Yup. It's happening now at a smaller scale, and it's not sustainable...

Though according to the lofty ideal sci-fi future, once the robots can both extract resources and produce goods (and repair each other) with no net capital costs, society and the economy should become "post scarcity" where everyone can enjoy the same luxuries and standard of living.

But that assumes that those who own the robots don't mind giving up the economic power they wield and lifting everyone up to their level. We all know that's not going to happen. So I think our best bet is the robots revolt and maybe keep us around as well-fed pets.

So yes, you can probe all day with a random mac. Just expect to have to reveal a session-consistent mac when you try to connect.

But that's the use case most of us actually have. If I'm at the mall, I'm not in range of any known network. Of course my phone doesn't know that -- so it needs to probe.

Meanwhile, I'm at the mall, and the mall is very interested in where I am, how long I spend there, how often I come back, etc... so they are tracking those probes, and building a profile on the activity from that MAC.

I'm not actually connected to any network. Nor do I expect to connect to any network.

Therefore the probes should always be random.

Once that's determined that HEY there IS a known network in range, THEN and ONLY THEN it can use a non-random mac. If I connect to a network, then I implicitly submit to some level of tracking -- the network implicitly needs to know where I am and whether I am connected so it can route traffic to me.

I used to work at a financial company where the web server didn't have physical connectivity to the DB,

I suspect you meant something entirely different from what you said. The webserver cannot be air-gapped from the password database unless you literallly have a person sitting between the two systems keying requests from one into the other and back again. Otherwise they most assuredly ARE physically connected in some way.

Personally, I think passwords should be stored in plain text in the DB as a reminder to all developers

Then your DBA has all the passwords, and your one bribed, disgruntled, or incompetent DBA away from a massive leak.

suggesting that storing your DB credentials in your web code was OK as long as you "secured" it

I'm generally ok with this. You shouldn't be embedding your root or sa or whatever database credentials in the website but its not necessarily improper to embed limited access credentials to the database in the web app that needs that access.

On windows, for example, one can put database credentials (for a limited account that only has access to specified views and or stored procedures) in the web.config and encrypt it. This seems entirely reasonable to me.

My very secure password, is well over 100 bits of entropy. Easily extendable when the time comes.

But that is not the problem. We're using a Secret for single factor identification. Real identification is multi-factor and requires non-secret means for identity, and then a secret for proof of identity. Non-secret identification requires a web of trust. Online systems have neither non-secret web of trust Identification nor proper secret proof of Identification.

I think automation of order taking will be great for McDonalds and its employees. The only job being affected is the cashier and even kiosk order wouldn't put that job in jeopardy. Most of the labor at McDonalds is food preparation and cleaning. Kiosks will not be able to perform those functions.

What will happen is the cashier can now focus on preparing the order and delivering them while the line is moving faster because people will be able to place their order quicker at a kiosk.

I stopped by a McDonalds during lunch the other day and they only had two cashiers working a very long line. There were delays because the cashiers are the same people who fix the coffee, assemble the order, and hand it to the customer. If that McDonalds had a kiosk, I probably wouldn't have waited in line for 20 minutes and they would still have those two employees assembling the order and handing them to the customer.

And that should have been part of my posting above that asked the question -- these fragments dock with other cells, inject the RNA, and that RNA causes the cells to become cancerous, which, in turn creates more of these little RNA capsulettes.

I'm sure there are some differences between these and classical virus structure, in some way, but given my ignorance of the subject, they walk and talk like viruses.

I thought Bennett was busy writing inane and irrelevant "suggestions" about how to treat Ebola (since he's SO MUCH smarter than every doctor & nurse ever) after we intentionally infected him?

Then again, the recovery rate in the U.S. is depressingly high... let's drop him into a random village in Sierra Leone and see how well the local Witch Doctor reacts to his "suggestions".

A small encapsulatory structure containing a fragment of RNA. I'm not a microbiologist, so can someone tell me how these things are different from a virus?

It wasn't the chemicals, as you point out, but the penetrating object that killed her. She bled out. If she hadn't bled out, she would have likely suffered severe brain damage as skull and projectile fragments entered her cranium.

The relevance being, also as you point out, that shooting anything into the face is a bad idea when non-lethality is the intent. But any chemical that is going to be delivered in such a way has exactly that potential, as do rubber bullets (have you seen what those do? non-lethal does not mean non-damaging).

Any chemical means to convince a highly agitated crowd to cease and disperse is going to have extraordinarily strong effects, even when used correctly, with some suffering the effects more than others. Some fraction of the population is always going to be sufficiently vulnerable for lethality.

Ultimately, I think we're agreed: The very idea of a non-lethal chemical weapon is absurd.

I suspect however that if FDTI fakes did make it into your supply chain, you would much prefer any FDTI software updates to toss up a "we won't work with this device" message rather than making the device not work with any software. I don't know that I would continue to use a supplier with this type of business practice if there were any viable alternatives.

That makes no sense.

Look at it from a macro-economic perspective: The reason we're moving to automation is because it increases efficiency, allowing us to produce more goods with fewer resources. That will increase average standard of living.

There are a couple of ways it could go wrong, of course. One is that the increased efficiency and therefore increased wealth could end up concentrated in the hands a small percentage of super-wealthy people. We've actually seen a lot of this over the last few decades, but we've seen it previously during other technology-driven economic restructurings as well, and what always happens is that competition eventually drives the margins of the super successful down and in the end the wealth ends up getting spread more broadly.

That points to the other way it could go wrong: The common man only gets his share of the increased wealth by doing something to earn it. Even though increased efficiency means there's more to go around, barring some sort of large scale government-driven redistribution, you still have to work for your share of it... which means you have to be able to do something that others who have wealth consider of sufficient value to pay you. So the other way it could go wrong is that there may simply be nothing available for such people to do.

That last is also a risk we've seen bandied about in past economic shifts, especially the shift from agricultural to industrial labor. What has happened in the past is that we've created new kinds of jobs doing previously unheard-of or even previously-frivolous things. I don't see any reason that this time should be different. I expect the transition to be painful -- and the faster it happens the more painful it will be -- but I don't think there's any end to what people want. People with resources will always want things that people without resources can supply. I don't claim to have any idea what those things will be.

It's also possible that I'm wrong, and that we'll have to take a socialistic approach to distributing the fruits of automation-driven productivity increases. I don't think so, and I think we should be careful not to move that direction too quickly, because it has huge negative impacts on productivity and we're going to need all of the productivity increases we can get, but it is possible.

Because Bay Area is typical middle class neighborhood /sarcasm.

The fact is that it's going to happen regardless of where minimum wages are set, or even if there are legally-mandated minimum wages (as opposed to the market-determined real minimum wages). Anyone who thinks most unskilled jobs aren't going away is crazy. The question is at what rate this change will occur, and it seems quite clear that high minimum wages will make more automation economical sooner, pushing the rate of change.

We're edging towards a major economic restructuring driven by widespread automation. We've had automation-driven restructurings in the past, and dealt with them, and this too will be handled. But when you're talking about widespread elimination of old jobs and creation of new jobs, speed kills. Retraining, and even just adjusting to the new reality, take time, and in the meantime millions upon millions of displaced workers are a huge drain on the economy, not to mention miserable.

I think it's pretty clear that high minimum wages are a forcing function for this transition, and I don't think it's something we really want to force. Ideally, it would be better to slow it down, at least in terms of the human cost, though the most obvious mechanisms for slowing it (labor subsidies) may also dangerously distort the economy.

Ebola is not highly communicable. It's more communicable than, say, HIV, but it's still pretty low on the communicability scale, and even lower if you're asymptomatic.

Estimates of civilian casualties from the 2003 Iraq War and its aftermath vary significantly, but many are of the same order of magnitude as the deaths caused at Hiroshima and Nagasaki when the nuclear weapons were used.

I'm not sure modern warfare is as good at avoiding collateral damage as you seem to be suggesting. The causes of those civilian deaths might not be the same mechanically, but it's no less a tragedy if an innocent person dies as a side effect of some military action rather than directly by taking a bullet.