PlanetMath Administration (Main Document)

PlanetMath Administration (Main Document)

Administration of PlanetMath

February 9, 2018

1 About this document (jac)

I wrote this document to address what I think is an overwhelming needfor improved communication among PlanetMath participants aboutactivities, goals, and resources. It is early in its life cycle;however, I hope it can serve a useful organizational role as itdevelops. Over the course of the next couple of months, I intend toflesh it out with many additional details on goings-on aroundPlanetMath, and supplement it with links to other documents ofadministrative interest. At the same time, I have made the documentworld-editable, so that other PlanetMath users can add details or ifnecessary make corrections directly.

1.1 Formatting Convention (Wkbj79, jac)

In order to make it clear who said what, the following convention isused: In any section or subsection, the author(s) of that section willput their username in parentheses. Unsigned subsections of a signedsection should be assumed to have been created by the main section’smain author. Contributing authors should feel free to modify existingsections, although major thematic changes should certainly be signedso that everyone knows roughly who said what.

1.2 References to Other Administrative Documents (jac)

This document will be a high-level map, and other documentswill go into detail on specific areas. In general referencesto these documents will appear in the text, with URLs in footnotes.However, here is a short list of several documents that areeither especially important or that just haven’t been integratedinto the text here in the manner I described.

1. PM Community Guidelines http://planetmath.org/?op=getobj&from=collab&id=112

2. PM Content Committee http://planetmath.org/?op=getobj&from=collab&id=113

3. PM Member Rights http://planetmath.org/?op=getobj&from=collab&id=116

4. PM Point System http://planetmath.org/?op=getobj&from=collab&id=114

5. PM Task List http://planetmath.org/?op=getobj&from=collab&id=118

2 Administration (jac)

We should try to understand the terms better. When I say“administration”, really you could just think of a live updating mapof what is going on in the project, but it should be both as detailedand as foldable as any individual wants, and capable of being viewedfrom on at least as many different sides as there are viewers.

In other words, this is what I have been talking about with thescholium system or “Arxana” all along– except that it is not just some abstract construction – theadministrative map should of course be filled with actual details suchas who is doing what and who is responsible for what, so that ifanyone has a question about the project, either they will find ananswer on the map, or they will find a good place to stick theirquestion to get it answered.

Now despite my optimism that some version of Arxana will be availableand consumable by the end of the summer, people have administeredprojects for millennia without it, and I think it would be very apt ofus to look around for some older technology which we can use to buildour administrative map.

I would like to suggest a simple latex file saved as a PlanetMathcollaboration. Sections can be cut and added, content can be movedaround or changed, questions can be inserted. Additionalcollaboration documents can supplement this one.

Thus, over the last few paragraphs I have done more than just make anabstract suggestion – I have kicked off an actual solution. Let mecontinue that briefly before I get on to my second question.

2.1 An Administrative Matrix (jac)

One of the key pieces of data about an organization is its tasklist. The tasklist is populated with tasks, which can bedefined broadly or finely, and which can be finite or ongoing.Furthermore, each task can have subtasks, and this progession cancontinue, down through the minutæ of meticulous Fordism, or up intothe ætherial realms of an organizational Mission.

But, we need not, at any given point in time, be so detailed.However, the tasklist itself should be complemented by some basicinformation about the tasks involved. For example, a task might beannotated with a description of its participants, its consumers (or “stakeholders”), its schedule, and itsbudget (or “requirements”). Other suchlike fields canbe added as interest and expediency indicate.

Having described what I mean by an administrative matrix, Iwill now construct the beginnings of such a matrix for PlanetMath, asan example (see Table 1). Since I do not know enough aboutall the tasks or all the details of all the tasks going on atPlanetMath, this will not be particularly detailed. I will focus ontasks that I know something about, i.e., things I have been working onor following.

Please feel free to add or update with details on the tasks you havebeen working on, if you like. Mainly I am posting this fordiscussion: however, I would like to see a more industrial-strengthversion of this sort of thing put into play on PlanetMath soon.

label	description	schedule	requirements
Arxana	Make browser and interactive store, retrieve, display, edit features.Try to get clusions and occlusions working.	Maybe add a few more tiny things now, butit seems like a higher priority to get elephantand noosphere stuff worked out first.	Internally, this should follow our Elephant storageand query design; externally, it should follow thenoosphere (re)design.
SLIME	Can SLIME connect to a MMTN server?Can we once again use SLIME to connect to a regular server?	Low priority to connect to MMTN.Get regular remote connection back up soon.	Long-term goal may depend on getting MMTN set up properly.Short-term may depend on help from Nick Thomas?
MMTN	Implement proper daemon system.Desocket for local connections (?).Write a Lisp client.	Cf. chat client “erc” soon to see how it connects to Emacs.	Understand the current system.Come up with trial application.
Elephant	Chalk out ontology and seach interface well-suited to our apps.Locking and permissions.	Start right away listing desiderata.	Match our desiderata to the elephant docs.Help from Ian Eslick and pals as needed.
aserve	Start with interactive web interface to elephant.	Perhaps get a REPL soon, as a demo?	Some time with the docs if we want to know what all thisthing can do.
noosphere rewrite	Survey to find featureset and reusable modules.	Our rewrite must be done by August 20 for GSoC review.	Would be good to talk more with APK as this develops.Would be good to have a central place to do thedesign and fill in pieces of code as they appear.
sys. admin.	Persist configs and work files.Make laptops usably useful.	ASAP.	Internet connection. Superuser passwords. Physicalaccess to the computers in question.
math/tech talk	Prepare (and give) 15 minute talk on PM and HDM for Mathfest.Use a holistic perspective and try to convey mathematicalconnections/relevance.	Must be finished by the beginning of August. May givea variant talk around August 8?	Need a place to develop. Feedback from Ray, Aaron, others.Perhaps some background reading/references.
“main” for PM	Keep maintaining this document.Survey various tasks that other PM people areworking on, and provide annotated index here.	So far this is a “whenever I want to do it”thing. But interestingly, or ironically, thissort of matrix/map (especially in an improved format)would be the sort of thing that would be good toupdate everyday.	For this to work extremely well,it should be networked into people’s methods forparticipation, and have many participants.
content committee	This is Chi, Alvaro, Ray, et al’s thing.I’ll want to watch and record what theyare doing, or involve them directly in asuitable administrative approach to thiseffort.	Would be nice to be updated of changes/progess,and act on them to have them reflected rightaway.	Some degree of cooperation from other thecontent committee’s participants will be necessary.
networking	Figure out possible places to getsupport for our efforts.	Want to have a number of good optionslined up by the end of the summer.Should build a more detailed schedule/planof action.	Phone (check); travel budget for West Coast trip;some sort of sanction from PM to act as itsrepresentative.
fundraising	Figure out and share value propositions with possiblefunders. (This is in some sense a projection of “networking”.Note that by building PM’s budget, our efforts may be bettersupported at PM; but this is not the only approach worth pursuing.)	Want to have a number of good optionslined up by the end of the summer. (Hopefullyenough of a sample to see if our approaches aregoing to pan out.)	Any individual working on this will need a PM sanction.Other than that, collectively we will want a coordinated plan.
FEM	Make a print version of the PM encyclopedia.Ideally it would be rebuilt frequently.	This project has been stalled for a long time:it would be nice to revive it this summer, butso far I don’t know of a specific plan or timeline.First goal would be to make one	A major requirement that we noticed, but were notable to institute, was a system for unified notationamong articles included in the print collection.There are various other technical steps that we wouldwant to make this work well.

Table 1: An example of an administrative matrix

Note that the format used here is pretty far from ideal, certainlyfrom the point of data entry (since it would be nice to just be ableto zoom in on the tasks and fields that one is particularly interestedin updating), and also, at least in some senses, from the point ofview of display (since it would be nice to be able to fold certainrows and columns out of the way).

One of the makes this effort somewhat difficult is that differentlabels (e.g. “noosphere”) may mean different things to differentpeople, so there is, at least at first, some degree of collisionbetween points of view on what a given task is about. I think theambiguity can be resolved with enough creative relabeling.

Another more complicated way to resolve this sort of conflict would beto give each person their own matrix where they would be free to usewhatever labels they like. But it would take a lot of work toreconcile this approach with the properties of “being anadministration” as I have defined them above.

3 Roles of PlanetMath participants (jac)

A quick mental check indicates that we have at least these broadcategories of participants:

• President

• Board Members

• Site “Admins”

• Content Contributors

• Interns

• Consultants

• Bounty Hunters

In the subsections that follow, I will discuss some of the roles takenon by people in these categories – I do not assert that thisdiscussion is in any way “complete” – rather I’ll just say, we cancome back and add details as appropriate, at any time.

3.1 Role of the President

• Almost every web-related issue gets routed through Aaron.

• Almost every financial issue gets routed through Aaron.

• Most practical “beaurocratic” decisions are made by Aaron.

• Many key “social networking” issues are routed only through Aaron.

3.2 Role of the Board of Directors

• Less frequently, decisions are made by members of the board.

• Given the fact that the board does not have public meetings veryoften, the ability of general members of the community to oversee orengage the board members vanishes to zero.

• See the section below on Governance for more details.

3.3 Role of Site Administrators

• There are some arcane and non-obvious rules and responsibilitiesthat devolve onto persons with so-called “admin” status on thewebsite: I do not know the structure of this continent, so I willnot say any more about it; it may be important for issues like thosethat Chi has been talking about, may not be so important for theactual business of “running the non-profit”, but who knows?

3.4 Generic Roles of Community Members

• Obviously the most frequent role engaged in by community membersis developing mathematical content for the site. Currently most ofthis content is in the Ecyclopedia, but discussion of math topicsin the forums also plays a significant role.

• Media-wise, most “important” (at least, in the sense of beingat all “persistent” and “public”) conversations among the powersthat be get mapped through the cumbersome forum software deployed onPlanetMath’s home page.

• Other than that, there are various unofficial and more (or less,in the case of the email list) transitory media

• Various informal decisions and actions are taken by thequote-unquote “powers that be” around PlanetMath, which includemyself and too many engaged PM contributors to list here – but notethat these powers are almost universally “unofficial” powers (allbinding decisions must go through Aaron or, possibly, the board).

3.5 Community Initiatives

• content committee (otherwise known as “community guidelines”; I don’tknow what the latest perspective is, but see footnote for reference toone writeup)

• maintaining this administrative document

• fundraising and other outreach efforts

• coordinating the PlanetMath membership situation (cf. section onGovernance below)

• …

3.6 Proposals for change (jac)

• Well, I think that Aaron’s current set of responsibilities madehandleable by more people. Not because he has in any way “done abad job” – far from it – but because the rest of us have doneessentially nothing, due to the existence of an administrativebottleneck that he just happens to mostly fill.

• I would like to see somewhere an official record of thethings people are working on – including me. This could eventuallylead into either “job titles” or “volunteer roles” (such as theyhave at Wikipedia), which I think would do quite a lot to givePlanetMath more than one “official” public face.

• Extending this (and this is something we have talked a lot aboutbut I think left sitting for a long time), I would like to see aclear description of “members rights” – hopefully made with atleast some serious participation by members.

• I think that many of the problems around PlanetMath – probablymost of the serious ones in my view – are properly administrativein nature (according to my definition above: we lack a good map!).But I think that this has repercussions for “morale” (among otherthings). Participants who do not have any institutional authorityare not likely to do much in the way of outreach, for example – orto be able to do much.

3.7 Comments on Administration (Wkbj79)

Please add additional subsections if you wish to make further commentson what has been said in or about the section on Administration.

With regards to the first point in the second bulleted list above. It seems unfair that Aaron has had most of the burden placed on him.

As for jac’s “proposals for change”, why not add such an officialrecord to this document? [Done or at least begun: see “CommunityInitiatives”. –jac] On a related note, information about the soon to be formed content committee is available in other collaborations and documents which are similar to this one.

4 Business (jac)

4.1 Basic Data on PlanetMath as a Business

• The nature of the institution. We describe PlanetMath as being“a non-profit”, “a commons-based peer production system”, and “acommunity”. As the name indicates, “math” is the main focus.(Although if we follow Marnita’s advice, we may build severalspin-off, partner, or child institutions.) At present, there is nopermanent paid staff retained by the non-profit.

• Legal status. The organization is (not without some technicalsnafus, at least potential ones) a tax-exempt charity based inthe United States of America; and the content and code is madeavailable worldwide under some “free” license or other.

• Revenue streams. Subject to limitations placed on charityorganizations, PlanetMath can receive donations, have financialsponsors, get federal and other grants, and sell a few things (butnot too many).

• Work product. The content of the PM site (encyclopedia, forms)and the code behind or adjacent to it (Noosphere,NNexus, Arxana,etc.).

4.2 Human Resources in a Mostly-Volunteer Organization

(And, in particular, how to have them without upsetting any applecarts.)

Experience indicates that paying someone for a service is not always agood idea. A volunteer might find the idea of being payed to workdistasteful: “I’m doing this because I want to, man, not formoney!”

The person speaking could be an ideologically motivated hippie or anideologically motivated millionaire, but in either case, offering topay them may seem like an attempt to rob them of their chance to give.

At the same time, many people do like – and most need – to be paidfor some form of work. Different payment schemes can be more or lessmotivating for different people, in different contexts. I’ve known anumber of people who have gotten paid a flat hourly rate and whohappened to learn that they could get away with very little or no workon the job. Certainly even for wage-laborers, money is not the solemotivator. There is also something to be said for such things aspride in one’s work, a sense of growth and achievement!

Which further highlights the fact that a “payment scheme” is notgoing to answer all questions of personal motivation.

Speaking for myself, there are a number of different things I like todo – but I might not be as motivated if I was compelled for somereason to focus exclusively on one of these things. (At the sametime, complete dissipation wouldn’t work so well either.)

• Payments for administrative work. Since I might be getinto a payed arrangement with PlanetMath that includesadministrative work, I want to think about how to do this “right”– for my own personal, subjective, reasons. However, other peoplemay get into similar relationships, so, it is also good to thinkabout suitably general ways to make payment for services work inthis community. (Cf. the section on Work Plan, below.)

• Payments for coding work. In a setting in which mostparticipants are not programmers, payments for coding may be theonly way to get a sufficient flow of code into the organization. Atthe same time, I think the software projects associated withPlanetMath should be very carefully administered, especially so thatknowledge transfer can make it possible for more participants tobecome volunteer programmers.

4.3 How to employ people in a non-profit organization (alozano)

This section describes an ideal policy for hiring.

1. First, an active board should be elected, that closely oversees theoperations of the organization (the current board at PlanetMath is sadlyinactive in the community, except Aaron).

2. Second, the board needs to explain (or somebody needs to explain to theboard) the position to be created, the duties, responsibilities, etc attachedto the position. It needs to be stated clearly how and why such position needsto be created, and determine what is the salary (if a salary will be awarded).

3. If a salary is to be awarded, the board (or whoever proposes the creation of a position) needs to guarantee that the salary will be sustainable within the current (and foreseeable) budget and finances of the organization.Most importantly, since the organization is built in volunteer andcollaborative efforts, it needs to be stated and explained the need of payingfor such position and why the duties cannot be fulfilled by the volunteer body.The board also needs to determine the minimum necessary qualifications for anapplicant to be eligible for such position.

4. The position needs to be approved by the board.

5. Once and if the position is approved, an application process shall occur.Since the company is a non-profit organization, the company needs to find theperson most suited to fulfill that role. As in the case of “code bounties”the position should be made public, so everyone is informed about the jobopportunity. It is of the outmost importance that the organization hires theperson most qualified for the job.

6. The position shall be advertised in and by all reasonable channels. Inparticular, the board should advertise the position in channels where the mostdesirable applicants may hear about the position.

7. A reasonable period should be given for the position to be reasonablyadvertised. Once the period elapses, the board will review the applications andchoose the best candidate available to fulfill the position.

5 Economics (jac)

Personally, I am hopeful, that in a time when “Y-Combinator” isbecoming a household word (see Newsweek Magazine, May 21 2007), thereis money out there to support even some of our most outrageousschemes.

But in this case, I think good things do not come to those who wait –they come to those who ask. And to ask a good question, you need agood formulation of the question, hopefully together with anillustrative example, model, or sketch.

5.1 Free/Libre/Open {Software, Content} Business Models

In case you are getting the name of this section is a bit awkward, Iagree! I think people in the world of Free/Libre/Open Source Softwareand content (acronym: FLOSS) have not found very clear ways to talkabout the economics side of what they do.

At the same time, there is at least one “economicsy” term that is inuse to talk about the way people produce this stuff. The termis “Commons-Based Peer Production” (acronym: CBPP). The connectionto economics is the suggestion that, for participants, their voluntarycontribution to the commons is a rational behavior.

Whatever else the analysis behind this term may get into, I don’tthink it has much to say about the ways in which commons-based systemsrelate economically to other systems.

I would like to simplify matters a great deal and point out thatfrequently the things that these systems are generating havepublic good-like aspects from the point of view of agentsoutside of the system.

This is important, since it says something about how money might beexpected to change hands to support the projects. Specifically,public goods tend to be paid for, and regulated, by governmentalbodies.

It would be great to get into a further analysis, but for now I amjust going to list a few words of wisdom on the topic of FLOSSbusiness.

• “You can’t sell something you’re giving away for free.”

• “The best way to make money related to free content orsoftware is to sell services.”

• “There are no restrictions on selling copies of freecontent or software.” – this is relevant to, e.g., the FEM, if weever get that together. (However, there may in fact be somerestrictions coming from our legal status as a charityorganization.)

5.2 Semantics of Cash Flow

(This might also be called the Cash Flow of Semantics. I haven’tgotten all the details worked out yet.)

These “brainstorms” are a tentative sketch, based on my thinkingabout Arxana. For some further thoughts along these lines, see myessay on political applications of the scholium system.

• Every connection, every link, is an arbitrage opportunity. Themoment I realized that, I was pretty excited – since a lot of whatI think about is links. This is why Google Answers worked (for awhile) – but since they stopped working, we have to be careful.Not every arbitrage opportunity is “viable” and one can fail tomake money for myriad reasons, whereas one must wait until the checkclears to actually make any money (and, OK, even then, there areexpenses to worry about!).

• By the time a link has been made, you might say that thearbitrage opportunity is already gone – but despite that fact, anew link may open up a whole “plane” of new arbitrageopportunities in the form of potential links, i.e. annotations,running perpendicular to it.

5.3 Ways to Make Money

• One idea is to get people to pay us to customize Noosphere; butan easy complaint is that we have not been able to pay to customizeNoosphere in the directions we want. The synthesis is to seeksynergistic relationships – customizing Noosphere in directions wewant, and having someone else pay us to do that!

• There are a number of similar ideas related to “other” or“next-generation” software systems we have been discussing(Arxana,CodeMarket).Given that a number of our proposals fall outside of PlanetMathproper, I think that we would do well to put together an“über-administration” that covers all of the things that wemight want to share work on.

For any project or scheme for making money, I think it would beworthwhile for us to come up with what’s called a“value-proposition” (or, in some cases, an atlas of suchpropositions).

6 Governance (jac)

6.1 Bylaws (jac)

The organization was incorporated with bylaws thathelp it meet the legal requirements of incorporation.These bylaws were revised.

(I’m not completely sure whether the second draft of the bylaws wasever really “put into effect”.)

6.2 Board of Directors (jac)

The role of the board of directors is described in the bylawsmentioned above. However, it is not completely clear that the boardhas been following the protocol described in these bylaws.This suggests that a review and possible a reconstitution ofthe board and of the protocol is in order.

6.3 Member Rights (Wkbj79)

Members, especially new members, should be aware of what their rights are. One aspect that instantly comes to mind is freedom of mathematical expression. After all, PlanetMath is supposed to be “Math for the people, by the people.” A good example of freedom of mathematical expression is the Smarandache entries. A poll was taken on PlanetMath regarding such entries, and although most active members here have doubts about the mathematical significance of the Smarandache entries, the decision to keep them here was the right one. If this were a rightthat were clearly expressed, then such a poll would not have had to be taken.

7 Processing Results of Words on Fire Consultation (jac)

The new info we got from Words on Fire will certainly need tobe processed somewhat; and indeed, following through on this is certainlyan administrative function. At the same time, some of this contentmay be worked into the other sections of the current document.

7.1 Mission (quoted form WOF)

PlanetMath.Org’s mission is to build and sustain an engagedcollaborative math community that is free and open for all individualswith a serious interest in mathematics.

This is the jumping off point for your mission. Before you can fullyarticulate what you do, you have to agree on this basic element of theorganization. Everything you do flows from and to your mission.

7.2 Work Plan

Some version of this is in the WOF documents, and Aaron is currentlyputting together a working version this work plan.

8 Technology (jac)

8.1 Current Technology

8.1.1 Noosphere

PlanetMath runs Noosphere. Understanding whatthis means is in some sense an administrative problem – and may alsorequire some data archeology.

8.1.2 LaTeX

Noosphere is integrated with LaTeX.

8.1.3 latex2html

This is the current rendering engine for Noosphere.

8.1.4 Wiki

A goodly amount of project-coordinating information is available onthe AsteroidMeta wiki, althoughthe wiki is not in very good shape and will hopefully be replaced soon.(An effort to do this is underway thanks to effort thanks to theGhestalt team.)

8.2 Current Technological Efforts

8.2.1 Coding

Right now the primary coding efforts are related to work being done onGoogle Summer of Code (see the wikipagefor basic coordination information, and the project proposalssubmitted by our interns for further details).

• NNexus is being extended

• MUSNis being developed

• a ratings/reputation system is being developed

It would be nice if more communication was taking place betweenpeople working on code and other members of the community.

8.2.2 System Administration

As mentioned above, this is currently routed almost completely throughAaron.

8.3 Ideas for Technological Changes

There are many ideas for features around, includinga recent compilation in another collaboration documenton PlanetMathand the huge ongoing feature request list from the wiki.

For the current document I am going to mention some of the ideas abouttechnology that I think are most related to “administration” in thesense of this document.

8.3.1 Issues

This is a short sequence of points that develop a basic theme.

• Keeping track of things. On all of the differenttask-areas mentioned here, it would be great to know who is doingwhat. Similarly with mathematics items in the encyclopedia – whichitems have been modified recently? Linked to recently? Viewedrecently? What, in general, is going on? – The question seems toogeneral, but in an administrative setting, we should be try to bemost aware of the things people are most concerned about. What arethe biggest annoyances? What are the best hopes we have for doinganything about them? And so forth and so on.

• The multidimensionality of information landscapes. Wemight be interested in integrating data from some field ofinformation, and we could put this data into some sort of aggregator(see below). But then, when we have a collection of aggregators,we’ll want to see which ones people are using. And if Alicerealizes no one is addressing issues of concern to her, then she’llhave to think about which factors are relevant here. Why is doesshe care about something other people aren’t interested in? Ormaybe other people are interested, and she is not using the rightset of aggregators. Or maybe the aggregators are not pointed at theright dataflow. In either event, she will have to debug her ownthought processes to find the answers – which will require otherintrospective data aggregators, and probably some field research.As the cycle continues, there’s more and more information to manage!

• Network effects. Is any given task ever completedperfectly? Or will there always be problems that come up? Isuppose it depends on how your point of view; I suppose my point ofview is that these actually aren’t very good questions. In thephysical theories, systems flow towards a lower energy state. Theydo not get to their destinations instantaneously, nor do they traveldirectly to the lowest energy state. To call this either“perfect” or “problem-ridden” seems to stretch the languagepretty thin. In human affairs, the things we call “problems” areoften opportunities for greater efficiency that we notice. A quitegeneral challenge lies in the fact that many interpersonal ortechnological arrangements could be made more efficient in manydifferent ways. Any given event or class of events might havenumerous criticisms associated with it – indicating improvementsthat can be made along many different dimensions. The first step infacing this challenge is to find a way to get these critical remarks“networked in”, so that they can be processed further and notforgotten or neglected.

8.3.2 Proposals

• The basic interface is might be some collection of data“aggregators”. Just the other day I was looking at the PlanetMathhomepage, and thinking about how it would be nice to have it workmore customizable (like Mac desktops or Google homepages… orGNU/Linux systems!), allowing users to select or even create newwidget-like tools to inspect the data flows they are most interestedin.