Yes, they are difficult to measure. The only way to measure a brown dwarf mass is if it is in a binary system. The binary frequency of brown dwarfs is low - perhaps 10-15%.
Having found your binary system, then it has to be wide enough to resolve with your telescope, but not so wide that Kepler's third law means it takes centuries to do an orbit. The low mass of brown dwarfs does not help here. To have an orbit of say a decade or less means they have to orbit close to the other object. This in turn means adaptive optics are required.
Having said all that, the masses of quite a few brown dwarfs have been measured, so it is not extraordinarily difficult.
The object talked about in the article is not in a close binary, so it's mass cannot be measured, it can only be estimated indirectly. But here come the difficulties. To estimate the mass requires accurate calculations of the evolutionary tracks of brown dwarfs. The measurements of brown dwarf masses in binaries indicates that we do not have these. Secondly, to use the evolutionary tracks requires a luminosity (and therefore a distance), but because brown dwarfs are cooling objects, you also need an age. In general, unless the brown dwarf can be associated with a group of other stars with known age then this isn't possible. Measuring the temperature of a brown dwarf isn't helpful because the cooling tracks are very close together in the Hertzsprung-Russell diagram, and in any case, it is very difficult because brown dwarfs are nowhere near blackbodies and their atmospheres are still poorly understood.
In this case, it seems that the brown dwarf has been newly associated with an (assumed to be) coeval group of stars at 200 Myr. Thus its luminosity can give a (highly) model-dependent mass. I cannot examine the details as the claimed published research cannot be found!
