Q: How many programmers does it take to change a lightbulb ?
A: Trick question. Programmers dont do hardware. (same answer really as None. Its a hardware problem.)

Q: How many Americans does it take to change a lightbulb ?
A: 250,000,000, one to change it and 249,999,999 to debate whether it it was politically correct.

Ten, one to do it and the rest to dance around, play the
tambourine, chant, and sing.

Q: How many SAS programmers does it take to change a light bulb?
A: One to analyze the historical failure rates of lightbulbs using PROC LIFEREG, so as to anticipate the failure of the lightbulb before the user actually has to report it, one to explain why SAS is better for changing lightbulbs than S-Plus, SPSS/X, BMDP, SYSTAT, MINITAB or a spreadsheet, one to write a custom interface in AF/SCL allowing the user to manually request the changing of the light bulb after its failure (prematurely) occurs, one to write a report with PROC SQL and PROC REPORT which will summarize the lightbulbs needing to be changed, sorted twelve different ways, cross-indexed (by wattage, type, and prematureness-of-failure) and totaled, one to actually spin the light bulb into the socket using SAS/Insight, one to call Cary to try to get them to explain when a new version of the lightbulb will ship, how much well pay to keep using lightbulbs for another year, and what well do if our site sends all its lightbulbs to Europe where 120V/60Hz lightbulbs tend to explode upon insertion in 220V/50Hz circuits, one to write an incomprehensible ten line SAS macro program which will perpetually insert new filaments into all mission critical lightbulbs until its author is fired, at which point the SAS macro will automatically encode itself into a copy of the latest SAS/ETS usage notes, one to write a graphical front end to the lightbulb changing process using SAS/EIS, with little speedometers showing the number of lightbulbs changed per hour, so that management can understand why we need to buy bigger lightbulbs, one to prepare a SUGI paper summarizing the entire lightbulb project, taking credit for the design and execution of the lightbulb project itself as well as the invention of the light bulb itself, another one to prepare a second SUGI paper benchmarking lightbulb replacement on twelve different types of light sockets, with separate graphs for florescent and incandescent bulbs (made with SAS/Graph, except for the titling, axes, color, polylines, and background, which were all added manually with Cricket Graph on a Mac), ten to push the dollie loaded with SAS/Lightbulb manuals, *and*, One more to ask SAS-L for help when you really need to change that bulb, NOW.

How many professors does it take to change a light bulb?

Only one, but they get three reports out of it, each of which will count as 10% of your final grade.

How many Einsteins does it take to change a light bulb?

That depends on the speed of the changer, and the mass of the bulb. Or vice versa, of course. Then it just might be easier to leave the bulb alone and change the room. Its all relative.

Q: How many GLC workers does it take to change a lightbulb ?
A: Four. One to do it and three to go round putting up posters announcing that the GLC, working for London, is going to change the lightbulb.

Q: How many Prolog programmers does it take to change a light bulb?
A: False.

Q: How many Macintosh users does it take to screw in a light bulb?
A: None – theres no documentation available, so you have to wait until a third-party supplier comes out with a solution.

Q: How many Pisceans does it take to change a lightbulb ?
A: Huh? The lights out?