[time-nuts] CE Mark

Magnus Danielson magnus at rubidium.dyndns.org
Tue Aug 19 15:13:52 EDT 2008


Poul-Henning Kamp wrote:
> In message <f8eb30bbe974cf183c9189189c874654.squirrel at rubidium.dyndns.org>, "Ma
> gnus Danielson" writes:
>>> In message <001701c901d4$6ae978b0$0300a8c0 at pc52>, "Tom Van Baak" writes:
>>>
>>>> Is software covered by CE regulations, or just hardware?
>>> Just hardware.
>> Wrong. See my other notice.
> 
> I hate to nitpick you, but there are no requirements to software
> anywhere, only to the resulting products behaviour.  If you do
> some of the stuff with software, that's your choice.
> 
> But there are no requirement to software as such.
> 

I agree, but the point is that software is included in the product and 
its operation and cannot be separated from it. If it is configuration, 
software, firmware, hardware or mechanics does not really make much 
difference. If it triggers a behavour which breaks whatever rules and 
regulation there is, the products breaks it. If you alter the behaviour, 
be it configuration, software, firmware, hardware or mechanics in such a 
way that you break the regulation the products breaks it. How the part 
of the product is implemented is irrelevant.

So, the separation between software and hardware is meaningless. 
Software can be an issue. Thus, saying only hardware matters is wrong. 
For most cases software is not much of an issue, but it can be an issue 
and it can be the cause of breaking the regulations for FCC or CE marks.

To cover this point, there is no requirements specific to 
implementation. The rules must be black box oriented. So if we invent 
another implementation method the rules still applies.

Cheers,
Magnus



More information about the time-nuts mailing list