It really is incredible that this old chestnut still appears in aviation circles. As quoted by Voidhawk with reference to Doc 4444, and by 604guy, Tinstaafl, and many others with reference to common sense, the definitions of Heading and Track were carved in stone decades ago. I think Dan W was slightly misleading in his post as it did not actually clarify the definition of Heading in Pans-Ops.
Doc 8168 (PANS-OPS), the modern equivalent of stone, defines it thus (1-1) :-
Heading. The direction in which the longitudinal axis of the aircraft is pointed, usually expressed in degrees from North (true, magnetic, compass or grid).
Track. The projection on the earth's surface of the path of an aircraft, the direction of which path at any point is usually expressed in degrees from North (true, magnetic or grid).
While PANS-ATM and PANS-OPS cover different aspects of the business, these ICAO docs are co-ordinated where definitions are concerned and virtually every nation is a signatory to them.
The Americans seem to have no problem with this, nor do most other nations. The UK has officially dropped "runway heading", as per bookworm's post, because of the apparent confusion involved; amply illustrated on this thread. By issuing Heading instructions, controllers know that all within their tactical plan will be affected equally by wind effect.
The notion posted by many here that Heading on suitable occasions actually means Track is laughable. If it sounds wrong to you, query it. If, as muppet experienced, you get a bollocking for correctly following an incorrect instruction, or the instruction would compromise obstacle or noise-abate paths, file an ASR/MOR (or whatever your system) after you land. It works wonders and will usually improve something.