M
Matt B
Guest
Nick wrote:
> Matt B wrote:
>> Nick wrote:
>>> Matt B wrote:
>>>> Nick wrote:
>>>>> Matt B wrote:
>>>>>> Nick wrote:
>>>>>
>>>>> I would have thought this was obvious?
>>>>
>>>> Not entirely. There is more to it than simply "providing a net
>>>> benefit" I think.
>>>
>>> What. You find a few counter examples where a man got a ticket for
>>> parking where it didn't cause a problem and you want to get rid of all
>>> parking restrictions?
>>
>> No, I want to only act against parkers for whom it can be proven that
>> they have caused, or are very likely to cause a victim.
>>
>
> But some one who fills only one chamber of a 6 shooter has only a 1/6
> chance of killing someone when he pulls the trigger so that makes it ok?
What is the chance that someone parking with a wheel one inch over the
line creates a "victim"?
>>>>>>> However presumably a lot of them don't have any income or assets
>>>>>>> which may mean the model breaks down.
>>>>>>
>>>>>> Ah, so we let them off, or we change the model?
>>>>>
>>>>> You only use models where they fit.
>>>>
>>>> Should models for crime prevention be designed to reduce crime, or
>>>> should models be created then crimes that they don't fit be left
>>>> unchallenged?
>>>
>>> No, like most models you should only apply them where they fit and not
>>> where they don't.
>>
>> That doesn't answer the question.
>
> It does actually. It means that alternative models have to be used to
> challenge different crimes.
>
>>> The fact that a model works for parking and not for bike thieves does
>>> not mean that the model is wrong it just means that another model is
>>> needed.
>>
>> The model needs to be flexible enough to cover /all/ "crimes".
>
> What? Now you really are talking nonsense.
i.e. you disagree - fine, but say that then.
Why can't "one model" cover all "crimes"[1]? All "crimes" have many
things in common, so should be treated similarly. Departing from /the/
model is only required for "crimes" which are no such thing, but which
for political reasons, spin, or whatever, or to foster public support,
or to hide or mask some other deeper problem, require retribution, or
punishment of some sort to elevate them to "crime" status.
> I use various models for
> getting from A-B. From one side of the bed to the other I generally roll
> over. From the bedroom to the kitchen I generally walk. From my house to
> work I ride a bike. From England to the USA I go by plane.
You could say you use one model. Your input is a "journey" you process
it based on distance, and possibly other criteria, the output being the
method of travel.
That can work too for dealing with "crimes".
<ramble>
The input is an "action" (taking someone's bike to sell, parking at the
roadside, spinning the chamber and pulling the trigger), the process is
establishing who the victim(s) is/are, or weighing the likelihood of
creating victims, establishing the size or likely size of the cost to
the victim(s), deciding, if the suspect is culpable, and what the
penalty should be. Deciding the threshold of "cost to the victim(s)"
above which the action becomes a "crime" is a political job. To ensure
fairness and consistency across actions, the threshold should be fixed,
not varied dependent on the "action". The calculation of the cost to
the victim (which could be "society") should be done dispassionately,
scientifically, and independently of political influence. The
"punishment" should also result from an objective calculation (inputs
might include "wickedness", intent, "social factors", extenuating
circumstances, etc.), with the algorithm for deciding the mix of "fine",
"prison", "community service", whatever, and whether the penalty is
suspended also politically defined.
</ramble>
If a victim is deprived of, say, a bike thus costing him a total of,
say, £1000, that action should be dealt with in exactly the same way as
stealing £1000 from the same person, or negligently causing £1000 damage
to his car, and the culprit dealt with in the same way.
> But you tell me one model has to be flexible enough for all cases. Yeah
> right.
It depends how well you define the model.
>> I don't care whether it is a bike thief, a bad parker, or someone
>> carrying a ladder - I don't want to be a victim of their actions or
>> their negligence - and if I am, I want whoever it was to be dealt with
>> proportionately and consistently, and by the same process. What I
>> don't want is a new law to specifically ban ladder carrying where a
>> certain number of lines of a certain colour are painted on the ground.
[1] By "crime" I mean an action performed either knowing that a victim
would be inevitable, or knowing that a victim would be likely, or
reckless as to whether one would result or not.
--
Matt B
> Matt B wrote:
>> Nick wrote:
>>> Matt B wrote:
>>>> Nick wrote:
>>>>> Matt B wrote:
>>>>>> Nick wrote:
>>>>>
>>>>> I would have thought this was obvious?
>>>>
>>>> Not entirely. There is more to it than simply "providing a net
>>>> benefit" I think.
>>>
>>> What. You find a few counter examples where a man got a ticket for
>>> parking where it didn't cause a problem and you want to get rid of all
>>> parking restrictions?
>>
>> No, I want to only act against parkers for whom it can be proven that
>> they have caused, or are very likely to cause a victim.
>>
>
> But some one who fills only one chamber of a 6 shooter has only a 1/6
> chance of killing someone when he pulls the trigger so that makes it ok?
What is the chance that someone parking with a wheel one inch over the
line creates a "victim"?
>>>>>>> However presumably a lot of them don't have any income or assets
>>>>>>> which may mean the model breaks down.
>>>>>>
>>>>>> Ah, so we let them off, or we change the model?
>>>>>
>>>>> You only use models where they fit.
>>>>
>>>> Should models for crime prevention be designed to reduce crime, or
>>>> should models be created then crimes that they don't fit be left
>>>> unchallenged?
>>>
>>> No, like most models you should only apply them where they fit and not
>>> where they don't.
>>
>> That doesn't answer the question.
>
> It does actually. It means that alternative models have to be used to
> challenge different crimes.
>
>>> The fact that a model works for parking and not for bike thieves does
>>> not mean that the model is wrong it just means that another model is
>>> needed.
>>
>> The model needs to be flexible enough to cover /all/ "crimes".
>
> What? Now you really are talking nonsense.
i.e. you disagree - fine, but say that then.
Why can't "one model" cover all "crimes"[1]? All "crimes" have many
things in common, so should be treated similarly. Departing from /the/
model is only required for "crimes" which are no such thing, but which
for political reasons, spin, or whatever, or to foster public support,
or to hide or mask some other deeper problem, require retribution, or
punishment of some sort to elevate them to "crime" status.
> I use various models for
> getting from A-B. From one side of the bed to the other I generally roll
> over. From the bedroom to the kitchen I generally walk. From my house to
> work I ride a bike. From England to the USA I go by plane.
You could say you use one model. Your input is a "journey" you process
it based on distance, and possibly other criteria, the output being the
method of travel.
That can work too for dealing with "crimes".
<ramble>
The input is an "action" (taking someone's bike to sell, parking at the
roadside, spinning the chamber and pulling the trigger), the process is
establishing who the victim(s) is/are, or weighing the likelihood of
creating victims, establishing the size or likely size of the cost to
the victim(s), deciding, if the suspect is culpable, and what the
penalty should be. Deciding the threshold of "cost to the victim(s)"
above which the action becomes a "crime" is a political job. To ensure
fairness and consistency across actions, the threshold should be fixed,
not varied dependent on the "action". The calculation of the cost to
the victim (which could be "society") should be done dispassionately,
scientifically, and independently of political influence. The
"punishment" should also result from an objective calculation (inputs
might include "wickedness", intent, "social factors", extenuating
circumstances, etc.), with the algorithm for deciding the mix of "fine",
"prison", "community service", whatever, and whether the penalty is
suspended also politically defined.
</ramble>
If a victim is deprived of, say, a bike thus costing him a total of,
say, £1000, that action should be dealt with in exactly the same way as
stealing £1000 from the same person, or negligently causing £1000 damage
to his car, and the culprit dealt with in the same way.
> But you tell me one model has to be flexible enough for all cases. Yeah
> right.
It depends how well you define the model.
>> I don't care whether it is a bike thief, a bad parker, or someone
>> carrying a ladder - I don't want to be a victim of their actions or
>> their negligence - and if I am, I want whoever it was to be dealt with
>> proportionately and consistently, and by the same process. What I
>> don't want is a new law to specifically ban ladder carrying where a
>> certain number of lines of a certain colour are painted on the ground.
[1] By "crime" I mean an action performed either knowing that a victim
would be inevitable, or knowing that a victim would be likely, or
reckless as to whether one would result or not.
--
Matt B