In my own opinion, a good specialist armed, once and for all, with solid real world knowledge in politics in general and about a given country or region can make surprisingly good analysis, and even accurate forecasts in politics and strategy on the sole basis of open sources.
However, it is understood that, in that case, analysis and forecast are likely to loose in accuracy with time since real world knowledge are not exempt from changes and evolutions. The pace of this observable decrease in accuracy depends of the observed country or region and of the significant political events or upheavals (publicly known or not) happening in them.

I see that correct deductions and forecasts exclusively based upon open sources owe to two ways of analyzing open sources which are: the formal analysis of the open sources per se; and perception management analysis. Other factors linked to the specifics of the observed country (when the object of our attention happens to be a country in particular) will provide the analyst a basis which will help detecting deception attempts. I’ll probably not tell anything new if I say that if the formal content of a given media known for its inclination to practice deception is therefore of relatively poor value, then we may consider it otherwise when analyzing it under the angle of perception management.

What happens when it comes to Army intelligence operations (a field about which I bloody don't know anything)? I admit that things might be ruled otherwise, then; especially if the enemy is not a nation or a country with a government and its own national media and is, or is not, acting as proxy.

So I’ll read this manual with great interest.