It is the ongoing problem of writing factual articles and books, upon what source or sources of information to base your account of the subject?
There are of course a plethora of choices ranging from the totally superficial to the incredibly well indexed and deeply researched and whilst one would obviously prefer the latter it is rare that the item you are working on has been covered in such detail. Should the information not exist in such desirable form then it’s a question of getting as close to the ideal as possible within the time and, more importantly, budget available. It would be wonderful to spend the days chasing down the finer points of French Motorcycle manufacture in the 1930’s but it’s finding somebody to pay me to do it that’s the issue and explains why it doesn’t get done too often by anyone.
If there are no academic papers available then books by respected authors are the next best thing and this is where the greater part of the worlds useful information is held. The web is often cited as being an extraordinary encyclopedia but it becomes less useful as a primary source with each passing year as it descends into little more than a glorified shopping channel. Certainly sites like Wikipedia can be useful in providing an introduction to a subject but unfortunately it can only be considered a starting point for investigation rather than the ultimate authority. I am a great fan of its intention but there are just too many people and organisations that would wish to manipulate it to further their own interests which is a terrible but unavoidable consequence of its open nature.
Of course there is always the option of talking to those with a knowledge of the subject but as with the web one has to be careful about considering such information as chapter and verse. Some folk do have a firm grasp of the subject (machinery dealers for instance obviously know their goods) whilst others are just good at talking. It can be a mistake to take everything at face value until you have checked it out against other sources or there are other indicators of the information’s integrity. The classic example I have of this is a photograph of a vintage tractor proudly displaying its build date as being seven years after the model ceased to be produced! It’s a nice image with an interesting story, but discretion insists it’s not used.
If there was one single font of knowledge that was trustworthy and complete then writers and authors would have a much easier time of it, but if such a virtuous spring did come into existence then we would as likely or not be found surplus to requirements and that would be a tragedy of its own.