in
       portable Scheme code that could not possibly
       be achieved without the oligarchic assertion
       of sovereignty) to collect votes which
       give a retrospective appearance of legitimacy.)


  2e. because revised reports cause terminological
      confusion and impose unfair economic costs
      on dissenters

      (I am sure I'm not alone in wondering, should
       R6 pass, what I will call the implementation
       I'm working on since the proper name "Scheme"
       will have been so radically redefined.)


  2f. because this draft in particular piles on
      many complex features yet no rationale has
      been offered that shows they can not be usefully
      reduced to a few more general, less restricted
      constructs

      (An example: the draft's Unicode support could
       have been enabled mostly by removing a few
       restrictions in R5, making the CHAR type
       more general. Instead, the draft proposes
       some hefty new restrictions on the CHAR type.)


3. There are *reasons to* publish as a SRFI

  3a. because the SRFI process is relatively fast
      so that later corrections, amplifications,
      and alternatives to the draft can be published
      along side the draft in a timely manner

      (This is inherent in the SRFI process.)


  3b. because the SRFI process is relatively non-rival:
      it does not propose to define the borders of
      "the Scheme community"

      (This is inherent in the SRFI process.)


  3c. because the authority to publish a SRFI is available

      (This is inherent in the SRFI process.)


  3d. because the SRFI process is democratic

      (This is only approximately true but it is a
       good approximation, inherent in the SRFI
       process. There is still a (formally) tyrannical
       oligarchic imposition of minimum standards of
       relevance, language, form, etc. however (a) those
       are relatively objective concerns so it is easy
       to see if they are being administered well;
       (b) should the "SRFI oligarchy" go insane
       and abuse that authority, it will be trivially
       easy to "fork" the SRFI process without much
       loss.)


  3e. because SRFI's provide a non-rival, structured
      name-space for standards rather than (further)
      muddying the meaning of the proper name "Scheme"

      (The literary function of the revised
       report series is to offer an authoritative
       definition of, to quote the title, "The Algorithmic
       Programming Language Scheme".

       I propose that that literary function is
       obsolete. In ordinary usage, the proper name
       "Scheme" is the name of a loosely defined set
       of design patterns for programming languages,
       where all programs that conform to the patterns
       contain a common, portable, tiny subset in
       terms of which all else in the language can
       (in theory) be defined. The report series
       has long aimed to define that subset, and
       the current draft diverges from that tradition
       substantially.

       It is surely useful to have a variety of
       formal documents, the product of public debate,
       to which to refer as we do and promote our work
       in Scheme. The SRFI process provides us with
       a non-rival name-space for such documents.)



Final Remarks About Authority

  Paulo J. Matos also wrote:

> You can vote no to the ratification! What more do you want? There
> will always have to be some group at the top making the hard
> decisions. If you don't like them, vote no but putting all the
> hard decisions in the hand of a community is not wise, IMO.


  What you are saying, in effect, is that the Scheme community
  is populated by a bunch of superstitious idiots for whom
  the magical title "R6RS" will hold some important totemic
  significance not found in a title like "SRFI-<N>: First Report
  on the Scheme Workshop Programming Environment".

  I have higher expectations of this community than you do,
  I think, although I admit I have fears about equally as low.

-t
Received on Fri Jun 15 2007 - 13:54:56 UTC

This archive was generated by hypermail 2.3.0 : Wed Oct 23 2024 - 09:15:01 UTC