Follow Slashdot stories on Twitter


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re:Lift the gag order first... (Score 1) 363

tiering should still work, the thing being blocked is certain companies/products have a bit of a faster lane to get through the network. Paid priority will no longer exist

The one legitimately valuable business arrangement that I can see being blocked by net neutrality is where I buy the lowest possible tier of service from my ISP, because email doesn't require much bandwidth, but then buy premium delivery from Netflix, where Netflix pays my ISP to deliver data faster than my service tier.

Comment: Re:Taste of their own medicine (Score 1) 363

Already been done. The conservatives had a shitfit when their ads were blocked.

What beautiful irony. The proponents of small government and deregulation, running to a government agency complaining that a private company won't run ads calling for government regulators to block a business transaction.

Comment: Re:Lift the gag order first... (Score 5, Insightful) 363

I hate the big ISPs too. Everyone does. But the solution to them is competition. Not government regulation. Just remove the stupid laws that make it illegal for rival companies to lay cable in their territory.

Those laws don't exist in general. The primary thing preventing Time Warner from running cable to my house is the fact that Comcast already has a wire there. Comcast has already spent the millions of dollars required to wire my neighborhood, and the tens of millions required to wire my town. Whatever price Time Warner can offer, Comcast can beat, because they've already sunk costs. Time Warner can, optimistically, hope to get 50% of cable subscribers, meaning at most half the revenue that Comcast projected to pay off their capital. There is no way for a new cable company to compete effectively with one that's already laid out the major capital expenses. The only reason DSL is competitive is it doesn't require laying new copper to every home.

Likewise, there's no way multiple electric or gas companies could compete with an incumbent who had already wired/plumbed a neighborhood. When cities deregulate gas/electric service, they do so by transferring the wires to one company, and forcing that company to sell transit to all comers at regulated rates. If you want to see competition among ISPs, nationalize the coax, copper and fiber, and let the ISPs rent bandwidth to subscribers' homes and manage their access.

Comment: Re:Lots of weird crap coming out of Congress latel (Score 1) 486

by tburkhol (#49187925) Attached to: White House Threatens Veto Over EPA "Secret Science" Bills

I am prepared to engage with you on this issue rationally. But you should be warned now... I am extremely rational. Doubletalk, sophistry, and other fallacious nonsense will simply get vivisected and pinned to an examination table while I take it apart bit by bit putting each little piece in its own little formaldehyde jar with its own little label.

That's fine. You're exactly the audience this piece of legislation targets. I mean, really, who would make a rational argument against using good, reproducible, published science to support their policy.

This legislation is social engineering that depends on the distinction between theory and practice. In theory, it's great to use data to support your policies, to have good science backing your regulations, and to know the exact effects you're trying to block or induce. In practice, those data do not exist. There's 150+ chemicals at my job site, and the MSDSs for half of them report "Data not available." This legislation is intended to delay for as long as possible any regulatory action.

Example, pentachlorophenol, one of the most common preservatives for telephone poles and RR ties over the past 30 years, still lacks any human carcinogenicty studies. It looks like penta is metabolized in vivo to a much more carcinogenic compound, and I imagine that another 10 years or so will produce pretty convincing data. Meanwhile, the EPA has been allowed to regulate penta as a probable carcinogen since the early 80s. This includes banning the sale to private individuals (it used to be a popular fungicide for people to spray around their basements). Penta is one of the better studied molecules because of its popularity and explicitly toxic function, and after 60 years, one can still argue that there is no reproducible scientific data demonstrating its role in cancer.

Comment: Re:Lots of weird crap coming out of Congress latel (Score 5, Informative) 486

by tburkhol (#49187705) Attached to: White House Threatens Veto Over EPA "Secret Science" Bills

You mean the compounds so secret that there's a wikipedia page listing them all?

First, there is no reason to believe that list is exhaustive. According to the page itself, it is "a partial list of the chemical constituents in additives that are used or have been used in fracturing operations." It was only released in 2011 in response to a congressional investigation, having been held secret for 60 years. Nor does it help you know whether fracking fluid is mostly toluene or mostly liquid nitrogen (personally, my guess is that there is very little, if any, liquid nitrogen in fracking fluids, but it's on the list)

Second, from a random sampling of MSDS:

  • 2,2-Dibromomalonamide: No human toxicity studies have been carried out with this product. Not evaluated by IARC.
  • Poly(diallyldimethylammonium chloride): Not evaluated by IARC. No carcinogenicity information is available
  • Carboxymethyl hydroxypropyl guar: Carcinogenic effects: Not available; Mutagenic effects: Not available; Developmental toxicity: Not available

So, under the proposed legislation, even if you know what the chemicals are, you have to wait for someone to get interested enough in them to perform ecological, carcinogenic, and mutagenic studies with those specific chemicals and publishes the results. Until someone proves that a compound is carcinogenic, it would be regulated like it is not carcinogenic.

Perhaps you are willing to have your dinner grown next to a factory that can hold its chemical waste secret for 60 years, and then be unable to regulate that waste for another few years or decades, waiting for someone to bother to measure their health effects. Maybe you believe that no company would knowingly or accidentally release chemicals without clear confidence in their non-toxicity (even if they can't release that data to the public). Maybe you trust those companies, more than the politicized EPA, to balance their profits against potential harm to humans and environment.

Comment: Re:NOT TO BE TRUSTED (Score 1) 130

by tburkhol (#49179389) Attached to: Schneier: Either Everyone Is Cyber-secure Or No One Is

1 Companies that sell software... better have all code open sourced (not same as free) or should be labelled "NOT TO BE TRUSTED".

No way to tell whether the provided source code matches the provided firmware

Code (including scripts and updates) is then compiled locally and before first execution hash checked automatically against non-centralized database (p2p technology similar to bitcoin block chain)

1) binary code will vary depending on the specific architecture, optimizations, and libraries during compilation. 2) a hash can be falsified as easily as a binary.

3. All hardware sold with precise technical diagrams... or should be labelled "NOT TO BE TRUSTED"

At least an order of magnitude less effective than open source, and we've seen that even "important" OSS like openssl can go decades without independent code review.

4. All encryption always on client side.

Quite sensible, although I suspect that people will rapidly become frustrated when they forget their pass phrase, or lose their private key, and 5 years of family snapshots disappear. Or when grandma dies, taking access to her archive of family history with her.

5. Get rid of centralized authorities for security (looking at you SSL) Centralized servers have big fat sign that say "NOT TO BE TRUSTED". P2P.

Because you'd rather trust 1000 amateurs to secure all of their systems than one professional to secure his server?

7. Shaming lists on NGOs (applause to EFF). Any politician that votes for mass surveillance or doesn't adhere to above principles. put on NGO lists as "HUMAN RIGHTS VIOLATORS"

Yeah, ranks right up there with executing journalists and kidnapping babies. Among the most certain ways to get people to ignore you is to blow your cause completely out of proportion. If you use the same words to describe digital surveillance as other people use to describe the Khmer Rouge or Stalin, then people are going to think you're a nutcase.

Comment: Re:Jail time (Score 1) 535

Well, if anybody else in government did this, they'd get fired, lose their pension, and possibly face criminal charges.

Still waiting on charges against Sarah Palin, for the same offense. I'm guessing it will be a cold day in hell before either sees any consequences beyond partisan propaganda. In fact, I'm pretty sure this is one of those rules, like declaring any gifts over $50, that gets employees a firm warning not to do it again.

Comment: Re:Good grief... (Score 1) 677

by tburkhol (#49110461) Attached to: Bill Nye Disses "Regular" Software Writers' Science Knowledge

"Computer Systems Engineering" covers it pretty well -- it's a mix of EE and CS so you end up with a ground-up understanding from transistor to circuit to chipset to architecture to OS to software.

But it's often not "Computer systems engineering" anymore, it's CS: computer science. Dropping the E allows you to skimp on or abstract the transistor/circuit, and focus on architecture and software. CSE these days seems to mean "Computational science and engineering," which is a completely other thing, having more to do with the simulation of experiments than the design of computer/software (although you may need to write some software to simulate your expeirments).

Comment: Re:Good grief... (Score 1) 677

by tburkhol (#49110371) Attached to: Bill Nye Disses "Regular" Software Writers' Science Knowledge

Oh, because dave420 is soooo much more identifiable, right Einstein? Pot, Kettle...

AC posts a joke. Sardaukar86 disses AC for not 'hav[ing] the guts' to put his name to it. dave420 points out the hypocrisy in pseudonymously railing against anonymity.

dave420 may or may not agree with sardaukar86's point (that one should have the courage to post insults under his own name), though likely not. dave420 is making a completely different point, that distinguishing between anonymous and pseudonymous is silly. Your 'pot, kettle' reference is appropriately applied to Sardaukar86, not to dave420.

Comment: Re:But CNN Said... (Score 5, Interesting) 266

by tburkhol (#49099807) Attached to: The Robots That Will Put Coders Out of Work

We will have self driving cars long before we have robots that write code.

But we already have robots that write code. Almost no one actually writes machine code anymore, depending instead on assemblers, compilers, templates, or interpreters to do it for them. Those 'robots' have gotten progressively more complex and progressively better at figuring out what the programmer means by larger language constructions. The languages have moved closer to natural languages.

Already, it seems like the difficult part is getting the managers to properly specify the desired functionality. It's not a huge leap to imagine that one might construct a formal language for program specification that would allow you to automate translation of the spec into a code skeleton.

Comment: Re:Port 443 makes it even worse... (Score 1) 153

by tburkhol (#49086707) Attached to: Samsung Smart TVs Don't Encrypt the Voice Data They Collect

But my point was that usually 443 is a clear indicator of encryption, and hackers don't bother to try it, let alone run a packet sniffer on the port.

Maybe if you're talking about a web browser. If you're talking about a bit of custom software embedded in a TV, then ports 80 and 443 only say "traffic that will probably be allowed by firewall rules."

Comment: Re:No Trust (Score 2) 153

by tburkhol (#49086653) Attached to: Samsung Smart TVs Don't Encrypt the Voice Data They Collect

The problem is the date being sent in the fist place. A likely application is a nice speech-sample database that can then be used to identify people where other means do not work.

You have a microphone in people's living room, broadcasting every conversation they have, and the application you come up with is voice-print identification? Not listening for people reciting strings of numbers like account or social security. Not people discussing passwords, drug deals, or plots to blow up the Capitol. Not people talking about a new car, a new pregnancy, or an imminent wedding. The content of these conversations is (presumably) being sent home at least to do Siri-like speech to text, so even Samsung clearly has the processing power to generate transcripts for all of those conversations, easily searchable, tied to a specific consumer, and salable to marketing or security services.

Comment: Re:Web site gets hacked... (Score 3, Informative) 125

by tburkhol (#49086095) Attached to: Jamie Oliver's Website Serving Malware

Why, always, 11, ... ?

In the US, the traditional time for networks to show their nightly news is 11pm, after the 'prime time' entertainment and kids have gone to bed. Any unsold prime-time commercial slots are filled with teasers for these news programs, generally of the form "Shocking ways that Foo can kill you! Details at 11," or "Weird tricks to save you money! News at 11."

Comment: Re:C4 (Score 2) 389

by tburkhol (#49075319) Attached to: What To Do After Robots Take Your Job

If there are fewer jobs, how can it not mean unemployment?

Not fewer jobs, different jobs. When the cotton gin put all the seed-pullers out of work, it created demand for cotton pickers. When steamships put the wind jammers out of business, it created demand for longshoremen. You (and I) may not be clever enough to figure out what to do when they automate elevator operators or McWendyKing burger flippers, but there will be something, even for unskilled workers. Think about how many baristas there were in 1980. Or how many microbreweries in 1990. Kids today are going to work in fields that didn't exist 10 years ago.

Comment: Re:For targeted advertising? (Score 1) 227

How are they planning on delivering that? Through injecting ads in your traffic, email spam or letterbox spam?

From AT&T's faq:

For example, if you search for a car online, you may receive an email notifying you of a local dealership's sale.

So expect this to mean out-of-channel advertising. In fact, it sounds like they mean primarily to deliver targeted email, rather than to inject html. I can't really imagine that an email address is worth $30/month to advertisers, so this really does sound like a punitive charge on people who are concerned about privacy.

Mommy, what happens to your files when you die?