That definition is the same as it was 10 years ago and is still unchanged because when an aircraft is de-iced and anti-iced and it continues to snow, by the time it begins its take-off roll, the wings may very well be covered in snow, but snow that does not adhere to the critical surfaces of the aircraft thanks to the anti-icing fluid.
This is a totally incorrect and misleading statement which, in my opinion, should be removed before the less well informed start to believe it's true..!!
The glycol present in the anti-icing fluid will effectively melt the snow falling on it and the fluid will become progressively more and more diluted as time passes. Eventually the glycol will be diluted to a level where it becomes totally ineffective and the snow will start to form or accumulate on the treated surfaces. At this time the so called 'Holdover Time' will have run out and the aircraft will require a further de-icing/anti-icing treatment.
While under certain conditions any anti-icing fluid residues remaining in contact with the structure may prevent the accumulating snow from bonding or freezing to the surface of the aircraft, there is no guarantee of this and certainly no guarantee that this partially melted snow will effectively flow off on any anti-icing fluid residues that may be still present.
To quote from Section 3.15 in Edition 26 of the AEA Recommendations for De-icing/Anti-icing Aeroplanes on the Ground, "..Holdover time will have effectively run out when when frozen deposits start to form/accumulate on treated aeroplane surfaces.." At this time a further de-icing/anti-icing treatment would of course be required.