Notes On Process
From SchemePunks
Various notes on "Process". We are going to have to develop some processes for inherently difficult things, most especially making decisions on the content of the Standards we produce.
Here's some notes on all this, especially things that have been shown to work or not work. Hga 01:11, 1 September 2007 (PDT)
Contents |
[edit] The Democratic Technical Decision Making AntiPattern
(This first bit is from memory, since my copy of AntiPatterns was lost in one of my recent series of moves.)
In the final section of AntiPatterns there is a very interesting one the essence of which goes something like this:
Only 1 in 5 programmers "get" abstraction. No source cited, but that would seem to roughly match what I've observed in the field.
Therefore, "democratic" design processes result in the people who get abstraction being outvoted, and this is a major reason why so much software has such bad architecture.
Therefore, democratic design processes do not work. (Note that in this self-selected Scheme Punks group, the 1/5 rule almost certainly does not apply. One would hope that by definition, anyone who really likes Scheme "gets" abstraction, but Scheme has sufficient merits for those who don't that I'm sure this is not true.)
I'll add my own Original Research, which I think is self-evident: the random political processes that elevate people to decision making positions, to the extent they are not based on a true merit filled history of successful projects, will also tend to elevate people who don't "get" abstraction (purely by the simple distribution of the aptitude). And so this kind of non-democratic process also does not work (on average).
Moving beyond the specific of abstraction and to the reason we've all gathered here, it's hard to look at some of the YES statements for R6RS ratification:
and to retain much 1000 faith in a democratic decision making process for this domain....
What instead is currently left as an exercise for the reader, or whomever wants to make comments below in What Is To Be Done. Hga 01:11, 1 September 2007 (PDT)
- (in reference to "perfect" Hga 09:33, 14 September 2007 (PDT))
- Please be fair: the real 'R6RS is "perfect"' is out there. -- MF
- I believe I'm being entirely fair. The real 'R6RS is "perfect"' is "out there" ... but wasn't referenced in the vote. As stated elsewhere, the author knew his statement would be recorded for posterity, knew people would be reading it, if not now, in the future when there was no chance to know the context.
- To hammer the point home, it's hard to view his statement as at best anything less than a deliberate provocation. In any event, it betrays a fairly deep contempt for some important parts of useful democratic processes---and the fact that people have done this since the dawn of democracy is one of the reasons its potential fit to our needs should be carefully examined. Hga 09:33, 14 September 2007 (PDT)
[edit] The Myth of the Flat Earth
I've been wondering about this, since I knew the Greeks had figured out both its shape (roughly, it's a bit flattened at the poles as I remember) and its diameter.
Summary by Jeffrey Burton Russell
for the American Scientific Affiliation Conference
August 4, 1997 at Westmont College
How does investigating the myth of the flat earth help teachers of the history of science?
First, as a historian, I have to admit that it tells us something about the precariousness of history. History is precarious for three reasons: the good reason that it is extraordinarily difficult to determine "what really happened" in any series of events; the bad reason that historical scholarship is often sloppy; and the appalling reason that far too much historical scholarship consists of contorting the evidence to fit ideological models. The worst examples of such contortions are the Nazi and Communist histories of the early- and mid-twentieth century.
Contortions that are common today, if not widely recognized, are produced by the incessant attacks on Christianity and religion in general by secular writers during the past century and a half, attacks that are largely responsible for the academic and journalistic sneers at Christianity today.
A curious example of this mistreatment of the past for the purpose of slandering Christians is a widespread historical error, an error that the Historical Society of Britain some years back listed as number one in its short compendium of the ten most common historical illusions. It is the notion that people used to believe that the earth was flat--especially medieval Christians.
It must first be reiterated that with extraordinary few exceptions no educated person in the history of Western Civilization from the third century B.C. onward believed that the earth was flat.
[...]
Fashion is a strong driver in "consensus". It has no place in an honest search for the truth, but there is no denying its very strong power amongst "intellectuals", which all of us are by definition (you can't be interested in the domain of this group without being one, I claim). Hga 01:11, 1 September 2007 (PDT)
[edit] What Is To Be Done
No long term idea as of yet. For now, play the wiki game with mutual respect, make decisions by "unanimous consent" (as in the US Federal Legislature phrase, but less pro-forma), and go from there. Hga 01:11, 1 September 2007 (PDT)
[edit] What will happen on this wiki
Kumoyuki 03:18, 1 September 2007 (PDT) - Pardon me as I quote Mr. Heinlein, even as I realize that doing so comes perilously close to a preemptive strike under Godwin's Law 1000 .
From The Moon is a Harsh Mistress when the revolution on the moon had succeeded and they were trying to develop a government for a society that had essentially no legacy of centralized government (and yes, some of RAH's sociology is suspect, but I think this part is valid):
I'm a Royalist because I am a democrat...A King is the people's only protection against tyranny, especially against the worst of all tyrants, themselves.
This is the internet. And the internet is truly a web of tiny tyrannies giving an illusion of anarchy. This is generally viewed as a democracy because the tyrants generally want the approval of the communities they serve with the resources they control. In this case, the wiki owner has certain capabilities and responsibilities which don't truly devolve on to people who post their ideas. One of the main capabilities concerns the structure of the MediaWiki pages. There are two methods available for structuring our content, given the software that is being used to publish it:
- Page naming conventions
- Custom namespaces
So the pages relating to Kumoyuki's proposal have been moved so that they are named with the prefix K1, and all tentative proposals which use multiple pages should follow this same practice. Once a proposal has reached a critical mass, the wiki owner will 'promote' it to a separate custom namespace, which should allow the promoted proposal to grow more easily.
[edit] Decision Making and Formal Standards
Note that the preceding discussion is entirely about the technical management of the content on the wiki. It says nothing about whether a proposal gets an 'official' SchemePunks blessing as a 'standard' or anything else that the group may care about. Those processes should rightly be determined by some community mechanism, although I realize that the technical management of the content will have an impact on that. So I suggest the following life cycle:
- A proposal is floated, using page naming conventions to keep the content segmented
- If it achieves a large enough buy-in to justify a custom namespace, it will be considered a Project
- the pages will be moved into it
- a separate mailing list will be created for the Project
- A community process will determine when the Project should be considered finished
- The content pages will be frozen
- A public announcement will be made on (minimally) the Junta list and c.l.s
Major revisions to frozen documents will need to restart the process as a new proposal. Minor revisions should probably be able to be handled through the Talk pages.
