Guys,
I appreciate your comments on the new Camillus site. Thank you for the tour and I really appreciate your feedback. You'll be seeing a lot of growth on this one over time.
I guess a couple of points in reading your comments ...
Bottom line is that frame-based websites are less usable and "findable" than non frames-based ones. That's all, it's up to you -or whomever build a site- to make it as s/he wants.
No offense, but I'm wondering what you mean, or how you're arriving at the position that frame based sites are "less usable"? In our business, the only acid test that we would even consider is in the "daily user stats". Do you have any data to back up your position?
Not trying to be an A$$ or anything, but I can produce more than 1 million visitors a month in traffic data to back our choice of navigational options. I can tell you, without question, that a left-hand frame based navigation system is the most effective option in terms of "page depth, time on page, time on site and partial content retrieval" that you can design. We've done them all, and still do. In fact, it almost doubles the numbers of a standard "deck of cards" layout, whether running SSI or template driven aspects. In our case, design difficulty or budget was not an issue at all. We build what the numbers tell us will work best, and we can produce the numbers.
Actually, the Camillus site runs in frames, but also uses SSI, PHP and Perl. The Perl running both the internal search engine and also the cloaking software for keeping the critical meta content out of public view, which, in this case, is important considering the nature of the client's business market. And leads me to ...
As far as a frame based site being something that someone personally doesn't care for, this is a risk you run on any given visitor. But, professionally, you design for the numbers. The argument against framed sites being less index-able (or as you mentioned ... less findable?), I've read that for years on all the web master training type sites. I suppose the only argument I can make that would make any case against this position is that we currently have 1-10 rankings under 90% of the major key words in the knife industry. Actually, I think this is more an argument against the deep linked pages being opened outside the frameset from a targeted url listing. The point being, yes ... this is the common issue in the frameset design for most amateur web designers. In our case, we employ coding that eliminates the issue. Here's a deep linked example:
http://camillusknives.com/cuda/cuda_tal220.shtml
Voila ... opens inside the nav set! All the pages on this site perform in this way.
Anything beyond the numbers, style and personal preference aspects for instance, I can't defend. But, the aspects being discussed above, frameset index-ability ... usability ... and such, the numbers don't lie ... and we have the numbers.
I'm not all that sure that professionally using SSI and/or templates in the design to make your work simple is really a good position to follow. I would argue that these elements should be used if you need to design for long term ease of updating, or global content change across a deep site. The Camillus site is marginally deep, but they are a very dynamic and large company, so the reason we employed SSI was to simplify the content page navigation during global product line updates, which will be often.
The props go to Will, Bill, Phil and of course, Jim. These guys are the one's that did the hard work for Camillus ... all we do at atlantavirtual.com is type ... and hype

!
Again, thanks for the feedback. Plus or minus, it's all good here.
Alex