Thursday, August 28, 2008

If a Hobby relates to your work, is it a Hobby?

In yesterdays blog Lance mentions that he didn't see other math people or trade problems when he was on vacation, and that it wouldn't have been a vacation if he had.

This raises a question: Can you ever use a hobby in your work. I give a few examples below.
  1. Steve Skiena wrote a book Calculated Bets about mathematical modelling and betting in sports. This combined Steve's lifelong hobby of watching and betting on Jai-Alai with serious math and computer science of interest.
  2. Ken Regan's hobby is chess and he has some research there, as seen here.
  3. My hobby is collecting novelty songs and I've gotten some blogs out of that. (Is blogging part of my job?) Other bloggers also use their hobbies on their blogs-- Luca and photography comes to mind.
  4. Steve Rudich's hobby is magic, and I've seen him entertain at conferences. Some of the magic is based on math.
  5. Lance has commented on the show NUMB3RS in the blog, and I wrote a reviewed Season 1 my SIGACT NEWS column. (If I buy the DVD and review it again, is that tax deducbible?)

If your hobby is some game then there may be some math of interest surrounding the game that you could get involved with research on. If your hobby is collecting paintings then you may have a harder time connecting it up with your work. (Maybe Game Theory of Auctions.) Do you even want to connect up your hobby with your work? Hobby is almost defined as not work.

Since most of us (I assume) like math, the line between hobby and work may be hard to see. If I work on Math problems in my leisure time, is that a hobby or work? If one of them inspires a paper then does it cease being a hobby? How does Sudoko fit into this? If I actually use the ZK Sudoko protocol to convince someone that I solved it, is that work or play?


  1. What's the deal with airplane peanuts.

  2. A hobby requires freedom. How closely a job relates to your hobby depends on how much freedom you have in your job.

    If you are self-employed, then your work can be hard to distinguish from your hobby.

    In academia, the situation is a bit unclear since grants require peer review.

    Peer review takes away some of your freedom unless perhaps you are very famous and people let you do whatever you want.

  3. What about considering that work is what your are forced to do, and that whether you like it or not, it does not matter?

    Everything is based on definitions...

    Oh, and there is a lot of fun to get with maths about wikipedia, too.

  4. Bill, there's a word for people like you... Actually several, like nerd, geek, etc.

  5. What's with the binary 0/1 model of work/hobby? Something can be work and a hobby, or neither, or partly both, or ...

  6. Thanks for the unexpected mention. In my case the "hobby" has morphed into real research, and in three separate ways that are also different from computer-playing. The above link is to my "for-chessplayers" page (where I need to update/fix the Kasparov-World material)---here are direct links to the others:

    (1) My anti-cheating work here is now part of a larger project to assess performance intrinsically from move-choices rather than from results of games. In chess it aims to answer controversies such as whether Elo ratings have "inflated", "deflated", or stayed the same over 40+ years, but the methods are chess-neutral and aim to quantify general decision-making activities. The stats program I described at the Complexity open session has since grown from 3000 to 8000 lines of C++ code, and I hope shortly to make it and a draft paper available---the program will also exemplify this blog's June "why C++?" item.

    (2) Here and more recently here are examples of reproducible hash-table faults common to quite a few of the major chess-playing programs, several of which I know to generate their forever-fixed hash-key bits (called Zobrist keys) pseudo-randomly rather than by grabbing 50,000 bits from here or here or here. The lure is whether the chess programs can be abstracted into an asymptotically-exptime but concretely-efficient discriminator of pseudo-randomness---not by the mere frequency of collisions (which is only a linear test) but by their interaction with depth-first search. In cases I've demo'ed but not yet posted I've isolated faults to hash-table size, including one where telling a program the "50-move-rule horizon" is 49 vs. 50 moves away makes a wacky 6.5-pawn difference in its assessment, the difference reversing when hash is set to 32MB not 64MB.

    (3) Whether an NxN chess position wins for the player to move is canonical-complete for PSPACE or EXPTIME, depending on whether a generalized 50-move-rule is in effect. Scott Aaronson built this into his prime example `to explain complexity to laypeople' here. Concretely, my work after being "called out" by the world's premier chess blog on the instance here (with only 12 pieces!) has grown past 500 pages of analysis, and even after importing half a terabyte of precomputed 6-piece results I still don't know the answer! But I've found many testbeds for the question of whether BPP search heuristics (of "evolutionary" flavor) can replicate parts where my analysis is definite---some I say yes, others "no way" even if quantumly D-Wave's stock hits $1,000.

    Certainly (2) and (3) hit the heart of our field, but in real time I've focused on getting to the plateau on (1)---while also trying to keep up with my "real work" :-).

  7. I thought this was an interesting question, so I decided to do a quick Google search to find out how the IRS determines if your work is a business, or just a hobby.

    I found an article here which indicates that if you make money on your hobby, then the IRS calls it a business, however if you lose money, then you have to take special steps to get them to believe that it is a business before they allow you to deduct the loss.

  8. Actually, if you make money from a hobby, you have to report the income to the IRS. However, it is not business unless it satisfies their definition of "business."

    "An activity qualifies as a business if your primary purpose
    for engaging in the activity is for income or profit and you
    are involved in the activity with continuity and regularity.
    For example, a sporadic activity or a hobby does not qualify
    as a business. To report income from a nonbusiness activity, see the instructions..." - From the 2007 Instructions for Schedule C