I don't really agree. Yes in some respects and for some purposes the independent claims are the most important parts of a patent, but reading only the independent claims is very different from reading a patent.
The specification and drawings are important as they, among other things, define the meaning of the claims. Usually, if you just skip ahead to the independent claims you will have no idea what the claims are talking about. This is especially true in the electronics/software/online arts where people tend to make up their terminology as they go along.
There's a great comment on the article itself by a Max Kennerly, who appears to be a practicing attorney. He cites a Federal Circuit court opinion that basically says - yes, the claims must and do stand alone. You're quite right that if you know nothing about the patent you may be lost in the claims, but from a legal standpoint, they should be sufficient. And more practically, if you come to the patent knowing a lot about the field, you'll usually get a very good sense of the thing from the claims alone.
And if you've only got a minute, you're time will be far better spent in the claims than in the abstract, description, or title, which is what most people do!
I am also a practicing patent attorney, and the case Mr. Kennerly cites to is the well known Phillips case. In any event, that case says that claim terms can be defined by the specification.
I agree that if you have a minute only and you are worried about infringement it may be a good idea to start from the independent claims. But you should not fool yourself that you have read the whole patent. Because if you only read the claims, and even if you think you understand the claims perfectly you can be in for a nasty surprise if you do not read the spec.
Also important: understanding the difference between a patent application and an actual issued patent. The rule of thumb is that if it has a number in the millions, like 6,123,456, then it's a patent. If it starts with a year, like 2008/1234567, then it's an application. The slash may or may not be omitted.
This approach does seem to cut to the heart of the matter... I have had a patent in the system for a while, and just to reassure myself, I read just the independent claims.
While reviewing my patent I have always become worn out by the time I get to reviewing the claims so I have always been a little wary of them (especially because I am used to being as precise as I can, but claims are written to be interpreted broadly).
Sure enough, when I looked this time, the claims spoke loudly to me. What needs to be there is there in the independent claims.
It is a little disturbing that the "code" of a patent (the claims) only takes about 30 seconds to read and the rest of the ream of paper is just comments and/or obfuscation. (Most likely the latter.)
Nevertheless, from the description you give to a lawyer to the final patent, there is both a notable expansion factor and a large drop in comprehensability. Last time I had to do this, the expansion factor was 10, although I think that's above average.
If there is a drop in comprehensibility, then you probably need a better patent attorney. Patent applications should be as clear as possible.
Regarding expansion, this is sometimes necessary. A good patent lawyer may need to add a lot of material to your description. There are many possible reasons for that: for example, you may be so advanced in your field that things that may seem obvious to you are actually pretty complex and must be described in more detail. Or if you have defined your invention too narrowly, a good patent lawyer will describe your invention in different scopes (both broadly and narrowly) and think of all kinds of different fields and applications for your invention, thus ensuring you get a better value out of your patent.
