These three different philosophical perspectives are based on three different approaches to engaging with a tradition. I call these three perspectives "traditional lore, reconstructionist" and the middle ground of "objectified traditional lore," or "revisionism." If you ever wanted to be entertained, just get together three individuals who are die-hard adherents of these three different perspectives, introduce them to a strategic point of disagreement, and then let the fur fly.
According to Barrabbas' definitions, Traditionalists are defined as members of a particular initiatic system or lineage who adhere to to the teachings of that tradition regardless of outside evidence to the contrary. Reconstructivists seek to restore the practices of a particular group at a particular period in time according to academic writing and research about the tradition. Finally, Revisionists validate and augment their lore by researching academic and scientific information. Like this. By those definitions I'm clearly a Revisionist and proud of it. In fact, I have no idea why anyone would want to be a Traditionalist or Reconstructionist if they're at all interested in doing magick.
Purely Traditionalist systems suffer from the accumulation of dogma. Without any sort of peer review inaccurate information can wind up being disseminated. For example, the Golden Dawn origin story about Anna Sprengel and her body of European adepts made for a great plot element in my novel but most experts agree that it is probably not historically true. From the fragmentary technical documents that have been published on the Golden Dawn tradition sometime between Aleister Crowley's work with Macgregor Mathers ("Liber O vel Manus et Sagittae") and Israel Regardie's publication of the Stella Matutina documents ("The Golden Dawn") it looks like the Lesser Invoking Ritual of the Hexagram was replaced by the Lesser Banishing Ritual of the Hexagram, a change which is taught by most of the modern Golden Dawn orders but which according to my own empirical research on probability shifts massively weakens the system.
Reconstructionists sometimes make important new historical discoveries about their respective traditions that prove useful, but without revision any such system is always going to contain elements that were appropriate in a particular time and place but are no longer relevant to the modern world. For example, there a million books out there that purport to teach "Celtic" Paganism. Many experts believe that the discovery of bog men clearly demonstrates that the ancient Celts practiced human sacrifice. Should we? Obviously a modern person is going to respond in the negative to that one, but as soon as you start changing even extreme elements of the system to fit the modern world you effectively are a Revisionist anyway. I also think the idea that if a magical practice is old it automatically works well needs to be discarded once and for all. My electric refrigerator works a lot better than an icebox.
When Traditionalists and Reconstructions went to war over whether or not Ronald Hutton's ideas about the history of Neo-Paganism were correct my Revisionist response was pretty much a big yawn. Who cares? Seriously. Either the techniques work or they don't. Whether or not they're ancient or modern shouldn't make any difference, especially since the debate seemed to be between two sides that couldn't agree over what percentage of modern Pagan practices have ancient roots. The more important question is what percentage of modern Pagan techniques work and what percentage don't. If all the people kicking up a huge fuss about whether the correct percentage of ancient lore is 10% or 90% instead performed a series of experiments with various Pagan magical techniques and recorded their results something useful would be generated rather than an emotional but pointless debate.
Magick is a technology. That means if something works better it is better, plain and simple. Some of its effects can be subjective, but that only makes evaluation of those aspects difficult, not impossible. Psychologists research subjective mental phenomena all the time. And as far as practical techniques go, all you need to do is set up experiments to test the probability shifts those techniques produce. Ancient methods are still worth studying because the human mind has changed little over the last several millenia, but they should be researched, tested, and integrated into modern systems only if they are found to be effective. The "sifting" method that most proto-sciences use to accumulate knowledge can occasionally produce spectacular successes that the formal scientific method will miss, but at the same time it can produce some spectacular failures, superstitions that manage to live on as the tradition evolves despite their ineffectiveness.
What amazes me from following this whole debate is how ridiculous it is. If your spiritual system works for you and produces the results you want, why should you care if it's ancient or modern? If it doesn't work for you, why should you practice it? Thelema works for me and "The Book of the Law" was written in 1904. I've certainly never lost any sleep over its relatively recent origins. As a matter of fact, from the standpoint of study, being in possession of the original manuscript is quite useful, and if the book were thousands of years old it almost certainly would have been lost. Furthermore, Thelema's modern elements such as "the method of science, the aim of religion" are a big part of its appeal to me and had it originated even hundreds of years ago those elements would probably not be present.
UPDATE: The comment thread for this article has drifted over into discussing the issue of individuals in the Pagan community who do not practice magick but still want to be treated as authority figures. That conversation is continuing over at Rob's place.