Wikipedia:Reference desk/Archives/Miscellaneous/2016 November 30

= November 30 =

Jim Walters Mining Disaster in Brookwood, Alabama
This disaster killed one miner and twelve rescuers in two explosions at the Jim Walters Coal Mine on September 23, 2001. It was of course overshadowed by the events of 9-11, and the loss of life was small in comparison, but the injustice it represented was a rather big deal as the mine was fined only 5000 dollars for its long history of violations. However, it is only referenced on two Wiki pages and there is no article about it.

From the Walter Energy page:

2001 mine disaster[edit source] At approximately 5:15 p.m. on September 23, 2001, at the Jim Walter Resources No. 5 coal mine in Brookwood, 40 miles southwest of Birmingham, a cave-in caused a release of methane gas that sparked two major explosions, killing thirteen miners.[2]

From the Sago Mine Disaster page:

The Sago Mine disaster was a coal mine explosion on January 2, 2006, at the Sago Mine in Sago, West Virginia, United States, near the Upshur County seat of Buckhannon. The blast and collapse trapped 13 miners for nearly two days; only one survived.[1] It was the worst mining disaster in the United States since the Jim Walter Resources Mine disaster in Alabama on September 23, 2001,[2][3] and the worst disaster in West Virginia since the 1968 Farmington Mine disaster. It was exceeded four years later by the Upper Big Branch Mine disaster, also a coal mine explosion in West Virginia, which killed 29 miners in April 2010.

The link on the Sago page takes the reader back to that bare mention on the Walter Energy page.

My question is: would it be more appropriate to expand the section on the Walter Energy page or to write an article? Aside from the 1 working reference on the Sago article, I have some newspaper articles as online references but I cannot find the disaster in any printed books that I've been able to get hold of. Can you give guidance on what action I should take?

Thank you.

Viccivarner (talk) 01:59, 30 November 2016 (UTC)


 * I would advise gathering all of the reliable sources you do have, and writing a draft article, in either draft-space or your own userspace. If it winds up being so extensive that it would dominate the article on Walter Energy, then it should be its own page. Is there anywhere you haven't looked for sources that we could help, or any sources you know of, but couldn't get a hold of? Someguy1221 (talk) 02:03, 30 November 2016 (UTC)


 * DuckDuckGo has search results for Jim Walters Mining Disaster in Brookwood, Alabama.
 * —Wavelength (talk) 03:17, 30 November 2016 (UTC)

How do you stop an AI from hacking itself?
Suppose you are building a robot to achieve some goal. It must have some way of knowing that the goal has been achieved, through some measuring device. If it's an intelligent robot with advanced manipulators, it might learn how to hack the electronics of its own measuring device to make it think that it has achieved the goal, instead of actually achieving it. How would you program the robot to prevent it from doing this? PeterPresent (talk) 03:42, 30 November 2016 (UTC)


 * By being more clever than the robot. Philosophically, this is the same as any problem asking how to prevent any entity from following the letter of a law or rule without following its intent. Any time you see someone scheming to find a loophole in the law, or in any work of fiction where a genie grants a wish a bit too literally, this is the problem you're dealing with. We even see this in genetics - you engineer a fly to fluoresce when a gene is on, then look for a mutation that shuts down that gene. But since your measurement is fluorescence, the easiest way to appear to shut down the gene is to simply delete the gene for fluorescence! So more formally, you task a robot with achieving outcome X repeatedly. You permit the robot to alter its behavior or even electronics/mechanics to achieve X more efficiently. But how did you teach the robot to determine that X has been achieved? If it's something like flipping a transistor, the easiest way to achieve that goal is for the robot to simply rewire itself so that said transistor flips on its own. You would prevent this by writing clever rules regarding what the robot can or cannot alter, and making more complicated rules to determine that a task has been achieved, basically an audit of the robot's activities. Basically, the same sorts of things you do to stop a human from cheating, just with wires and code instead of laws and regulations. Someguy1221 (talk) 03:54, 30 November 2016 (UTC)


 * The OP may be interested in reading about the technological singularity. -- Jayron 32 03:58, 30 November 2016 (UTC)
 * Or possibly Existential risk from artificial general intelligence. -Arch dude (talk) 17:47, 30 November 2016 (UTC)


 * The OP's question is a central concern of the book Superintelligence by Nick Bostrom. The book is largely wild speculation, but he raises interesting questions and it's a quick, non-technical read. μηδείς (talk) 18:52, 30 November 2016 (UTC)
 * The OP may also want to read up on Futures studies and prominent Futurists, which are people who have convinced the world they should be paid for telling us when we'll have flying cars. -- Jayron 32 02:13, 1 December 2016 (UTC)
 * I've seen plenty of "flying" cars on our highways. ←Baseball Bugs What's up, Doc? carrots→ 08:07, 1 December 2016 (UTC)


 * We're the product of billions of years of evolution and yet you still get people hacking themselves by taking drugs or gambling or not believing things because they don't want them to be true. If we are so liable to acting less than optimally, in fact plain stupidly, I don't see that there can be any straightforward solution. And if you do find one it may be quite useful generally if adapted for people. Dmcq (talk) 22:42, 4 December 2016 (UTC)

Moving phone and broadband from different suppliers
I am in the UK, and have phone and broadband from two different suppliers, neither of them BT. (They both go over BT physical lines of course; but Ihave no relationship with BT).

Next week I'm moving (just a few yards, so I am keeping the phone number). There's currently no phone line connected to the house I'm moving to, so my phone supplier will be getting BT Openreach to install whatever is necessary. I rang my broadband company to ask about moving, and they first said that if the line is not yet in, they cannot take the order to move it. Then they said that if I could get the SIM (possibly called the LORN) for the line, they could accept the order in advance. I tried to get these from the phone company the same day, and the agent went away and talked to the techies, and said they didn't know, as the order hadn't even gone in yet. So I waited a week, and tried again. This time, after talking to the techies, the agent said that they will not be able to provide me with that ever.

So as it stands, when the phone moves over on Tuesday my Broadband will vanish, and I can only order a new broadband connection then, which takes 7 to 10 days.

Does anybody have any explanations that I can give to the phone company, or suggestions about how I can get through this? (I've deliberately not named the companies, because it seems doubtful that that is relevant). Obviously I've thought about workrounds like using mobile (reception's very poor here) or using a neighbour's connection (which is probably what I'll do). --ColinFine (talk) 13:13, 30 November 2016 (UTC)


 * Try reading this - http://www.webologist.co.uk/blog/bt-woes-and-the-account-is-not-activated-yet - and take note of the comment that the Sim transfer has to be requested at the same time as the order is placed for the line - you cannot order the line installed, then get a LORN later. You may have to cancel the order, and start again making sure that the Sim transfer is booked as part of the BTOutreach order. Wymspen (talk) 17:12, 30 November 2016 (UTC)


 * Thanks, . At least I know now. That makes things clearer. I have no relationship with BT, and no confidence that if my phone supplier put in a new order they would even know to ask for is, and I'd lose a week. I'm just going to have to accept the delay I think. --ColinFine (talk) 23:16, 1 December 2016 (UTC)