Ingenieria

PlanetMath SWOT: Strengths, Weaknesses,
Opportunities, and Threats
Joseph Corneli
March 6, 2010
I. Strengths

mathematics content on the web”, without doubt, the content we have is an asset. For comparison’s sake, there are
approximately 23488 mathematics articles
on Wikipedia2 and Springer claims to
deal with more than 50000 concepts in
their online encyclopedia of mathematics.3 Still,PlanetMath tends to have a reputation for providing in-depth treatment
of topics that others may deal with more
superficially. (It would be nice to back this
up with some solid evidence!)

Community and visibility I’ve estimated that PlanetMath users have contributed up to $0.75mil of writing.1 During
that same time frame, PlanetMath’s board
members and major monetary donors
havecontributed about $0.25mil towards
operational overhead.
In the same period, from 2005 to 2009,
we’ve recieved an estimated 30 million
hits. If each hit was valued at $0.50 (an arbitrary figure, chosen to match the value
of one Point in the estimates given above),
then we’ve delivered $15 million worth of
mathematical content, at a cost of about
$1 million.
Apart from this back-of-the-envelopestuff, we’re quite popular: currently the
#2 hit on google for “math encyclopedia”
and #5 hit on google for “mathematics encyclopedia”. We show up on the first page
of search results for “math”.

Free/Open Anyone can contribute to
the project, and anyone can use the things
the project produces.

II. Weaknesses
No command-line or email-based access It may seem like a minor point, but
for peoplewho are familiar with mainstream open source software development patterns, the restriction to a web
interface is a big annoyance. Lack of
email integration in the forums leaves us
lightyears behind platforms like Google
Groups or Posterous in terms of useability and features.

Math-specific
software PlanetMath
A X source code and
works well with LTE
has
state-of-the-artmath-on-the-web
rendering, making it an attractive option
for mathematicians.

Content and coverage We currently No access to the actual data Planethave 8786 entries dealing with 15342 con- Math’s databases are not public. There
cepts. While we don’t have “the most are presumably many ways that the interested public could add value to Plan1
That estimate was made about a year ago; etMath’s data, if thisdata was available.
the same instrument today suggests that our total
2
http://en.wikipedia.org/wiki/Portal:
value has depreciated over the past year, to something more like $0.6mil. See http://metameso. Mathematics
3
http://eom.springer.de/
org/~joe/docs/transparency.pdf for details.

1

Even though we provide tarballs of content, our lack of procedural openness
compromises ourextensibility and even
our legitimacy as an “open” project.

the forums, on the wiki, and at our rare
in-person meetings – but many of these
ideas are now languishing. Various other
related projects (e.g. SAGE), both large
and small, have successfully used ticket
systems and so forth to sync requests to
development. (Note: this critique of PlanetMath’s workflow covers nonprofit management, codedevelopment and math
content development.)

No objective standards for quality Although a tool for rating articles was developed and deployed on the production
server, there is no easy way to poll PlanetMath to find out more objectively useful things like “how many theorems are
illustrated with examples” or other simple
structured queries. This is a sign that our
data model may not besufficiently rich.
(And, if these questions are possible to
ask in the backend, the methods have not
been exposed to the public.)

Money-poor When critical comments
about development come up, we remind ourselves that PlanetMath is almost
ridiculously poor in terms of financial resources. Many of the people involved with
running the project are also quite short on
No strategy for growth and...