CNC Retrofit knowlegebase - Request for your input

I've been dreaming of a CNC conversion of a knee mill for some time
now. For the past couple of years, I've been reading all the familiar
Yahoo groups (DIY CNC, CAD CAM, Mach, Gecko, etc) along with the
boards on CNCzone. I'm finally in a position to begin working on this
project, but I'm surprised just how little of the vast amounts of
knowledge & expertise that flow through this community have actually
been captured in a central place.

Now, I realize there is no cookie cutter path to a CNC retrofit. This
is a complex systems integration effort with many possible paths to
success. However, that doesn't preclude the collection of "best
practices" or pros & cons or "if/then" project planners, and the
like. If I had been diligently making notes and saving bookmarks over
the years I could fill a good sized book. Sadly, I haven't, but more
poignantly, I don't believe the larger community has done much better.

I'd love to be shown that I'm wrong, but let me expand upon this
comment with a little "for example". Here's a sort of FAQ topic list
(and only for a mill conversion at that), but I challenge anyone to
show me where one can get insights (pictures, plans, sources,
personal experiences, reference articles, etc) on these issues *in a
centralized place*. I know the info is out there, but it's highly
scattered, and much of the best stuff simply slips by like the
proverbial sand through fingers:

- What are the pros and cons of different types of machines to use
as a starting point? Old Bridgeport vs New imports? Knee vs square
column?

- Steppers vs Servos? (so often asked I was surprised how buried
the "reference" posts are)

- If steppers, what are the popular drive (and motor and accessory)
options? By pricepoint? By application? (e.g. small/med/large mill)
Real world experiences? Advantages/disadvantages? Ditto the above for
servos.

- Given a chosen path (e.g. Servos on a Bridgeport-type knee mill),
are there any sample controller box plans? What's a breakout board
and why do I need one? Wiring schematics? Build reports? Recommended
parts suppliers? Tricks & traps?

- What IPM is recommended given the application? What are the calcs
to determine a possible reduction to get there?

- Motor selection/testing? Power supply matching? Encoders?
Suggested resolution?

- Continuing the example of a Bridgeport-type retrofit (although
there are related issues for any path chosen), what are the Z-axis
pros and cons of driving the quill vs knee? What are some admired
approaches for the Z-axis quill drive? (good examples here are
particularly difficult to find, yet this problem has been solved
hundreds of times by people in this community who are willing to
share this knowledge)

- What are similar admired approaches to the X & Y axis (although
examples here are easier to find)?

- Ballscrews - are ground worth it vs. rolled? Who sells them?
What's involved in their installation?
Is it ever appropriate to use leadscrews with modifications? What are
those modifications?

- What are my software options? This is a huge topic in itself and
goes way beyond the choice of Mach vs EMC, and is ever changing,
particularly when you consider CAD & CAM options in the software
chain.

And the list could go on and on and on (pendant options? Mach 3
screen customizations? Benefits of a dedicated pulse generator like
the G100?)

What's my point in asking you to read this rather long post? I'm
impressed with the knowledge of this community and the spirit of
sharing that exists within it. However, I'm frustrated by the fact
that this knowledge is not being captured efficiently, compendium
style. Some of this is due to the temporal nature of the mailing-list
centric groups that dominate the community. If you don't keep up with
the posts, the knowledge they contain is as good as gone.

There are any number of technical approaches that could help here,
including hyperlinked FAQs, "sticky" topics within forums, and wikis.
Of these, it strikes me that a wiki would be the ideal knowledge
repository given the multiple author contribution and maintenance
requirement of the task at hand.

However, the first and most important thing is to collect the
knowledge to be centralized in the first place, if only as a starting
point. I gladly volunteer to be part of this effort and welcome the
help of others. The ideal response to this post would be, "Hey,
dummy, that exists and it's all at this link". Failing that, I'd
really appreciate your favorite sites, links, posts, build reports,
and any specific piece of knowledge you're willing to share.

I will summarize all responses, add them to my research, and --
working with any other volunteers-- figure out a good place to put it
all. No or poor responses will simply result in no or a poor end
product, so if this is interesting to you, please take some time
to "pour some knowledge". With better knowledge management, more
people will get involved in this community, which will lead to more
perspectives, more creativity, more (and possibly cheaper) products,
and more fun for all of us.

I will cross-post this on a few lists to try to maximize response. My
apologies beforehand if you see it more than once.