This seems pretty interesting to me. Before I start learning algs I'll probably look into orienting all the edges during the first 3 F2L pairs, as I don't really know how that would slow me down. Also some of the algs seem pretty long and with recognition time it might be quicker to just do OLL. Thanks for showing the concept though, it's a neat idea.
http://cube.crider.co.uk/algtrans.html
Put the alg in and look at the "Mirror L/R" although you probably can do it without this.
I don't really keep up on the developments and variations of the Fridrich Method, but here's my 2 cents:
First check out my response on the second page of this thread.
If you are dedicated to doing the cross, then CE pairs, then learning WV would probably be of limited value. You already need to know 57 (shudder) OLL algorithms, and 21 PLL algorithms. Unless you happen to know some nifty moves to always force the LL edges to be oriented by the time you have the first 3 CE pairs solved,
and force the fused FR CE pair to sit at UFL and UF, you would have to learn the mirrors.
As I am biased towards learning as few algorithms as possible, I think that 57 OLL algorithms plus 21 PLL algorithms is at or near the total number of acceptable algorithms for a speed solving system. If there are plenty of world class cubers out there who routinely use more than 78 algorithms, then disregard this paragraph.
I still think that the best use of WV is with the Petrus Method. If you follow the advanced method where you orient and permute the corners of the last layer, you need 42 algorithms, plus 4 to permute edges. I think this is an improvement over Fridrich in the total required algorithm count, and I believe that the method has a lower average move count than Fridrich (please correct me if this is not the case). The problem I have with the Petrus Method is that you still have a 2 look LL, despite investing the time and moves to twist the edges. sure, you save 32 algorithms-worth of required learning, but it is still a 2 look LL.
If you do WV instead, making sure that you solve the FR CE pair last, that works out to 27 WV algorithms, plus 21 PLL algorithms. For a Petrus solver, that would only mean learning 17 PLL algorithms, as they would already know the edges only permutation algorithms. A Fridrich solver would have to learn 0 new PLL algorithms.
I think that the most significant barrier for a Fridrich solver to learning WV in the way I intended it (with Petrus), is the severe conceptual/perceptual shift from doing the cross, then CE pairs to block building and edge twisting. In the beginning, it would be agonizingly slow, and I suspect it would just feel wrong. This would probably be discouraging enough to make them revert back to Fridrich.
I think that the relatively low average move count of the WV algorithms and the relatively few algorithms to learn make it worth exploring, but
not using it in conjunction with Petrus would probably restrict its usefulness.
