As another example of why I claim you're wrong, let me give an example of a company I worked at for far too long. It was a small family run business that sold and serviced automobile radios. At that time, car manufacturers contracted that service out to hundreds of smaller repair shops. This was one of them.

As part of their service program, a dealership could call up and request a radio to take care of a customer's problem. The radio would get shipped, the customer would get a call from the dealership to come by and swap it out. This way, there would be no hole in the dash, as most drivers would go insane if they didn't have some sort of radio (or CD or cassette) playing in their car. At least one dealership figured out how to game the system this company used (we figured that they ordered the top of the line radios to use as upgrades for the sales department). Consequently, they were stealing more than $1k/month of product from the company. Mostly it was due to crappy paperwork that ended up in a filing cabinet, never to be seen again. I came up with a simple VB/Access application that used a barcode reader and a barcode printer. I also made the mistake of using my own money to buy a PDA with built-in barcode reader (specifically, a Symbol SPT1500) and writing an app to track paperwork for the employees.

The boss hated it because his wife hated it. She hated it because it would make her have to do things slightly differently. Furthermore, the boss also hated it because it showed a flaw in his business system, and he treated it as if it were a bad mark on himself: he looked bad, therefore he was a bad person (that was his psychological take on the matter).

Between bad customers (the service department at that dealer wasn't the only crooked one), bad employees (at least one went into business for himself filling his garage with inventory before quitting), and a bad workflow system, he was losing between $50k and $150k per year that I could figure out.

Anything that involves more than 1 human will involve politics. And academia isn't any better or worse than any other business when it comes to the stupid nonsense that humans do to each other.

Several years ago, I was the webmaster for the Hearing Research Center at Boston University. It's a collection of 20 or so labs, each run by a professor, with associated grad students, etc..

The first thing I noticed was that most of the researchers were universally and remarkably happy people. (I had come from a background in Physics, which had more angst.) Then I noticed a huge number of collaborations as I was compiling the professors online CVs, and a huge number of publications. Before a certain date, for the older professors, there were individual authorships, but not after. This was a tantalizing mystery.

The department hosted an annual party for hearing researchers worldwide, and I hung out with the diehards at the end of the night - the old timers - and asked them what the story was. It's fascinating.

Those who are department heads now were all in school together at MIT thirty years ago and they all had the same professor who sat them down and told them that research didn't have to be competitive and backbiting. While these people were grad students, they decided to take control of how science would be practiced in their discipline.

Hearing research doesn't get anything like the same kind of grant money that, for example, vision does. In order to advance the discipline the most efficient way, with limited funds, they COULDN'T afford to duplicate research. They need to share data, yet give researchers the freedom to use the data differently if they have different ideas.

So patiently, these people took over their discipline, and groomed their grad students into the culture where people are rewarded based on sharing and collaboration. Perhaps there's a little less thunder, but people are publishing very frequently and the discipline is advancing much more quickly than it would have otherwise. They took control and set up the reward system differently.

Did I mention that these scientists were some of the happiest people I've ever met?

It is an interesting question in Game Theory, what strategy is optimal for academic peers.

Another thing to keep in mind is that academic papers are supposed to be about the truth, and so there is ostensibly an "objective" standard to which papers are held. Of course this varies a lot by quality of the conference, but the best conferences will generally do a fairly good job of letting the best papers in and keeping the bad papers out (most of the confusion/"unfairness" comes at the borderline papers, where it is understandably more difficult to make a judgement).

That being said, I think some types of competition will always be a problem in any discipline, although it's not at all clear that the corporate world is at all better.

Theoretically, yes. In practice, anonymizing work is hard to do -- if you're prominent enough in a field to be asked to review papers, you probably have a good idea who is doing what work, and you'll probably be able to recognize authors' individual writing styles.

I've been asked to review "anonymized" papers in the past, but with the exception of one which wasn't really in my field (I wrote back to tell the editor that I didn't think I was qualified to review the paper, and he found other referees) I've always been pretty sure who the authors were.

There aren't that many prominent people in a given discipline (and some of them do a poor job in reviewing anyway - slow, etc). It's quite normal to have a nonprominent person do the reviewing.

This is, after all, peer reviewing.

And BTW, it is exactly the desire for objective evaluation that has penalized interesting ideas (e.g., for web 2.0 apps) that are hard to evaluate using experiments/proofs. That's rather unfortunate.

Try being successful in a way that makes your boss look bad, and you'll discover he isn't the least bit interested in your "success".

But not all environments are equally bad. There are green patches on both sides, or at least greenish :) Your job is not to settle for a brown one.

When your million-dollar grant gets funded, you buy your grad students a beer.

When your million-dollar contract gets signed, you buy yourself a car.

Unfortunately, the current system in academia of peer-reviews always felt awkward to me simply because you are not aligning self-interests. Any researcher in a field knows that by approving someone else's work, he is creating more competition for precious funds. Sure, a terrible boss can shoot down a good idea of yours for pride's sake/other reasons, but that would hurt the company in the long-run. At least when you are working for a boss in industry, your self-interests are aligned and you both are aiming for the same goal.

In academia, you write grant proposals and research papers that are evaluated by your peers.

Part of what I was getting at is the analogy isn't perfect because each have their own unique characteristics. We can make a mapping from one to the other, but we probably won't learn anything from it.