Here's a new signal mirror training video on YouTube,
and some comparison of results using various methods of aiming signal mirrors.

The new video teaches the "V-finger" method of aiming signal mirrors, a method taught in USAF survival manuals for aiming improvised signal mirrors.

There's a lot of good stuff in this <4 minute video - I think this video was extremely well done. The author has other survival-themed videos I plan to look at.

- The video illustrates the V-finger method from many points of view:

• signaler's point of view
• profile view of signaler
• side view of signaler
• receiver view of signaler

- It also adds

• tips on making the flash look artificial and deliberate
• safety tips
• some physics
• mention of signaling with moonlight
• notes that many of your flashes will miss using V-finger
• graphic illustration of the narrow beam of light.

- There's also humor

• Sound bite at 1:36
• Sight gag at 2:35
  
  
  
  ( Full screen version at YouTube )
  
  While knowing how to use an improvised reflector as a signal mirror is important, I stress that you will put many more flashes on your target if you use a mirror with a retroreflective aiming aid (e.g. the Doug Ritter designed Rescue Flash, the Ultimate Survival StarFlash used by the USAF and USCG, Coghlan's glass "Survival Signal mirror", Coglan's floating "Sight-Grid Signal Mirror" etc.).
  
  Does a good aimer make a big difference?
  
  You bet!
  
  In the video above, the signaler gets about 22 solid flashes/minute using V-finger. In the video below, a newbie gets over 60 flashes/minute with a mesh retroreflective aimer.
  
  In the flashing sequence of the first video (2:40-3:03) I see 8 solid flashes* in 22 seconds (about 22 flashes/minute), at short range, and if I look closely, I see 13 faint flickers as well, so at least 58 flashes/minute are going out, but most are missing, as his voiceover notes.
  
  In contrast, in the video below, my friend's flashes hit me nearly continuously, using a mirror with a mesh retroreflector aimer. This is all the more impressive when you know this is my friend's first session with this mirror (which I loaned him), and I'm seeing these nearly continuous flashes with my naked eye at a range of 22 miles. (My counting in the narration is to try to see how many seconds of essentially continuous illumination he could manage.)
  
  
  
  ( Full screen version at YouTube )
  This is a solid clue that hitting the target using V-finger aiming isn't as easy as you'd like - but why is it hard?
  
  The first video also illustrates some of the problems you face with V-finger aiming.
  
  First, the signal mirror beam is very narrow. At 3:47 into the first video, you can see the mirror spot on the stop sign. See how small that spot is - a fraction of the apparent width of a finger, and an even smaller fraction of the space between the V-finger. At my arm length, the spot is the apparent size of Lincoln's head on a penny, as seen on the right photo at this link.
  
  PHoto illustrating apparent mirror beam size vs. penny
  
  Since the mirror lights both fingers in the V-finger, the naive may think that any target they see between the V-finger will be hit. As you can see from the spot on the stop sign - not even close.
  
  Your flash will be on target with the V-finger technique if:
  (a) Your target is exactly (remember Lincoln's head) centered between your V-fingers.
  (b) Your mirror flash is exactly centered on your V-fingers ( with a small mirror, the V-finger spread is slightly wider than the flash, so you can see the edges of the light rectangle on your fingers and use that to center the mirror flash on your fingers).
  (c) Your eye is exactly centered on the physical mirror ( I would mark the center of the bottom edge of the mirror, hold the mirror right over my eye, and look under it, right below the center of the mirror.)
  
  Even if you get two of these three exactly right, if the third is wrong by the apparent width of Lincoln's head, then your target won't get the full flash. You've got to get all three right. ( At short range, you get a bit of a break - they may see the (much) fainter diffuse glint that occurs around the main flash.)
  
  Fine, you say, but for those in trouble who don't have a signal mirror on them, what are they to do other than V-finger?
  
  My first suggestion would be - make your own aimer on the spot - if you have a mirror, it isn't hard to make an aimer that will really give you a step up.
  
  === Field-expedient signal mirror aimer - aiming hole ===
  
  The first "step up" in mirror aimers is to put a ~ 1/8" sighting hole in the reflector - you can scrape off the silvering on the back of a glass mirror to make a sighting hole, drill a hole in a metal/plastic mirror, or, if you've procrastinated until you are in trouble, auger a hole in a CD/DVD with your knife point.
  
  Now, you can look through that hole at the target, line up your finger tip with the target, catch the "shadow spot" cast by the hole on your finger, and carefully transfer the shadow spot back and forth from the fingertip to the target.
  
  In the next video, my friend is using this method for the first time, at 22 mile range, using the "Silent Universal Signals Mirror" a thin lightweight 2"x3" stainless steel
  
  This segment is twice as long as the 22 second flashing segment in the first video, and you can hear me counting out 22 hits - about 11 hits for every 8 with V-finger. However, that's at 22 mile range (though with 8x binoculars). Looking carefully at the video (full screen, HD), even at 22 miles I see more than 20 flashes in each of the first and second segment, so at a more realistic range of 5 miles or so (where the flashes will be over 19 times brighter) the receiver will be seeing many more flashes than I did here.
  
  ( View full screen movie in HD at YouTube - the flashes are too faint to see in a small-screen video)
  
  * Bright flashes at:
  2:41, 2:45, 2:47, 2:49, 2:56, 2:57, 3:02,
  faint at 2:40