|
|
Viewing Message
|
|
|
| ``some important things to notice about PlanetMath''
by jac on 2009-05-21 20:53:51 |
|
| These are some of the things that I would say if I was giving someone a tour of PlanetMath. I'm sure there are other important points -- but these seem to be the key issues (both areas that need improvement, and some of our strengths).
We have fewer or less prominent "general" articles than some of our competitors:
http://www.google.com/search?q=planetmath+spectral+graph
We have OK, but somewhat less descriptive articles about basic objects of mathematics (although perhaps our articles are a bit more mathematically precise):
http://planetmath.org/encyclopedia/LaplacianMatrixOfAGraph.html http://en.wikipedia.org/wiki/Kirchhoff_matrix
Our new jsMath rendering mode doesn't always work; our HTML with images mode isn't necessarily tuned for maximum readability.
http://planetmath.org/?method=js&from=objects&name=Orientation3&op=getobj http://planetmath.org/?method=l2h&from=objects&name=Orientation3&op=getobj
(compare with the darker and more readable text on the Kirchhoff matrix page at Wikipedia).
Our code repository is out of date (but I think this in the process of changing, now that the data is being recovered from our old server's drives -- although this does make me wonder whether code development has been following "best practices" for an open source project in the mean time):
http://aux.planetmath.org/noosphere/snapshots/
Although we do have a good way to look at "related" documents (through "See also" links, Attachments, and Parent links)
http://planetmath.org/encyclopedia/InfiniteProductOfSums1a_i.html Child: http://planetmath.org/?op=getobj&from=objects&id=11425 Parent: http://planetmath.org/?op=getobj&from=objects&id=4368 Grandparent: http://planetmath.org/?op=getobj&from=objects&id=4230 Cousin: http://planetmath.org/?op=getobj&from=objects&name=InfiniteProductOfDifferences1A_i
there is no easy way to use all of this information to gather the related articles on a topic into a "course packet" that one could use to *learn* the topic. (Such course packets would also provide us with a way of deciding how complete the treatment of a topic is -- e.g., are we missing overview articles? Are we missing useful basic examples? -- If we could gather articles into course packets, it would be easy to see what's missing!)
On the plus side, detailed information of this sort is almost wholly missing on our competitor's sites! It's long been noted that detailed coverage of proofs is PlanetMath's strong point. I'm inclined to agree -- but perhaps we're not that good at showing off or using this strong point.
Feature requests frequently sit around for a long time, sometimes the code is written but the feature is not installed or not documented.
http://wiki.planetmath.org/AsteroidMeta/email_posting_to_forums
Our list of donors is small (and so we don't have a huge budget to go into improving anything on the list of issues I've brought up above):
http://aux.planetmath.org/doc/donors.html
(This may change at any point in the future -- and our new "your logo here" banner may help leverage our web presence to invite large donations and important partners.)
Many ideas about how to improve things have been talked about by members of the community here, but they are not particularly well integrated with the "workflow" on or around PlanetMath, just for example, this project I've been working on for quite some time:
http://common-lisp.net/project/arxana/
It has frequently been suggested to me by my dad that we should try to get PlanetMath plugged into the One Laptop Per Child program,
http://laptop.org/en/vision/mission/index.shtml
But frankly, I don't think we're ready yet, despite the fact that "improving, porting, and developing math and science activities for students and teachers" is in demand in that program.
http://wiki.laptop.org/go/Contributors_program
A key ingredient for readiness to use PlanetMath as an educational platform would be for us to have *problems* that can either be graded automatically or by a teacher or self-graded (by comparing one's own solutions with archived solutions). We don't have anything like that currently. This would of course be useful to math learners at all levels and socio-economic classes! |
| | [ reply | up ] | |
|
|
|
|
|
|
|