Buzz & MF, I understand what you are both saying.
My experience is that better siting rarely eradicates the problem (cooking fumes and steam can have a fair old spread) but, yes, of course, better siting can reduce the problem. And, yes, the detector should be installed in the best possible position (in this case to reduce unwanted alarms) from the beginning (with due respect to BS recommendations).
I also, obviously, agree that it is better to have the smoke detector 7.5m away from the source of the problem and only 500m away from corner walls, rather than the centre of the room.
IMO the cause of problems with poorly sited detectors is rarely the fault of system designers. They just mark detector positions on drawings and generally the scale of these drawings is so small that they can normally only indicate that a detector should be installed somewhere in a particular space or area. Often there are numerous other symbols for equipment on the drawing as well and it can all become a bit confusing.
IMO it is the installers, in this case, who are often to blame. And all because they often don't know anything about the BS recommendations. Even the best 'proper fire alarm specialists' will still use electrical installers rather than fire alarm engineers to install the cabling, and the cable installers end up determining the installed detector position. They normally look at the drawing and install the wiring to exactly the point that it appears to be shown on the drawing. They need to learn that the detector positions shown in any space/area of a drawing is just a generalisation and that the 'correct' position needs to be chosen with careful due regard to obstructions, potential causes of unwanted alarms etc., but with regard to the BS recommendations for positioning.
In my experience, the better projects are invariably those where a proper fire alarm engineer visits site and 'marks' the exact required detector positions on the ceilings before the cable installers even begin their work.