Blanki g cooker isolation under sink to pass EICR

[JohnW2]

Not really. That just highlights the unsatisfactory situation about personal views being legally enforced (re: the 28 days).

Agreed.

Everything is C2.

That would be true were it not for the fact that we currently allow judgement/discretion on the part of the inspector - which (in my opinion) is totally unsatisfactory, particularly given the absence of any regulation of 'inspectors'.

Ok. Then that would explain the different views regarding EICR completion.

Different views regarding risk-aversion will account for some variation of coding, but in other cases is may reflect lack of knowledge/competence (and/or inadequate common sense), or even deliberate fraudulent attempts to 'create work'.

Therefore there is no solution to the problem.

Probably no perfect solution - but ("tight") regulation of inspectors coupled with much more prescriptive/explicit 'rules' as regards coding would go a long way to ameliorating the 'problem'

 

[JohnW2]

Are somethings more dangerous now than they were previously?

I can't think of any - can you?

Probably not but we are more risk averse ? Probably we are

I don't think that there is any 'probably' about it The increase in 'our' risk-averseness has been dramatic during my lifetime to date.

 

[morqthana]

Are somethings more dangerous now than they were previously?

I can't think of any - can you?

Main terminals in CUs.

Domestic LV distribution earthing system.

Plugs & sockets for domestic use.

ELV "adapters"/supplies.

Some white goods with motors.

 

[JohnW2]

Are somethings more dangerous now than they were previously? Probably not but we are more risk averse ? Probably we are

I can't think of any - can you?

Main terminals in CUs. ... Domestic LV distribution earthing system. .... Plugs & sockets for domestic use. .... ELV "adapters"/supplies. ... Some white goods with motors.

Aren't they all things which may have become more dangerous than they were previously because they have changed ?

Although not stated explicitly (by either ebee or myself), I thought it was fairly obvious that we were talking about 'things'/practices which have not changed, but which some people (in some cases regs) believe to be more dangerous now than they were in the past (e.g. inaccessible screwed joints, installations without RCDs etc.).

 

[morqthana]

Fair point.

 

[Harry Bloomfield]

And yet, we have (and have had for a fair number of years now) computers that can not only match, but absoloutely thrash the top chess grandmasters. Brute force number crunching/data lookup/machine learning can go a long way to replicating many of the tasks humans would have required intelligence for.

Yes, but have they achieved that by learning to play chess, or by simply working their way through all every possible move and counter move (high speed calculation, and taking the most beneficial route?

 

[Harry Bloomfield]

That would be true were it not for the fact that we currently allow judgement/discretion on the part of the inspector - which (in my opinion) is totally unsatisfactory, particularly given the absence of any regulation of 'inspectors'.

Discretion is essential, where an individual opinion is involved, and there is no definite line in the sand. Discretion, is a part of the MOT process, it has to be, but at least there is an appeals process, and body to oversee the inspections are carried out fairly, and properly - entirely lacking in the EICR, then landlords over a barrel, for repairs.

 

[SimonH2]

Yes, but have they achieved that by learning to play chess, or by simply working their way through all every possible move and counter move (high speed calculation, and taking the most beneficial route?

Actually, yes they have "learned" to play. The mumber of possible moves even when looking only a few moves ahead quickly becomes too large to handle - which is why early chess programs got got thrashed by even moderate skill players. At the start of a game, there's 20 possible moves, then 20 possible responses - so 400 permutations. Then the number just keep going up ...
Not something I've followed but IIRC modern programs have a view of strategies rather than trying to work out all possibilities.

 

[JohnW2]

Discretion is essential, where an individual opinion is involved ...

That is a statement of the obvious. My "judgement/discretion" and your "personal opinion" relate to the same process - a process which is undesirable, although sometimes unavoidable.

... and there is no definite line in the sand.

That's the problem. There are essentially no 'lines in the sand' with EICRs. The only 'coding instruction' in BS7671 is that absence of RCD protection required by BS7671 should be given "at least a C3" - which, of course, means essentially nothing! Everything else is down to discretion/judgement/personal opinion -particularly as regards what the inspector considers to be 'potentially dangerous'

Discretion, is a part of the MOT process, it has to be ...

Wherever possible, criteria for MOT inspections are explicit/detailed and, where possible, quantified, with only the 'unavoidable' (e.g. judging the severity of corrosion, an 'oil leak' or the readability of a number plate) being left to the inspector's judgement.

.... but at least there is an appeals process, and body to oversee the inspections are carried out fairly, and properly - entirely lacking in the EICR, then landlords over a barrel, for repairs.

Indeed so - that's one of the major problems. Also, in addition to the existence of an appeals process, there is strict regulation of who can undertake MOT inspections, and at least a degree of 'policing' of the the performance of MOT inspectors/'stations' - all of which is missing in relation to EICRs.

 

[JohnW2]

Actually, yes they have "learned" to play. The mumber of possible moves even when looking only a few moves ahead quickly becomes too large to handle - which is why early chess programs got got thrashed by even moderate skill players.

As you say, the number of possible moves in a game of chess is astronomical, probably too large even for current-day computers to sensibly consider - which, as you say, means that strategies other than 'pure brute force' have to be utilised.

I would imagine that a very similar issue arises with SatNavs, since there will again be an astronomical number of possible routes between A and B, particularly if the distance between them is large - yet, again, tiny bits of hardware (and associated software) can achieve that almost instantly.

However, I'm not at all sure that, in either case (chess or SatNav) what a computer does has anything to do with any 'intelligence' (in any sense I would recognise it as such) rather than simply 'doing what it has been programmed (by humans) to do.

Much the same could probably be said of things like 'predictive text'and 'grammar checkers' - which, again,I think only do what they have been programmed to do - even if the end result may 'look like' the exercise of 'intelligence'.

 

[ebee]

A Sat Nav must use intelligence surely?

Mine told me that fastest time and how many miles to get from Lands End to John O Groats very rapidly and I had only actually decided that morning! How clever is that.


( My clumsy attempt of a Joke for the New Year)


Seriously though the point about someone deciding, as with a C1, C2, C3 etc on an EICR etc I think can can only be done that way, because of all of the degrees of faults and numbers of faults blurs some lines therefore less absolute definites in all of the circumstances and some matter of degree. (Intelligence???)

 

[Harry Bloomfield]

I would imagine that a very similar issue arises with SatNavs, since there will again be an astronomical number of possible routes between A and B, particularly if the distance between them is large - yet, again, tiny bits of hardware (and associated software) can achieve that almost instantly.

Mybe something similar to what a human would do - draw an imaginary line on the map, from point A to B, then scan out from there, to where the roads actually run, give each road a value for speed possibilities, then try a few routes for time comparison.

 

[JohnW2]

A Sat Nav must use intelligence surely?

Even if that statement (and what followed) had not been intended as a joke, it would be meaningless without some definition/understanding of what "intelligence" means/is!

Seriously though the point about someone deciding, as with a C1, C2, C3 etc on an EICR etc I think can can only be done that way, because of all of the degrees of faults and numbers of faults blurs some lines therefore less absolute definites in all of the circumstances and some matter of degree.

Given the essentially endless list of possible situations, explicit/prescriptive rules obviously could never be exhaustive, but it would certainly be possible for the system to move a long way in that direction - at least as 'guidance' when hard-and-fast rules could not cover every possible situation.

(Intelligence???)

see above

Happy New Year!
Kind Regards, John

 

[davelx]

Satnavs use a modified version of Dikstra's Shortest Path First algorithm, which is the foundation of the OSPF routing algorithm used by Autonomous Systems in the Internet.

This is not really surprising, as the road system is a network too.

 

[morqthana]

Actually, yes they have "learned" to play.

Not chess, but... https://en.wikipedia.org/wiki/AlphaGo_Zero

And Go is much more complex than chess.

Trevanian Quote

Go is to Western chess what philosophy is to double-entry accounting.

libquotes.com

The mumber of possible moves even when looking only a few moves ahead quickly becomes too large to handle - which is why early chess programs got got thrashed by even moderate skill players.

They don't now, and they're up to several 10's, possibly 100's, of millions of moves per second.