``This solution does paradoxically not always provide the lowest energy efficiency neither the lowest `sky glow'. The reflection of the road and its surroundings must be taken into account when assessing the total upward light flux and thus `sky glow', but please not that there are different 'types of sky glow'(see remark later).'' Sky glow and total upward light flux are not simply proportional. Sky glow is very sensitive to the direction of upward emissions. So it is an error to write "and thus `sky glow'". ``Furthermore these designs are also not de facto the most energy efficient ones because horizontal light beams can contribute more to the visibility by the horizontal observer, especially when taking into account the specular road and object reflection.'' Horizontal beams from road luminaires produce no useful illumination, they produce just glare. The only situation, where they help, are traffic lights... (should be of course fainter at night then during the day). The angle, where the benefit of rays may become larger than the hindrance to visibility (due to glare), is below 72 degrees from nadir, or 18 degrees below horizontal direction! That's rather far from being horizontal: a 18 degrees slope is a 32 per cent slope, a really steep one! This is a well established rule for bidirectional lighting, every lighting expert knows it. ``Flat glass luminaires don't allow transmission of low angle (near horizontal) incident light rays due to increased reflection on the inner glass surface (e.g. Brester angle effect). Therefore flat glass or 'full cut off' luminaires are unable to offer wide beam luminaires.'' Even at the very limiting angle, the reflection on a flat glass (which is no prerequisite for 0 cd/klm luminaires) makes mere 32 per cent. At those angles, where the light is definitely useful, like at 65 degrees from nadir, the reflection is only 21 per cent, at 60 degrees it's just 16 per cent. Any curved refractor has 9 per cent reflection as a minimum. Brewster's (this is the spelling) angle has nothing to do with such large angles of incidence. It's 56 degrees for common glass, and this effect _contributes_ to the glass transparency at angles at 50 to 60 degrees. ``Luminaires with curved glass (e.g. semi-cut off luminaires) or refractors can create these more horizontal light beams more effective and should still be considered.'' Paradoxically, this sentence begins with a true claim. Flat glass (when employed, what is no ``must'' for 0 cd/klm luminaires) filter out rays at glaring angles nicely. They let just 1/4 of them escape the luminare at 85 degrees from nadir. Using flat glass helps road visibility a lot, without any demands to luminaire designers, simply by laws of physics. Luminaires with curved glass have a major drawback in this sense. Of course, when hidden in the opaque top of the luminaire, they may still be considered, but require very carefulr design of the mirrors inside the luminaire (and a lamp deep enough in it). ``The full cut-off luminaires have in general also more internal reflection and have therefore a lower efficiency.'' Efficiency in terms of total emissions from a luminaire, if the (organic) clear refractor is still clean and non-weathered, yes. But efficiency in terms of the amount of light hitting the target, not necesarilly. On the contrary, the harmful light avoided above 72 degrees (reflected backwards into the luminaire) can be reused within the luminaire to go to useful directions. ``Also it is required to install these luminaires with a lower spaceto-height ratio (SHR) due to the smaller output light beam, this means that more lighting poles are required.'' Luminaires producing 0 cd/klm at 90 degrees and above (the `fully shielded' ones) can be employed even for largest useful pole spacings, at SHR 5 or even 5.5. Larger spacings lead to unacceptable glare. Just a bit smaller spacings, at about 4.5, offer quality, low-glare lighting. Here the best `fully shielded' luminaires perform better than those which don't fit in this category. ``From chapter 4 it can also be deducted that lower wattage HID lamps have in general a lower efficiency and this is also not beneficial for solutions with 'flat glass' or 'full cut off' luminaires because of the increased number of light poles.'' Would be true for SHR >5 installations with suboptimally chosen fully shielded luminaires. ... The rest of the 6.1.4 text just perpetuates the demonstration of not understanding to the physics of road illumination and of spreading of light in nature. These topics have been very thoroughly investigated and described in a series of papers by Pierantonio Cinzano. These need to be given as references and the correct conclusions from them included into the 6.1.4 section instead of the current vague and confused (at best...) paragraphs.