After reading User Research Smoke & Mirrors..
This article describes the tough relation between designers and designer-hirers, and the hardships of designers trying to use scientific methods to prove their expert point of views.
Design is by default not science. By using scientific research methods to try to make sounds more convincing to people, I feel that it will only bring limitations and restrictions only. Design has their own rules and guidelines, but designers who are bold enough to break them and able to justify their actions will create breakthroughs. Science, on the other hand, adhere strictly to laws and results; hence by making scientific user research as the foundation of the design process, this is like asking designers live by numbers and results rather than experience and instinct.
Research as Design Tool
In this part, he mentioned about using eyetracking as a scientific research design tool. Seriously, does those few miliseconds that a user chose to place their focus on certain spot of the screen determines the usability of that page? The tool may tell where the user is looking at, but that does not mean the user is cognitively processing what he is seeing. ie: he may be just staring at that spot but thinking of something else. Personally, I feel that this eyetracking thing is something which people created to create a buzz only. It's usefullness is really debatable, and to zoom into where users may look at using technology may just seem like going a little too far. It should not be the case of technology driving design, but it should be technology facilitating design. Hence research should be at best to find out as much information required during the initial design stage, but those results should not be the determining steering factor for the later stages.
Research as Political Tool
I could definitely understand the usefulness of it being the only weapon for designers to retaliate against non-designer bosses. There are things which seem obvious to designers, but bosses usually find it hard to accept things without any support. Research helps to lend weight to the opinions of designers, but I also feel that it is something essential to CYA (cover your ass) when something goes wrong. At least the blame won't fall entirely on the designer when such things happen, and things do go wrong for most of the time.
Research as Bullshit
While I was reading up research papers for another module project, some of them really didn't seem impressive nor groundbreaking to me. In fact, I agree that some research are just bullshitting so as to make their results look outstanding, when there is actually not much practical usage for it. The article mentioned "Persona Rooms", which I feel, may be useful to a limited extend. When designers "walk into and live" in their users' space, the experience they are getting may not be the same as what the real users were experiencing, as experience is unique to individuals, and cultural / past experiences affects present experiences. It would actually be much better if the designer could arrange a homestay in potential users' houses. That may give a far more realistic experience than this mock up shell.
User Research isn't a Bad Thing
User research is never a bad thing, and the more user research being done, the deeper the understanding of users will be, and the better the product may possibly be too. It is just the extend to which it is being conducted, and knowing what knowledge or lessons that should extracted from the research that matters. Following results blindly will ruin an initial good design, but not following it may ruin the end product completely.
I believe that by utilizing moderate user research as a political tool and a design validator, designer will be able to work more effectively to create user-centric products while convincing their bosses of the design decisions. Over emphasis on the scientific research methods will only create more mess and trouble within the whole design process.