Marcus Berger wrote: "Say a worker spent 1h25min on a job which pays him 8 $/h. How much money should he be paid for that amount of work?"
That's easy -- $12. It's unlikely that his/her pay would be calculated to a resolution of 5 min. A more likely resolution is 0.5 h or 0.25 h (with actual time worked rounded up). Costing a job for accounting purposes is, of course, another issue -- probably not involving manual calculations. In any case, there's only one calculation, with your illustrated one reducing to 8*(85/60). If the person calculating the pay doesn't know that 1 h 25 min is 85 min, he/she is in the wrong job. Regarding your percentime hours, conceptually that's really nothing new. However, there's no need to change the length of the hour or screw around with the second.. I seem to remember seeing time clocks calibrated in hours and hundredths. (Perhaps others as long in the tooth as I am can confirm that.) They're perfectly normal hours (i.e., 3600 s long, with no special name), with no mention of minutes or seconds. The worker clocks on to the job at, say, 09:00. The starting time stamp will be 09.00 (decimal point, not colon). Then he clocks off after 85 min. The time stamp will say 10.42. The pay, in dollars, is then 8*(10.42-9.00). There's no reason a foreman's own analog watch shouldn't also be dual-calibrated -- hours and minutes and hours and hundredths. All of the above is, of course, irrelevant if the worker clocks in and out at some kind of online scanning device, using an ID card and a job card, for example. It's amazing what can be done with software. <g> Now, as I've dealt entirely with a unit (h) accepted for use with SI, we're back on topic. Doesn't that make you feel all warm and fuzzy? Bill Potts, CMS Roseville, CA http://metric1.org [SI Navigator]
