Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?
What's the story with these ads on Slashdot? Check out our new blog post to find out. ×

Comment Re:Too many of them aren't worth following (Score 2) 132

Too many seem to have the following structure:

90% of the time is dedicated to an episode specific narrative following a formula. Whether it's the detectives getting a case, the scientific guy chasing a new phenomenon, etc. For the most part the, the events in this portion are episode specific although usually there's some new morsel that exposes information the grand conspiracy and larger story arc when that episode's events are resolved.

10% of the time is dedicated to following/expositing the serial aspect of the story, usually some kind of conspiracy or larger story. Very little information is exposed, mostly just enough to let you remember there's this bigger (and often much more interesting) narrative arc taking place.

Mostly this just feels as if the series has been turned on its head. It should be about the 10% part that is the actual "meat" of the story. If (and only if) the dumb series runs enough seasons, the larger story arc might get resolved in some semi-satisfying way. Mostly it seems like the writer had a pretty cool idea but didn't know what to do with it, and fell back on the "case of the week" to fill it in because the bigger idea really didn't have much behind it.

In some cases, this can be tolerable but most of the time you just feel strung along, like there's this really cool story that's going to get broken wide open...and then nothing, or something entirely lame like Lost happens.

In contrast, really good series (like the Wire) manage to make the entire series about the story arc and the individual episodes expand and bring it out. Part of the Wire's specific genius was that it did this well and also had a seasonal anthology feel to it as the action shifted from the corner, to the port, to the dealers again without losing the larger momentum but giving us different characters and settings, too.

When I start a new series if I feel like I'm being strung along by episode 4 or 5, chances are I won't ever get resolution and I just drop it.

Comment Re:Cop video storage is a moral hazard for Taser (Score 1) 99

The thing is, I don't think it would be a stated business strategy.

The nature of most moral hazards isn't that they're obvious conspiracies to do the wrong thing, but a set of biases and bad incentives that lend themselves to creating a situation where bad choices get made.

As an example, drug addiction is a moral hazard for doctors. Doctors know that drugs can be habit forming. We expect doctors to be experts in administering them, to have reasonable ease of access to them for treating patients as best they can. The doctor believes his own expertise will prevent him from getting addicted to them. But expertise plus overconfidence in their own knowledge plus access results in a ton of doctors getting hooked on drugs.

Taser for the most part sells stuff to cops. Taser would like to keep cops happy and keep buying cop stuff. Taser "knows its market" and understands what they want. At some point the desire to make money selling stuff to cops and knowing what cops want lends itself to creating holes in accountability, not because some executive said "they're good guys and good customers, they shouldn't get dragged down because some douchebag criminal got a good attorney" but because they want to please their market for reasons that are independently all completely normal and reasonable.

With automated systems, it's much harder to argue that the problem wasn't deliberate.

"When asked why the body camera video of the police beating didn't exist, despite the system supposedly being automated to upload them to remote secure storage, officials noted that 'network limitations' caused by 'budget constraints' prevented the video from being immediately uploaded as originally designed. Police data networks were overwhelmed when the system was first rolled out and the vendor, Taser, Inc, added an on-site caching feature that uploaded the videos in a slower and more controlled fashion to prevent network overload. A problem with the caching server at Police HQ caused 'only a handful' of videos to be lost and Taser officials said this risk will be fixed in a new version available sometime next year."

Desire to sell your product + pleasing your customer = exploitable hole, even though nobody actually *conspired* to do this even though the design goal was the opposite. Had a vendor been selected whose first concern was guaranteeing data integrity, not necessarily accommodating the end user's specific desires, the hazard could be avoided. But this only happens if the vendor's allegiance can be to someone other than the cops, like some kind of oversight board whose principal interest is in data integrity.

This way the vendor's goals are aligned with the purchaser's goals and the hazard is avoided.

Comment Re:Cop video storage is a moral hazard for Taser (Score 1) 99

But that's thing with a moral hazard -- just look at banking and securities. If you jack around the majority of your customers, it will become public and cause a shitstorm, but it doesn't make the moral hazard go away nor has it prevented all manner of moral hazards in banking from being exploited.

And not every -- or any -- potentially "lost" video is going to be tied to some high profile incident where some innocent black woman in a wheel chair took a dozen rounds of 00-buck to her face. The most likely ones will be the low profile ones nobody cares about, where some obvious drunk got manhandled after bar closing and a dozen citizen eyewitness statements back the police version of events completely.

And it's also not likely that Taser would just delete videos themselves -- that's too obvious. Rather than running a system that's totally secure from police tampering, the inclination will be to provide a "friendly" system that offers soft points where the police can prevent videos from getting uploaded at all under the guise of technical glitch or something.

Comment Re:Should they only be in the layer-2 business? (Score 1) 162

The thing is, you're *already* "having a company come in" as a carrier ISP to supply uplink for the municipal fiber. And hopefully/presumably more than one carrier is being used for redundancy. I would also guess that these carrier facility equipment rollouts aren't just some 2U Cisco router with a fiber port and an ethernet jack. Chances are there's enough uplink brought in by all the carriers that they could easily resell uplink to other 'ISPs" in the muni NOC.

I don't know what equipment the muni is using for terminating the fiber connections, but what do you want to guess it's MPLS or something which would easily have the facility to map and aggregate endpoints to some other endpoint within their NOC which would then act like a local ISP. They're probably already doing this so the water department/school/etc can have a private LAN that spans sites.

AFAIK there are still a fair number of regional/smaller ISPs serving niche markets who might be interested in opening a branch for that many fiber connected customers or who could be tasked with acting as the "caretaker" of the L2 network (getting the muni out of that headache) and for whom adding layer 3 service would be no worse than break even if they are already paid for managing the L2 network. In Minneapolis they did something similar with city-wide wifi -- the network was built and managed by an ISP. I doubt the paying wifi customers meet their costs, but the added costs of retail wifi are really low when you're being paid to manage the physical network.

Splitting off layer-2 from layer-3 would also make a ton of sense from a business incubation perspective, because if you were slightly forward thinking the the muni fiber NOC wouldn't just be a spare room in the basement of city hall, but a datacenter-like space which would have room for colo for whoever wanted to be an ISP and for local businesses looking for an offsite location.

Now you've got a big-city style datacenter facility with a large geographic fiber plant connected. It might attract a lot of other interested business looking for a well-connected smaller town to open a branch office or take advantage of lower cost of doing business. It's about the equivalent of widening the county road to the Interstate and paving your gravel streets.

Comment Re:Speed isn't Everything (Score 1) 162

I think the model like DSL service should be the one that municipal fiber follows -- the municipal fiber just provides the layer 2 connectivity and you choose which ISP you want.

If somebody wants to start a geek-centric service with static IPs and where technical support is limited to setting reverse DNS, great, they can buy a rack or whatever at the municipal fiber hosting center and sell that service to whoever's interested.

If Comcast or AOL or whoever wants to offer their mega-consumer focused service with dynamic IPs, webmail, coupon offers, ad-injection, great, they can lease a rack, too and sell that.

Plenty of cheapskates and technophobes will pick the consumer service for all the add-ons and technical support and the geeks willing to spend the same or just slightly more for static IP service with none of the bullshit can pick that.

There was a time where a company I knew set themselves up as an ISP choice for DSL. Employees could get DSL from the phone company, choose their employer as their ISP and they had basically a hardwired VPN to work (that solution has some issues in terms of personal-vs-work access, but IIRC from the network guy at that company I talked to they had an entirely separate Internet provider they routed that traffic over). I think whatever setup and operational cost was greatly mitigated by reduced costs related to remote access and the legion of VIPs who wanted their personal ISP bill reimbursed because that "expense" got taken care off at wholesale.

The analogy that makes the most sense is the roads. The city builds 'em, fixes 'em and sets some pretty basic usage rules, but you buy your transportation and delivery services from other companies. If I want a pizza, I pick whoever provides the pizza I want and they just use the road to get it to me.

Comment Re:Cop video storage is a moral hazard for Taser (Score 2) 99

I think it is worse -- when the police control it, the moral hazard and control issues are pretty obvious.

When a third party controls it, it's more opaque. The police have plausible deniability to say "But we use a third party vendor, we didn't delete that video." The fact that Taser has a financial relationship with police departments is much less clear (to the general public at least) and it's a lot less clear that Taser has a neutral motivation with regard to these videos.

To me, the solution should more likely be that some police oversight entity selects/approves/controls the video storage contract and probably should be contracting with a vendor who doesn't have a specific dependency on the police as a target market. That may be more difficult if regulations regarding these videos lend themselves to market specialization and you end up needing vendors who specialize in those markets.

You'll end up with a similar moral hazard, but at least you'll have reduced the amount of financial influence the police have over the vendors.

Comment Should they only be in the layer-2 business? (Score 3, Insightful) 162

While I mostly think this is great, I wonder if they should be in the "business" of supplying actual layer-3 connectivity or whether they should just be maintaining the fiber plant and selling access to it to other companies willing to provide actual IP connectivity?

Maybe a purely internal municipal ISP makes sense for supplying IP connectivity to municipal offices, schools or other parts of the government.

The part that makes me kind of leery is the fact that the government is the ISP and this creates a certain conflict. Does the fact that the municipality runs it mean that the police have greater access to monitor the network or some increased motivation to use municipal control to go after "evildoers"?

It's not hard to see how this could also morph into the kind of local political control that those in power use to stay in power.

Comment Re:... less energy than a greenhouse (Score 1) 119

I'm not sure when greenhouses were meant to be net producers of energy.

I think their original purposes might have been:

1) growing things in climates that were otherwise too cold for them (where your energy "production") probably comes from. aka "the greenhouse effect" which I think in practice almost always has some kind of either supplementary heat (if its too cold outside for the sun to provide enough heat) or supplementary ventilation (to keep it from being too hot).

2) protect more sensitive plants from natural predators, weeds, etc -- like floral greenhouses

Comment Cop video storage is a moral hazard for Taser (Score 2) 99

Body cameras haven't been around long enough to really know whether they will be predominantly exculpatory for the police or provide evidence of misconduct.

But doesn't relying on a vendor who has a financial interest in continued sales to police organizations in charge of storing possible evidence of police misconduct create a significant moral hazard for Taser?

If they come to be seen as an organization "too cooperative" with enforcement of rules against police misconduct, doesn't this imperil their image with the police and potential sales of equipment to the police? It would seem this would provide them with a subtle pro-police bias which could undermine the entire point of video cameras from the public's perspective.

Comment Re:Three main types of bad jobs. (Score 5, Interesting) 468

Basically tech jobs are closer to blue collar than white collar

A peer and I once made the same comparison. We called ourselves digital maintenance men, because by and large that's what it is.

I've never worked for a company that had a significant manufacturing component, but I kind of wonder how the blue/white collar split works there for the people who setup, maintain and manage seriously complicated factory systems. I think they might have been called millwrights at one time.

Are they treated like blue collar people (probably, if the job involves any serious mechanical tools), or because of the sophistication of the equipment (all computer driven and complicated) are they treated like dirt, like other blue collar jobs, with all the usual management/labor hostility, clock punching, etc.

And why do "office" jobs seem to escape a lot of that labor/management hostility? Even the lowly marketing associate seems to get treated better than the most skilled blue collar worker. I've known some electricians who were really intelligent and used to sort out cabling issues in my data center better than I could, even though he didn't know how to configure the equipment. He'd make suggestions via some kind of intuition that never dawned on me.

Comment Policy recommendations aren't predictive? (Score 1) 153

That would make the techniques less interesting to many economists, who are usually more concerned about giving policy recommendations than in making forecasts.

Decision Maker: The opposition and the polls are beating me up over the jobs numbers. Give me some policy advice, economist.

Economist: I think you should implement this policy.

Decision Maker: Will it improve the jobs numbers?

Economist: I have no idea what the outcome of the policy will be, I just made some stuff up.

Isn't the entire point of policy recommendations to achieve some kind of desired goal? Even if the policy recommendation is based on pure ideology, usually the alignment with the ideology is based on some notion that the ideology produces the best outcomes. There may not be data to prove any of it, but it's not like the policy was selected because it has a cool logo or you like Hayek's suits or something.

Comment Re:Programmed behaviour is programmed behaviour. (Score 3, Informative) 437

Every time I've heard an expert (usually a college professor with a background in computer science, robotics, or automation) discuss existing self-driving cars (the Google car is almost always mentioned as an example), the experts always describe self-driving cars as something more highly programmed and rule-bound than actually autonomous.

They rely less on machine vision and more on extremely detailed and high-resolution saved maps versus driving the road they see in front of them. Sensors are used to determine hazards, but more for avoidance than some kind of self-guided navigation decisions.

Comment Re:exhibit A: OK Cupid's famous essay (Score 1) 310

I don't doubt that my made up example would be difficult to actually pull off in practice, although who knows. There may be enough women who are turned off by the meat market aspect of other dating sites that a service with a zero tolerance for weird behavior might find it appealing. And both sexes may find the idea that "the system" automatically weeds out inactive or unsuccessful daters appealing, knowing that they will be much less likely to waste time on "losers".

I think there are some "higher end" in-person dating services that cater to higher-income professionals looking for long-term relationships that have made something similar work. They cost a bundle and involve a lot of human interaction and these kinds may be doing the sort of active filtering that eliminates dead wood.

I also wonder if the pricing model of dating sites isn't skewed against more and better matches. If women (or even men) gain access at reduced costs, they may value it less and invest less in it personally. If men pay a higher cost for access, they may over-engage because they value it more than women and appear desperate when really they're just trying to get their money's worth.

I suspect that some minimal level of cost to participate is probably necessary -- without "skin in the game" it's too easy for people to willfully not participate and create imbalances in interest. You probably could also benefit from a "participation economy" -- credits against your bill for responding to messages, credits for going on any kind of a date, etc, with credits valuable and easy enough to obtain that people who are actively engaged in the site might actually end up having zero monthly cost. Encourage participation, discourage non-participation.

I'd also wager that some kind of moderation system would make sense -- I hear a lot of complaints from women who have used online dating that the creep factor is really high -- men who make lewd propositions to people whose profiles are listed as "seeking a relationship", etc. Perhaps users bothered by a message could submit it for moderation, and moderated messages would be anonymously displayed to other users who could vote them up or down and receive credits for it. I would probably limit moderation of messages to people seeking similar relationships, since those looking for longer term relationships would have a lower tolerance for messages suggestive of casual encounters. This would avoid an obvious values conflict between the two groups.

Users who have messages moderated as inappropriate would lose credits. Users who submit messages for moderation where their complaints are unsustained would also lose credits. This would enforce a kind of community standard for acceptable behavior as well as discourage people from being offended too casually, and I think the latter is probably equally important. I think there are people who are single not because they don't want to be in relationships but have really skewed, intolerant or unrealistic standards and are basically single because of it.

To downgrade the human mind is bad theology. - C. K. Chesterton