This question asks whether the bowline knot is better with the tail inside or outside the loop.
When climbing and tying in with a bowline on a bight, does it make a difference whether one starts with an inside or outside bowline?
The most common use of a bowline on the bight in climbing these days is to make a loop in the middle of the rope in a party of 3 or 4, as one of the comments above mentioned. In this case there's no inside or outside, and the knot is symmetrical with respect to the two ends.
As a tie-in knot, while I can see the difference between an inside and outside single bowline - especially the chance of the end getting caught - the bowline on a bight has the two ends coming out parallel to each other and taking the same turn round the bight. So I do not see any meaningful difference between the two.
We here at the Rigstar Training and Testing Center do a lot of break tests. Having the tail of the bowline to the inside or outside does not matter for strength or efficiency of the knot. The reaction is the same.
Here is the scientific reason why it's the same strength or have the same average breaking strength. The bights on the bowline cause compression to the inner part of the bowline rope, also there is the d/D ration of the line going around itself, then the rope when pulled is in tension which causes more compression with the two bights.
When the rope is in tension it causes the molecules to rub against each other which causes heat within the bites. All of these reactions are the same whether the tail is to the inside or outside of the bowline knot.
Tying the tail to the inside or outside does not matter and is a personal preference.
Short answer = 'Yes' - it does make a difference.
Long answer: If you are going to select and use a particular ti-in knot for life critical applications, you should select a fixed eye knot that is inherently secure.
'Bowlines' have 2 distinct advantage over the F8 eye knot (#1047):
The F8 eye knot fails on these 2 salient features. As climbing ropes get thinner and thinner - tie-in knots become increasingly vulnerable to jamming. Enter stage right - 'Bowlines'.
Inherently secure 'Bowlines' include:
With regard to the issue of tail inside versus tail outside... in circumferential (hoop stress) loading profiles, tail outside of the 'eye' withstands this type of loading much better than tail inside the 'eye'. The Simple Bowline (#1010) has the tail inside of the 'eye' - and it quickly deforms and fails when circumferentially loaded. In contrast, the #1034 1/2 Bowline (tail outside of the 'eye') is resistant to this type of loading profile. Circumferential loading is also known as 'ring loading'.
I would also comment that the #1080 Bowline-on-a-bight is not suitable for biaxial loading profiles as would occur if it were employed mid-line. It becomes unstable due to collar distortion. If you want to tie and use a 'Bowline' as a midline knot where it will be biaxially loaded, you should use #1074 Bowline with-a-bight. #1074 is both Tiable In the Bight (TIB) and remains stable in a biaxial loading profile loading. #1074 needs to be secured with Scotts locking maneuver or perhaps by a double overhand 'strangle' tied around the Standing Part (SPart).
It is not two knots if there is only way to tie. But apparently the feeds are not symmetric. And it would be possible (but not likely) for the feeds to flop.
According to this link page 7 the feeds are different. I don't paste as the link is copyrighted.
Hard to describe but the load is the side that is not wrapped.
As stated by Liam a double figure 8 is more common knot for this purpose.