Db :
Regarding bias, your the one who is claming that their is no bias in your work, not me. I take it by your post on this thread you are now changing that view?
Bias is a very broad term, you need to be a lot more specific. What I have mainly argued against was the usage such as cited by Buzzbait who argued that the review would be done differently if I liked or didn't like a knifemaker - thus it is biased. A less frequent, but still common arguement is that the work isn't "scientific" because it isn't done by machines and is thus biased.
All scientific work is an estimation as you are never actually measuring the desired quantity, or calculating one, but something which you feel is the best estimate for it. As you learn hopefully you can narrow the range of influences down and thus get more precise results. Of course this this doesn't mean the work is biased. The standard quantities used for such are the mean and standard deviation which are defined as "unbiased *estimates*".
As the reviews have evolved, methods have changed. I would not do much of what I did in the past becuase of systematic errors introduced because of factors that I didn't fully consider. As these come to light they are removed or at least reduced in magnitude for future work, and notes made in the reviews to illustrate the problems.
What is overlooked is that you generally can't actually say if a test (based on method) is biased or not. That is too simplistic. It is like saying a knife is good or bad. That is completely vague and therefore meaningless. Which is really ironic actually considering the arguement as a whole.
Bias is critical on the conclusions drawn. For example, specific to this thread, the rope cutting tests as described in the above, is *NOT* an unbiased estimator of relative steel performance for reasons described. However is is an unbiased estimator of relative edge lifetime from knife to knife.
Thus a test isn't biased or unbiased, it can be both depending on what it is used to determine. Just as a knife isn't good or bad. Just a steel isn't great or horrible - it depends on what you want it to do.
This goes back to the above quote from Brownshoe :
You claim that using a knife to cut something it wasn't designed for is a good test because ...
This is too simplistic a viewpoint. A test isn't good or bad. It depends on what is to be interpreted. If you use a fillet knife to chop through a piece of 2x4 and from that performance judge its suitability to process fish, that was a very bad test. However, if as Phil Wilson did with one of his fillet knives, use it to make a comment on the edge durability, it is a decent test for that steel, not of course for the knife.
I still clame that a knife preforms alot differently depending on the user, more than Cliff and alot of the rest of you give it credit for.
This is discussed extensively in the reviews. Some of them have specific examples of just how much of an effect can be seen when different methods are used. Not to mention the fact that the reviews contain links to work done by others so this perspective can be obtained.
I have always noted that the ideal review would be a collection fo work spanning the results of use by individuals over a range of skill levels, from novice to expert, and from the very weak to the very strong. This was origionally why I started sending out blades I had reviewed to others.
Of course a lot of performance isn't dependent on the use but is inhernet in the knife. Just like you can make a statement about the brightness of various flashlights, uniformity, whitness etc. . You can make general statements about knives that will hold for any user, such as if they find some knife brittle, then so will they find another. The Reeve grip is more abrasive than Cold Steels Kraton grip. This is the *critical* point, references must be given, the more the better.
For example a good ABS make like Ray Kirk will make a bowie that will vastly outcut the Trailmaster from Cold Steel. There isn't anyone who is going to see a reversal in performance from those two knives. Though the extent of performance gained will be directly proportional to skill and strength. Some tests are very user independent as well, just as push rope cutting, and just reflect geometry.
This is why I include comments in the review by which my skill and strength can be used, such as the chop depth on a class of wood, and the number of miss hits. I have also put up pictures of wood chops performed so the line following, staggering and such can be noted.
In addition to cutting ability, if you are very skilled and/or very weak, the functional limit of durability necessary will be very different than if you are very ignorant and/or very strong. This is why it is rare for knives to be graded in such a manner on their performance in the reviews as it can mean different things to different people.
There are some execeptions to this, some knives are so incoherent I simply could not find a use for them. In such cases however the maker or manufacturer (or just user) is always welcome to step in, either through email, or on the linked thread, and point out where the knife excells. However in general I will just state on the performance compared to other knives and opinions can be drawn.
Some parts of the review are almost completely subjective like the handle ergonomics. In this case all you can do is bring in as much references as possible, commenting on other knives, and hope that the reader has used one and thus obtained some perspective.
Brownshoe :
carcasses to cut can be found in many places, just go to your local animal control.
Locally this doesn't work. If I hit a moose and kill it, I can't claim it, even if it destroys my car - which you would think the goverment would then give you so it would not be a total loss. Fish and Wildlife will come in a sieze the animal.
Why cut carpet with a skinning knife?
As an aside, this would lead you to imply that I had actually done this and used it to grade the knife. "This is a horrible skinning knife it doesn't cut carpet as well as my stanley utiltity knife". Of course this wasn't done, which is why no actual link was given. The only knife which is skinner-like which was reviewed in such a manner was Pronghorn, however it isn't simply a skinning knife and the carpet cutting wasn't the full extent of the review.
Of course work should if at all possible be done in the intended range of work of the knife. However as noted in the above, and in many other places, the reviews are however often not that restricted, because they are not intended to be simply used to judge a knife but instead look at the performance inherent in the steel or the geometry or other factors.
swede79 :
... while the testing isn't perfect
BLASHPEMER!! If you keep refusing to say what we tell you to say we will be forced to remove you from the list of people we are impersonating.
-Cliff
