Warp wrote:
>   Your argument against "most illegal immigrants look mexican, so let's
> check the  mexicans" is, basically "assume that somewhere 90% of citizens
> are of mexican descent". That somehow invalidates the claim? I don't get it.

There's no reason to check mexicans *unless* the ratio of illegal immigrants 
that look mexican to legal residents that look mexican is especially high, 
*and* the number of residents that look mexican is especially low.

>   "Pointing out"? Maybe you are, but I just can't follow your logic.

It's simple math.

It's not more efficient to check 100,000 brown people, 1% of whom are 
illegal, than it is to check 10,000 brown people who are actually suspected 
of being illegal.

It's not more efficient to check the race with the highest absolute number 
of illegal immigrants without regard to the number of legal immigrants.

It's simple math. Your claim that disallowing profiling makes the process 
less efficient is factually incorrect.

>> Yes. But that's not what you suggested. You suggested checking in the ratios 
>> of illegal immigrants of different backgrounds, not legal immigrants or 
>> total residents of different backgounds.
> 
>   I was assuming a completely hypothetical ratio of 90% of illegal immigrants
> being mexican. Hence it makes sense to devote about 90% of resources to
> catching them. I really don't understand how the coincidence of 90% of the
> legal population of some place being of mexican descent refutes that idea.

Then you're missing the basic math.

>   (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?

>> If someone said "The town is 70% brown and 30% white, so that's the ratio 
>> we'll check in" then nobody would complain. But you're saying the "the 
>> illegals are 70% brown and 30% white, so let's check in that ratio", which 
>> is a different statement.
> 
>   Yes, I suppose I am saying that. You confused me with having the 90% of
> population being of mexican descent immediately after I gave an example of
> 90% of illegal immigrants looking like mexicans. I still don't understand
> what is it that your argument "proves".

It proves that the percentage of illegal immigrants who look mexican vs 
percentage of illegal immigrants who don't look mexican tells you *nothing* 
about the population of people you should be profiling.  If you want to 
catch more illegal immigrants by interviewing people without cause, you 
*must* know more information than simply the percentages of illegals from 
each country. You *also* have to know the percentages of legals from each 
country.

>> I'm simply showing the flaw in your logic.  You're suggested profiling 
>> *fails* to improve the efficiency, unless you use more information to tune it.
> 
>   Am I understanding correctly that you are arguing against my math rather
> than against the basic idea I was expressing (in other words, to distribute
> resources according to illegal immigration country of origin distribution,
> regardless of whether that's seen as "racism" by some people)?

No. I'm saying that distributing resources according to illegal immigration 
country of origin distribution *is* the flawed math.  Your basic idea is 
flawed because it's based on math that's flawed. I'm trying to point out how 
the math is flawed and hence how the results of applying that math won't 
have the effect you think it will.

>   Out of curiosity: What *should* a police officer do, in your opinion, if
> he asks someone for their ID (or whatever proof of citizenship) and he has
> nothing to show?

It depends why he's asking you. If he stops you while you're driving and you 
have no ID, he could do a number of things.  He would normally take you to 
the police station, book you (i.e., record your identifying information and 
issue you a summons to appear in court to prove you have a driver's 
license), and then call someone to come get you.

If he's asking you because he just wants to know, and you refuse, then he 
should say "have a good day."

In some states, they've decided that you're required to tell the policeman 
your name, but you don't have to prove it.

Same as if he comes to your door. If he says "May I come in?" and you say 
"No," then he should stand outside and talk to you. If he has to *ask*, then 
you get to say no. Makes perfect sense.

>   I suppose I can't deny that sounds reasonable. Basically it means that
> illegal immigrants are not actively "hunted down", but if they try to
> actually *do* something they won't be able to if they can't prove they
> have the right to (as legal citizens or immigrants).

Exactly.

>   Of course this assumes that nobody employs them without asking for their
> papers and without informing the authorities... I'm assuming that's one of
> the biggest problems in the US.

Yes. And it's already illegal to employ them without asking for their papers 
and informing the authorities you've done so. (I.e., paying taxes on their 
salaries, etc.) But police don't enforce *that* law, because it's the 
*white* people who are making money by employing them. Actually, that law 
*does* get enforced, which is how we know how many illegal immigrants are 
here. But not everyone gets caught, because not every employer reports them, 
and the fines for not reporting them are lower than the profit you make by 
exploiting illegal immigrants.

-- 
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