This is only my opinion but "Wiring a House" by Rex Cauldwell is one of the best practical wiring books, also his "Above The Code" recommendations are usually dead on as far as real life situations.

I vote for Cauldwell. Excellent book!

I saw that too, some really really good advice from someone who's learned from experience. Shapiro sounds alot more like a pencil pusher, or "desk jockey" as Caudwell says. Shapiro even says its a safety hazard to install a locknut upside down, now how many houses have burned because a locknut was upsidedown?

My recent experience in testing a newly installed GFI receptacle...pushed the test button on the receptacle and it tripped okay and figured it was all set. A few days later I was going through the house with a hand held tester (the $12.00 model with GFI test button) and plugged into this same outlet. Hit the test button on tester and nothing happened. Again pressed the test button on the GFI and it tripped. But, when I plugged in my tester there was still power to the receptacle.
In short, the GFI was faulty. It would trip with the test button, but still have power. Strange, but it made me a believer in testing with a hand held teste as well as the test button on the GFI.

Coolvt, your scenario is more indicative of a miswired GFCI, not a faulty one. It used to be that you could wire a GFCI with the load and line reversed and see exactly what you described. Newer GFCI receptacles will not operate if the wiring is reversed. This is precisely why Shapiro recommended the Wiggy test. It's no longer necessary with the new GFCIs.

BTW, both Shapiro and Caldwell are practicing master electricians, not pencil pushers.

My main tester is two pigtail Edison lamp holders in series with identical 15 watt lamps in each lampholder. Full brightness indicates 240 volts, dim indicates 120 volts. Two short pieces of insulated 14 gauge solid wirenutted to the free ends of the outside leads of the lampholders are the probes. Testing a GFCI receptacle is a simple matter of putting one probe in the grounding prong slot and the other probe in the (hot) narrow receptacle slot. If it does not trip the GFCI, there is a problem. Yes, I do have digital instruments, several of them. But this tester does 95% of the work.

I've used a plain old socket test light for testing for a long time, it's reliable, simple and easy to see in bright conditions. I've found that using rough service bulbs designed for garage door openers works better, bulbs last alot longer. It also draws plenty of current to trip a GFCI, can make grounding checks to appliances and can provide some jobsite lighting.

My specific reasons for the small lamps is that they break less easily and the resulting assembly is less bulky. Use of two in series enables use for 240 volt. 240 volts burns out 120 volt lamps almost instantly. Testing only for 120 volts on 240 volt circuits and equipment will often give misleading results due to feedback. For heating elements and fuses in 240 volt circuits, they can be open and still have 120 volts at each end due to feedback. The lamp tester is very low cost to build.

As far as testing a GFCI I still think the combination of my test light to ground and using the internal test function is the best way to determine reliability. I would trust my test light more than a wiggy or anything with internal circuitry or electronics. If the GFCI does not pass both tests it is either bad or wired improperly. Theres not much to go haywire on a test light in a keyless socket, besides a bad bulb.