I think Marty's getting a little too excited in his piece, though I agree with the basic concern that the tool has a lot of issues. I moderated the session, and I plan to do my own write-up in the future. But here's some of my take.

First of all, Google hasn't "broken ranks" with "15+ years" of open source search engine tradition in providing keyword data. The first few years, we could hardly get any. When we did, it was largely through Overture. The Google AdWords tool followed suit and continues to be a key resource. Nor did anything in the session come out as a sudden change. While Google's had interface changes, the real issue to me is that under the hood, things are buggy.

A key example was when I did a live search for "easter baskets," and the tool reported back something like 30,000 searches per month worldwide. Worldwide! In the end, the Google rep suggested there's probably a bug there. Similarly, the fact that Facebook didn't list "facebook" itself as the top term, he largely thought it was also another bug.

Key things though, like learning that only terms that are "commercial" in nature are listed -- IE, someone's buy some ads against them -- were a revelation. But that didn't sound new, simply something that had been going on.

The good news, omitted in Marty's write-up, is that Google said they plan to change the tool within a year so that you'll be able to download raw data and for free. Faster would be better, on that front.

After reading the article I took a look myself, and it definitely seems like things have changed. For awhile you could do searches for longtail keywords by choosing to look at phrases that had fewer than a certain amount of global searches.

However, now when you do that, nothing (or at least not much) seems to show up. This is huge and means that we're not going to find all the keywords we might want to find. It's like we're back in the 1990's or back to using KWR tools which get their info not from Google. If they're not going to show us all the kws, it's not very useful.

This is a giant leap backwards for SEOs. :(

And even more proof that Google absolutely can't stand SEOs. They don't want us doing SEO, they don't like us, and they never will. (Regardless of what Matt Cutts says.)

I've never had much faith in the tool from the get-go. The inquiry numbers are obviously inflated to justify increased keyword phrase pricing for paid ads. For both paid and organic results, I have not seen enarly the traffic commensurate with their claims. I use it for a relaive strength verification, and that's about it.

Danny, Overture KWs were enough. When Yahoo morphed it into Panama, then it actually got deeper after sucking for a bit, inventory wise.  Google started sharing KW data around 2002 when AdWords transitioned from strictly CPM to more CPC approach. I'd say that there's been awesome KW data around by 1 or more SE sources for at least the better part of a decade. Before that we could score KW data for Alta Vista Banner buys. We used THAT for SEO at the time, even though we never heard the term SEO. Yep, I'm pretty excited. That's because doing KW research is pretty F'ked now. To anyone out there on the street doing SEO, PPC, etc... this borders on devatstating.  Google has broken trust with me, their customer.  I  decide what is "commercial" to aimClear's clients.  :)

I’m not too crazy about using the Google's new Keyword Tool for keyword research. The best keyword research I think comes from Google Suggest.In fact, we’re offering a free software tool that scrapes Google Suggest from our blog.You can download it for free here: http://www.market-vantage.com/2010/10/free-keyword-research-using-google-suggest/