Darren New <dne### [at] sanrrcom> wrote:
> >> If you stop people for driving drunk even when you can't tell they're drunk, 
> >> how does that reduce accidents?
> > 
> >   Now you are not arguing only against me, but against the law of Finland.
> > Random sobriety tests are a common practice here, and nobody is complaining.

> That doesn't answer the question, tho. It doesn't matter whether it's the 
> law or not.

  The answer to the question is that it reduces accidents caused by drunk
drivers by both catching them before they have an accident and as a deterrent.

> >> You can disagree with simple baysian inference math, but you'd be wrong.
> >   Then I suppose our law is stupid for doing that...

> For one, it's a different circumstance, as I explain above. For another, do 
> you actually know if they profile the type of car you drive or anything like 
> that in pulling you over?  If not, you're begging the question.

  I said that I wouldn't be surprised if the police stopped more vehicles
of certain types (such as SUV's) and less of other types (such as buses and
trucks). However, that wasn't really the point.

> No, I'm writing as if you brought it up as a good example of why your ideas 
> work. Why did you bring it up if you don't think it's a good idea?

  I brought up the random sobriety testing as a counter-example to the idea
that testing people even when the police has no reason to suspect any crime
is wrong. Random driver sobriety testing is an example where testing people
even without justifiable cause makes sense.

  More specifically, "random testing is wrong" as an argument against testing
random people for citizenship is not a valid argument because "random testing
is wrong" is an invalid argument. There's a counter-example: Random sobriety
tests are not wrong.

  That doesn't mean that testing random people for citizenship is right.
It means that the *argument* that general random testing is wrong is invalid.

  That was the point of the example.

  (Of course *you* might disagree with random sobriety tests not being wrong,
but in Finland nobody is complaining, so I'm not the only one who agrees that
they are important.)

> See, this is where I get annoyed with these sorts of discussions with you. 
> You make assertions, you're argued against, and rather than seeing "Yes, I 
> see where that might be valid," you try to act like you weren't saying what 
> you did, or something.  It makes you sound dishonest, as well as unwilling 
> to admit you may be mistaken, at which point why continue the conversation?

  Taking into account the *context* where I make the assertions might help
understanding what I'm talking about.

-- 
                                                          - Warp

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> Warp wrote:
> > Darren New <dne### [at] sanrrcom> wrote:
> >>>   "Pointing out"? Maybe you are, but I just can't follow your logic.
> > 
> >> It's simple math.
> > 
> >   So you are, indeed, nitpicking on my math rather than the overall idea
> > I was expressing with my example.

> What was your overall idea?  I thought it was that if you use racial 
> profiling to target people of races who make up the majority of illegal 
> immigrants, then you'll have a more efficient way of finding illegal 
> immigrants, right?

  I said that if the vast majority of illegal immigrants can be distinguished
by how they look, it makes sense to concentrate more on those people than on
people who are much less likely to be illegal immigrants.

> >>>   (Sure, my suggestion of devoting x% of resources for x% of illegals is
> >>> cold, hard, inhuman math, and there are other more humanistic sides to the
> >>> whole issue. I'm not denying that. I'm just arguing that from a resource
> >>> distribution point of view it would make sense.)
> > 
> >> A percentage is a ratio between two numbers. You don't have a simple 
> >> "percentage of illegals."  Percentage of *what population* are illegals?
> > 
> >   Stop nitpicking on the numbers, and start discussing the idea I'm
> > expressing.

> I'm trying to. You're arguing that using statistics would make the process 
> more efficient. I'm showing you that mathematically, no, it wouldn't.

  No, you aren't. You are simply taking my simplified example and directly
applying it to a more complex situation and showing how it doesn't work
there. Then rather than readjusting the formula to account for that more
complex situation you simply claim that the whole idea of using statistics
is flawed. It's still a non-sequitur.

  If you argued that the idea doesn't work for practical reasons (eg. because
it causes unrest and riots), then that could be more valid of an argument.
Arguing about the math isn't the way to do it.

> Then 
> you say "Stop arguing the details of the math. Discuss how the math can make 
> things more efficient, instead."  You're not making sense.

  No. I'm saying that stop nitpicking on the specific example I gave
(using the 90% example, etc, because that's what you are doing), and start
discussing the idea this example is trying to convey.

> >   Although if your argument is "your math does not work on this more
> > complicated case, hence your idea is wrong", then I suppose there is
> > nothing to discuss. If you disagree with the idea, then say so rather
> > than nitpick about some percentages.

> I disagree with the idea that using statistics to stop people who have given 
> no indication of wrongdoing makes the process any more efficient. I disagree 
> *because* your math is wrong. When you use the actual math, it turns out 
> that it's less efficient.

  The opposite of concentrating resources on likely suspects is to spread
the resources equally to all people. This means that more resources will
be spent on people who are more unlikely to be illegal immigrants. Exactly
how do you get from this that it's actually more efficient to do it like
this? If more resources are spent on people who are more unlikely to be
illegals, then those resources are away from catching those people who are
more likely.

  What I think is happening here is that we are talking about different
things. I think you are talking about "use resources to check random
people" vs. "use resources to check only those who are very suspicious
because of eg. having been reported by someone", while what I am talking
about is "distribute random checks according to statistics" vs. "distribute
random checks equally".

> And you keep saying "ID" as if that's the same as "Proof of legal 
> residence."  It isn't.

  Maybe it could be a good idea to change that, then? It's not like it's
something unheard of.

> > (Why check them if it's completely inconsequential whether
> > they have it or not?)

> Why indeed? That's exactly why we have laws saying the police don't get to 
> do that.

  As I have commented much earlier in this thread: Because it gives the
wrong message to people thinking about moving illegally to the country,
as well as the human traffickers who are getting big money for getting
them there. It gives the message "once you are inside the borders of the
country, you are pretty much safe because nobody will ask you if you
really have the right to be there".

  (But as I have also commented many times, aggressively imposing the
immigration laws by using strict measures may anger some people, which
might not be politically wise. So it's a lose-lose situation.)

-- 
                                                          - Warp

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> Please rephrase what you were actually trying to say

  Again? I have already said it many times.

> Instead, you should say "Gee, I didn't realize I had the math wrong. Let me 
> try again." instead of complaining that I pointed out a flaw in your 
> thinking caused by a flaw in your math.

  You are still nitpicking about the "90%-90%" math example.

  Fine, nitpick all you like. If you don't want to discuss then you can
think whatever you want.

> And you still didn't answer whether you'd agree that it might be more 
> efficient to target people in churches than people who look central 
> american. :-)

  I found it so incoherent and outlandish that I didn't know what to say.
Maybe that's what you were after.

-- 
                                                          - Warp

Post a reply to this message

>>> Basically, people here are generally against getting arrested before a 
>>> cop knows a crime has been committed at all.
>>
>>   I'm not talking about arresting someone. I'm just talking about 
>> *checking*
>> the alcohol levels of drivers.
>
> That's being detained, at the least. We have this whole "innocent until 
> proven guilty" thing going on here. If there's *no* evidence you've done 
> anything wrong, why would you need to prove your innocence more than that?

In practice (in the UK at least) the police can always find some evidence 
that allows them to pull you over and breathalyse you if they want to.  No 
driver is perfect, and they probably only need to follow you for half a mile 
until you drive a few inches too far to one side, signal 1 second too early, 
go 1 mph too fast or too slow, then according to them they have "reasonable 
belief that you are under the influence"...

Post a reply to this message

Warp wrote:
>   That doesn't mean that testing random people for citizenship is right.
> It means that the *argument* that general random testing is wrong is invalid.

Except we aren't talking about random testing. You're arguing that based on 
statistics, racist testing should be applied.

>   (Of course *you* might disagree with random sobriety tests not being wrong,
> but in Finland nobody is complaining, so I'm not the only one who agrees that
> they are important.)

People complained less about random testing than about racists testing here too.

-- 
Darren New, San Diego CA, USA (PST)
   Linux: Now bringing the quality and usability of
   open source desktop apps to your personal electronics.

Post a reply to this message

Warp wrote:
>   I said that if the vast majority of illegal immigrants can be distinguished
> by how they look, it makes sense to concentrate more on those people than on
> people who are much less likely to be illegal immigrants.

And you're ignoring the fact that
A) they can't be, because they live where the majority of people look like 
them, and
B) there are dozens of traits more useful for distinguishing legal from 
illegal residents than looks.

>> I'm trying to. You're arguing that using statistics would make the process 
>> more efficient. I'm showing you that mathematically, no, it wouldn't.
> 
>   No, you aren't. You are simply taking my simplified example and directly
> applying it to a more complex situation and showing how it doesn't work
> there. 

Not at all. I'm showing you (A) it's more complicated than you think, and 
(B) not related to "looks".   I'm offering you examples of other 
distinguishing features (besides genetics) that will get you a much better 
chance of efficiently catching the criminal, and showing how if you base it 
on something irrelevant to immigration status (like looks or religion) you 
wind up with prejudiced yet ineffective enforcement.

>   If you argued that the idea doesn't work for practical reasons (eg. because
> it causes unrest and riots), then that could be more valid of an argument.

That too.

>   No. I'm saying that stop nitpicking on the specific example I gave
> (using the 90% example, etc, because that's what you are doing), and start
> discussing the idea this example is trying to convey.

I did. You ignore when I discuss the general idea.

>   The opposite of concentrating resources on likely suspects is to spread
> the resources equally to all people.

Now you're back-pedaling more. You were talking about "likely suspects" 
before. Even in this very post, you're using "looks mexican" as "likely 
suspects." I'm using the math to show you how "looks mexican" isn't the same 
as "likely suspects."  Yet you're ignoring that fact.

Concentrate on likely suspects?  No problem!

Concentrate on people who look mexican?  Bad idea!

>   What I think is happening here is that we are talking about different
> things. I think you are talking about "use resources to check random
> people" vs. "use resources to check only those who are very suspicious
> because of eg. having been reported by someone", while what I am talking
> about is "distribute random checks according to statistics" vs. "distribute
> random checks equally".

Yet you haven't mentioned any "statistics" other than "looks mexican."

>   I found it so incoherent and outlandish that I didn't know what to say.
> Maybe that's what you were after.

Why did you find it outlandish?  If 90% of illegal immigrants look mexican, 
and 90% of illegal immigrants are christian, why is it better to focus on 
people who look mexican than to focus on people who look christian?

The very fact that you're calling it outlandish is *exactly* the point I'm 
making. *That* is why focusing on people who look mexican is racist and not 
just statistics.

*This*, right here, is the crux of where you are wrong. You're asking me to 
adjust my nit-pick of your math so that you turn out right. OK, if 80% of 
illegals look mexican, and 90% are christian, would you still feel it's 
outrageous to stop people going into churches and ask them for proof of 
citizenship?  If not, then your proposal hasn't anything to do with 
statistics of efficient policing.

If you agree with that, then maybe you'll agree that when we find most 
illegals shop at Wal-mart, buy groceries from food stores specializing in 
mexican food, and have jobs paying minimum wage, then we should find people 
who engage in all three of those behaviors and go around banging on their 
doors, asking for proof of citizenship?

Why is that a bad idea?

>   Maybe it could be a good idea to change that, then? It's not like it's
> something unheard of.

Welcome to democracy.

> country, you are pretty much safe because nobody will ask you if you
> really have the right to be there".

No. Random people walking down the street won't have the right to ask you if 
you're doing nothing wrong.  Everyone whose behavior is affected by them 
being illegal already has to prove their legality.

-- 
Darren New, San Diego CA, USA (PST)
   Linux: Now bringing the quality and usability of
   open source desktop apps to your personal electronics.

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> Warp wrote:
> >   That doesn't mean that testing random people for citizenship is right.
> > It means that the *argument* that general random testing is wrong is invalid.

> Except we aren't talking about random testing.

  The subject which led to this sobriety test comparison was about the
morality of testing people without probable cause. I simply pointed out
that testing people without probable cause is not *always* immoral, as
there exists at least one legit example where it's not.

  So yes, in *this particular context* we were talking about random testing.
The original post might not have been talking about that, but this comparison
was not directly related to that.

  You are trying to invalidate my example by dragging it back to the
original subject (in other words, that Arizona law) and comparing it
directly to that, rather than what really prompted the comparison.

> You're arguing that based on statistics, racist testing should be applied.

  That's a different context, not related to this sobriety test example.

  Also, I don't appreciate your denigrating terminology on what I may be
arguing for.

> >   (Of course *you* might disagree with random sobriety tests not being wrong,
> > but in Finland nobody is complaining, so I'm not the only one who agrees that
> > they are important.)

> People complained less about random testing than about racists testing here too.

  No, your original argument which prompted this was that testing people
with no probable cause is always wrong. Don't change the context here.

-- 
                                                          - Warp

Post a reply to this message

Warp wrote:
>> You're arguing that based on statistics, racist testing should be applied.
>   That's a different context, not related to this sobriety test example.

Fair enough.

>   Also, I don't appreciate your denigrating terminology on what I may be
> arguing for.

Well, when you say "we should test people who look like illegal immigrants 
look", then that's pretty much the textbook definition of racist. You're 
saying "we should act as if those with the same genetics as criminals are 
more likely to also be criminals" is racist.  I don't know any other word 
for it, especially when you're claiming that genetics is a factor, but not 
religion or shopping habits or anything like that.

-- 
Darren New, San Diego CA, USA (PST)
   Linux: Now bringing the quality and usability of
   open source desktop apps to your personal electronics.

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> Warp wrote:
> >   I said that if the vast majority of illegal immigrants can be distinguished
> > by how they look, it makes sense to concentrate more on those people than on
> > people who are much less likely to be illegal immigrants.

> And you're ignoring the fact that
> A) they can't be, because they live where the majority of people look like 
> them, and

  You are using a different meaning of the word "distinguish" than I am.

  Take 100 illegal immigrants. If 90 of them are of mexican origin and
the rest are of European origin, you certainly *can* distinguish the majority
of them by how they look.

  That's different from what you are arguing, which is: Take 100 mexicans.
You know that 10 of them are illegals. Try to distinguish them by how they
look.

  What I am saying is:

  Take 100 mexicans and 100 canadians. You know that there are approximately
9 illegals of mexican origin and 1 of canadian origin (hence "90% of illegals
are of mexican origin", as in the example). Hence if you check all the 100
mexicans you will most probably get the 9 illegal mexicans even if you miss
that 1 illegal canadian.

  However, if you checked the 100 people completely at random, you will
only catch about 4 or 5 illegals in average.

  So exactly where is the flaw in this math?

> B) there are dozens of traits more useful for distinguishing legal from 
> illegal residents than looks.

  That may well be true.

> Concentrate on people who look mexican?  Bad idea!

  Maybe it's a bad idea, but not because of statistics, but because of
people's feelings and sense of privacy. That has nothing to do with math.

> >   I found it so incoherent and outlandish that I didn't know what to say.
> > Maybe that's what you were after.

> Why did you find it outlandish?  If 90% of illegal immigrants look mexican, 
> and 90% of illegal immigrants are christian, why is it better to focus on 
> people who look mexican than to focus on people who look christian?

  You might as well argue "100% of illegals look like people, so why not
concentrate on people?"

> The very fact that you're calling it outlandish is *exactly* the point I'm 
> making. *That* is why focusing on people who look mexican is racist and not 
> just statistics.

  What do you even mean by "racism"? Usually "racism" implies prejudice and
forming preconceptions about people based on their ethnicity. A policeman
thinking "most mexicans are criminals and deserve to be beaten" is racism
because it's a prejudiced opinion based on ethnicity.

  Statistical profiling based on ethnicity might be *technically* labelled
"racism" in the sense that it segregates people based on ethnicity. However,
if it's purely hard neutral statistics, there's no prejudiced intent. It's
simply the result of keeping up records and calculating the statistics from
them.

  One could argue that keeping statistics based on ethnicity is wrong, but
curiously that happens only if the statistics make *certain* ethnicities
look bad, while with other ethnicities it's ok. (For example, if a statistical
study showed that, let's say, Japanese people are in average more intelligent
than Finnish people, it wouldn't cause any commotion. However, heaven forbid
if such a statistic said that, for example, Finnish people are in average
more intelligent than, let's say, Nigerians. The outrage! That kind of
statistic would immediately be labelled "racist". Of course if it happened
to be the other way around, nobody would complain.)

> *This*, right here, is the crux of where you are wrong. You're asking me to 
> adjust my nit-pick of your math so that you turn out right. OK, if 80% of 
> illegals look mexican, and 90% are christian, would you still feel it's 
> outrageous to stop people going into churches and ask them for proof of 
> citizenship?  If not, then your proposal hasn't anything to do with 
> statistics of efficient policing.

  Well, if we have 100 christians and 100 non-christians, and we know that
approximately 9 of the christians are illegal immigrants and about 1 of the
non-christians is, then checking 100 christians would produce a higher
success rate than checking 100 randomly chosen people. It works the same.

  Statistically speaking it would be better to check only the christians
if we want to maximize the success rate of 100 random samples.

  You are arguing that "there are too many christians and there aren't
enough resources to check them all, hence it would be inefficient".
I didn't ever claim that checking all people of a certain group would be
*the most efficient* way. I said that it would be *more efficient* than
purely random checks. There's a difference between "more" and "most".

> >   Maybe it could be a good idea to change that, then? It's not like it's
> > something unheard of.

> Welcome to democracy.

  Democracy implies no citizenship-proving ID?

-- 
                                                          - Warp

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> Well, when you say "we should test people who look like illegal immigrants 
> look", then that's pretty much the textbook definition of racist.

  Except that's not what I'm saying. I'm saying that concentrating resources
on statistically more probable people is going to give a better success rate
than spreading the resources evenly on everybody. If the distinguishing
feature happens to be ethnicity, that doesn't change anything. You can call
it "racism" if you want. I am not prejudiced against any group of people
based on their ethnicity, nor am I claiming that some ethnicities are more
likely to be criminals than others, so I don't really appreciate the
insinuation.

> You're 
> saying "we should act as if those with the same genetics as criminals are 
> more likely to also be criminals" is racist.  I don't know any other word 
> for it, especially when you're claiming that genetics is a factor, but not 
> religion or shopping habits or anything like that.

  Now you are outright insulting me. Nowhere have I said that.

-- 
                                                          - Warp

Post a reply to this message