PlanetMath Q2 2012 Board Meeting - April 20th, Noon Eastern, 5PM UK, 6PM Germany

[holtzermann17]

Minutes

Status Reports (current progress)

Major change: now using github for all organizational docs and most org communication

• Github is like the new "PM-official AsteroidMeta"
• need Aaron's login info

Summary and discussion of the Advisory Board meeting

Present at the advisory meeting: Joe, Ray, Michael, Jamie, Lee. Notes taken during the meeting are here: http://piratenpad.de/p/planetmath-advisory-meeting. Some outcomes from continued work on the themes from the meeting are in the comments on #22. Quick overview follows:

• Education; what makes PlanetMath special?; etc.
• Licensing: most working mathematicians aren't
  going to be that concerned, timestamps etc. are
  likely to be enough, frames...
• Previews: Clarify what PlanetMath is so we can get people involved, e.g. how can people help out financially or materially?
• "PlanetMath is...": PM Previews #34 (comment) (Example: the Books Preview, in Books Preview #37.)
• Interns or grants would do a lot - bring up w/
  Jon?
• Hub: how does PlanetMath fit into the landscape?
• "Math on the Web prototyping center?
• "premium" membership in PlanetMath.org? (we need to build the "premium" features)... see plan a set of online services to complement existing offline offerings #18.
• FKI donations in Europe? @jac ask Wouter, follow
  up about other European donations
• When is the next meeting? (Get back to them
  after this meeting.)

Status report on beta testing

• starting this week, good opportunity to think about what
  people want from the site, will go into my thesis which will
  be wrapped up this month
• NIST mathematicians help? @jac: email Deyan, Pierre, Heow? -
  talk to Ian Short as well

Status report on NNexus

• this has been a good week for NNexus development: working
  rewritten system, does what it used to do before
• Allows links and "stand-off" annotations (e.g. useful for
  editors)
• Snapshots of PlanetMath, Wikipedia, DLMF ready to index ;
  could potentially use MathWorld's OAI as a source of data?
  (Maybe more convenient than crawling?) Currently there's a
  common framework for crawling.
• Right now everyone has an MSC classification.
• User documentation for NNexus? User docs are starting to
  appear on Github
• Old code for link-steering by Math Subject Classification is
  in the Noösphere codebase, Aaron could potentially retrieve
  that (If there's more than one match, go into a sub-routine
  to match.)
• Anywhere where you can map things to MSC you could use
  this
• There's a Jacobs student working on similarities between
  MSC classes - you could use this stuff for clustering of
  the links
• Can you make it so that you can plug in various
  link-steering mechanisms? Yes.
• Integration: plug into PM, ability to turn off this
  routine entirely for forums
• NNexus is ready to be a CPAN module, which is one step
  away from being a Debian distribution

Action Plan (stuff that needs to happen ASAP)

• Need to build a user community around the site - or at least need a plan for that, see
  MINIMAL PLAN FOR RELAUNCH #36
• during or prior to the meeting: make sure that everyone has
  a github acc't and knows how to make changes to the PlanetMath
  docs repo
• "Summer of PM" - can we recruit and mentor volunteers? Or scrape
  for funding?
• What are realistic things to expect about the magnitude
  of the dev effort for 2013? (Let's also define realistic
  in a sensible way.)
  - Joe can approximately keep up with the PlanetMath
  workflow
  - In order to do what Jamie needs we could help - well, he
  knows we're pretty much maxed out, and applied for money
  to help with development...
  - But concretely, it might not take THAT much work to set
  up something useful; maybe the best thing to do here
  would be to make a "PM Preview" for the features that
  we think Jamie wants, and share it with him.
• Who are we going to work with that might be interested?
• An open source style collaborative community - identifying
  people to make the pitch to... "the human element"
  - This is related to the "bundle" and "umbrella" ideas
  below

Entering the "Grand reopening" milestone

See #36 for details on a minimal plan; overview below

1. conclude a successful beta trial - "Program functionality"
   10-15 hours of programming time from Joe
2. other time from other people will help
3. update site documentation - Github is working for this
4. Ray is working on this, Joe is helping, beta testers
   will help as well
5. Deyan adding NNexus docs
6. community guidelines - will also go into the Github repo
7. Ray is working on this
8. Joe and Ray are admins
9. Dropping things directly from the database - this is OK
   for anything like spam or real problems
10. "Unpublish" any articles that need more processing
11. "Community moderation" - is a good aim, but more useful
    once we have the user community
12. initial content for problems and solutions
13. Ray: put Calculus book and 3 Project Gutenberg books,
    using collections features to add the books and
    associated problems things
14. Joe: Long term plan is more automation of this process
15. legal documents
16. Appropriate language for Common Carrier
17. Copyright officer - copyright and takedown language
18. New user agreement
19. @aaron will take care of this with Ray's help
20. inviting donations
21. @joe: Patronage model
    1. Deyan: intersection between bitcoin and math
    communities is probably quite tiny to deserve
    special attention
22. Can put a donate button right away
23. "Send check to Aaron"
24. PM Previews
25. appearance
26. Practice PHP problem for Ray
27. By next year maybe a whole new theme for the website
28. DG: also clear-cut Github milestones and timeframes
29. Print, circulate, and share with new devs...
30. maybe put on the website
31. advertising campaign
32. Still have flyers to circulate, can send to mailing list
33. Think more about this at next meeting
34. Announce at MathFest in August
35. install bibliography system - let's make a ticket about
    this, install library of congress, add copyright renewal
    tags
    1. Ray: Can install Jim Pitman's system in the next few
       months at least on an experimental basis
    2. populate with LoC QA section metadata (1GB of data)
    3. Add field(s) for copyright renewal (need to document the
       complexity, e.g. walk through one or two items so that
       people can contribute this info)

PM.org as an umbrella organization for PlanetPhysics and others

• NRI (pretty much the same people who've been involved with
  PlanetMath, doing similar projects; name PM as the successor
  organization), PlanetPhysics (hosting?), PlanetComputing,
  PlanetTechne, FTG Basic Platform
• Dissolve NRI and potentially gain any assets from the
  late Ion Bianu
• if we can sort out the hosting issues for PlanetPhysics,
  that would be good (if we're doing the hosting, we gain
  a bit of useful control)
• http://cicw.culturecraft.cc
  - note this concrete example of a project using Drupal,
  it would be good to stay in the loop with this and other
  projects to learn about features we'd like to use
• Passed, and resolution is rewrite this into our org
  docs
• Indie Bundle? Math Bundle? (SAGE etc.) see
  http://piratenpad.de/p/indie-edu-bundle
  (We could re-run the "bundle" within PlanetMath on an
  ongoing basis, assuming we have things we can provide.)
• Try the math people for comparison, see: http://wiki.planetmath.org/cgi-bin/wiki.pl/Surveying_the_Math_Metacommons
• Joe can report back in about a quarter about where he'll be
  this year?

After Meeting TODO items

Double check that any items from this agenda that need further work have been turned into tickets on Github.

1. Let's double check: what is the state of developer docs in
   Planetary and how easy is it for new people to jump in the
   project (or just install the system and use it for their
   site)

• at the moment, developer docs are just collecting in
  this issue:
  developer docs MathHubInfo/Legacy-planetary#88
• get an IRC channel set up for dev and community use,
  advertise to other Planetary developers and on
  PlanetMath website -- irc://irc.endernet.org, channel
  #planetmath
• Deyan: "for the record, I would be quite interested in
  doing Planetary hacking if there is activity in an IRC
  channel such as this one, where I can discuss where I
  get stuck."
• Let's make sure that the dev experience is as fun and worthwhile as possible for anyone who does want to join us!

2. Ray and Joe have moved their weekly voice chats to
   Thursdays, 6PM UK time - info about how to join the meetings
   is here:
   http://piratepad.net/paragogy-mumble - let's publicize that
3. Slight computer shortage: we can use li101-104.members.linode.com (which is where the backups go to) and li311-58.members.linode.com (which is where the current PMDEV site is) but we do need to do continual housecleaning on these small virtual machines
4. We also need to do some housecleaning on the U Waterloo
   server to free up space
5. Someone should turn this into a report for the PM Advisory
   board
6. Joe should set up a meeting with Jon sometime soon
7. Remind Aaron to create/share github username

Future Priorities (wishlists and mid-term goals)

• Using PM in teaching effort (now or postpone to 2014)
• Follow-up on DLMF and PlanetMath interaction
• Interested in adding interactive services, like "click on a
  formula see links to other services", can see links related
  to formulas, discussions related to formulas...
• PlanetMath and Planetary might be a good place to start -
  get one BA student for 3 months in the summer, will work
  with Deyan and Bruce's supervision
• No production site deployed, mostly Javascript stuff,
  Michael has someone working on JOBAD during the summer,
  hopefully they will be in touch with each other
  - Link to JOBAD repository:
  http://jomdoc.omdoc.org/wiki/JOBAD
• Someone from some Virginia University
• Funded + Supervised student
• Places to talk about the ideas? - Github Trac
• Interacting with external content via frames

For the record: our Initial Agenda

This integrates Ray's email and Deyan's points as an agenda suggestion:

Live-editable version: http://piratenpad.de/p/pm-board-meeting-spring13

1. Status Reports (current progress)

   • Summary and discussion of the Advisory Board meeting
     • Some outcomes from continued work on those themes here: build at least some basic documentations about how to get involved in dev work (basics of the FTG platform, "indie edu bundle", etc.) #22 -- it would be great if people could review that before the board meeting.
       • In particular, that issue talks about a «"premium" membership in PlanetMath.org (we need to build the "premium" features)» -- let's think some about what people will get with site membership. Indeed, there may be a whole range of things, see plan a set of online services to complement existing offline offerings #18. Ray made related points in his "PM Previews" proposal, which I've copied into PM Previews #34. (Example: the Books Preview, in Books Preview #37.)
   • Status report on the beta testing
   • Status report on NNexus

   ---

2. Action Plan (immediate next steps)

   • during or prior to the meeting: make sure that everyone has a github acc't and knows how to make changes to the PlanetMath docs repo
   • "Summer of PM" - can we recruit and mentor volunteers? Or scrape for funding?
     • What are realistic things to expect about the magnitude of the dev effort for 2013? (Let's also define realistic in a sensible way.)
   • Entering the "Grand reopening" milestone; see MINIMAL PLAN FOR RELAUNCH #36 for details on a minimal plan.

   'Minimal Viable Product' (MVP) goes hand-in-hand with the idea of fast lean production. It is not a wireframe website or a assemble-it-yourself cellphone. It is simply a product with the minimal set of features that accurately solves customers' problem and provides enough added value to make it competitive.

   1. conclude a successful beta trial
   2. update site documentation
   3. legal documents
   4. initial content for problems and solutions
   5. DG: also clear-cut Github milestones and timeframes
   6. advertising campaign
   • PM.org as an umbrella organization for PlanetPhysics and others
     • NRI, PlanetPhysics (hosting?), FTG Basic Platform and PlanetComputing
     • Indie Bundle? Math Bundle? (SAGE etc.) see http://piratenpad.de/p/indie-edu-bundle (We could re-run the "bundle" within PlanetMath on an ongoing basis, assuming we have things we can provide.)
   • Build a board consensus on an action plan
   • after meeting: Double check that any items from this agenda that need further work have been turned into tickets on Github.

   ---

3. Future Priorities (wishlists and mid-term goals)

   • Using PM in teaching effort (now or postpone to 2014)
   • Follow-up on DLMF and PlanetMath interaction

[dginev]

Something that just occurred to me - we can keep meeting minutes as comments and/or direct edits to the ticket body, then we have the full discussion archived here.

[holtzermann17]

@dginev I agree. Partly because the number of issues is looming, I'm interested to discuss a bit more: "What are realistic things to expect about the magnitude of the dev effort for 2013?". I've made some preliminary efforts at "social engineering" that might get some buy-in about the idea of a shared platform, and eventually some funding - but that's definitely the "long way round" (#22). The shorter way is to just get on with the dev work, and, where possible, help train other PlanetMath contributors on the basics of helping out. But I don't know if this train-in process is really realistic or not.

[rspuzio]

Entering the "Grand reopening" milestone:

One aspect of this should be making a "minimal plan" which we should be
able to carry out using current resources (i.e. 4 of us working part-time). The
idea is that, while there is plenty to do and we can always do more, we should
make sure that the basic things that need to be done don't get neglected or
put aside indefinitelty.

[holtzermann17]

Let's check whether or not the prototype plan sketched in the issues X milestones X people space here on Github is going to get us where we need to go. One thing that was missing in the initial sketch was "issues assigned to people". Let's get that sorted out by the time we have the meeting.

[rspuzio]

Here are my main impressions from the advisory board meeting:

As an initial meeting it was successful, but we will need further discussion.

I have concluded that we should not support "no derivative" content
for the following reasons:

1. The typical mathematician isn't that concerned about details of licensing and
   generally will be content with the default as long at it isn't completely wacky.
2. The combination of attribution licensing, timestamped archiving, and only
   logged-in user with accounts being allowed to post material (all of which we
   already have) should suffice to take care of plagiarism, so "no derivatives"
   terms don't add anything significant.
3. Should we want to do something with "no derivatives" content coming
   from some third party, we could use the frame system Deyan's came up
   with for NIST content.

By and large, we and our advisors agree as to what to do, the problem is
resources. Kohlhasse et al., of course, will continue work on Planetary,
which is extremmely helpful and important. However, the only other suggestion
we got was to try to recruit volunteers. While we should do our best to
interest and retain volunteers and make clear that we are interested in
having volunteers for certaini tasks, I am not so sure that that is enough.

Thus, I think that, in future interactions, we should make it clear that this
lack or resources is the limiting factor. With our current shoestring budget
and handful of volunteers working in spare time, work on implementing
new features and taking PM to the next level is creeping aliong slowly.
Any short-to-medium-term help in the form of, perhaps grants or student
interns or joint projects or development contracts or the like would go a
long way towards moving the process along.

[holtzermann17]

Another nice thing to do soon would be to create a "planetmath" organization on Github, and assign this repository to that organization, which will make it feel more official.

[holtzermann17]

This happened, and I think a lot got resolved! Anything that hasn't been turned into do-able tickets should be. But this ticket can close.

[holtzermann17]

So these don't get lost, some links and comments from Ray:

Following up the meeting, here are some links related to
items we discussed there:

Draft of New User Agreement

http://wiki.planetmath.org/cgi-bin/wiki.pl/Second_Draft_of_Pact

Discussion related to the issues behind the agreement

http://wiki.planetmath.org/cgi-bin/wiki.pl/PlanetMath_Copyright_Indemnification

Good Questions to ask of a Good Lawyer

http://wiki.planetmath.org/cgi-bin/wiki.pl/Good_Questions_to_ask_of_a_Good_Lawyer

(Reminiscent of what we're doing with the Advisory Board, but for legal issues.
At some point, perhaps we could recruit a copyright expert to our advisory board.)

BibServer home page

http://bibserver.org/

Documentation home page

http://bibserver.readthedocs.org/en/latest/

Their github page

https://github.com/okfn/bibserver

Installation and dependency instructions

http://bibserver.readthedocs.org/en/latest/install.html

Also, note that Pitman et al. were funded by Mark Shuttleworth. Perhaps
we should look into the possibility of getting some funds from there in
connection with integrating BibServer into Planetary when we talk to Pitman.