Forgot your password?
typodupeerror

Comment: Nice work, but this is kind of like cheating... (Score 1) 54

by going_the_2Rpi_way (#27023443) Attached to: Exoplanet Found In Old Hubble Image
... since knowing there's something there provides additional information that can be used to calibrate the extraction routine.

but not visible until new knowledge could see the picture in a fresh light.

This says it all. In fact, you could create a much simpler extraction technique consisting of a black box around the known item that meets this same standard. Can the new extraction technique do more than this? That, apparently, remains to be seen.

Comment: Cumbersome... (Score 1) 114

by going_the_2Rpi_way (#26938771) Attached to: Freeing and Forgetting Data With Science Commons
This is the just as likely to add burden as to remove it.

I can't count the number of times I've seen attempts to 'standardize' data, or even just notation, in a given field. It all works very well for data to that point, but then the field expands or changes, or new assumptions become important, and the whole thing becomes either unwieldy or obsolete. This is one reason why every different field, it seems, has their own standards in their literature.

Speaking of the literature, most of these proposals are quickly followed by a 'let's just ask authors to conform to this now' approach to adopting these things. Papers get rewritten (or rejected), key points get lost, and the community gets weaker, all so that some standard with a half life of 12 months can be implemented.

This might be different. I applaud people trying to solve hard problems, and this is certainly one. I do think that more of the burden should be on demonstrating that the standradization is applicable for 12 months or more AFTER final development in a given field, never mind several.

Generally, though we shouldn't fear context. We should embrace it.

Comment: Re:Backup Testing? (Score 1) 450

by going_the_2Rpi_way (#26931965) Attached to: Ma.gnolia User Data Is Gone For Good
Ouch... Isn't part of a backup strategy to sometimes attempt a recovery from a backup, on a test system?

Yes. He addresses this and acknowledges he did not either deliberately fail his system or conduct extensive tests to ensure his backup scheme was adequate.

He acknowledges this was one of many 'lessons learned' (aka huge mistakes made).

Comment: Re:Ok then... (Score 1) 244

by going_the_2Rpi_way (#26897615) Attached to: Researchers Hack Biometric Faces
Here's how you do it with a face: instead of using your own face, you a photo of Brad Pitt on your Iphone or related device. When they brute force that, you switch to a picture of Jennifer Anniston. You can change your 'biometric-based' password just as easily as they can brute force it. Just don't limit yourself to your own biometrics.

Comment: Re:Ok then... (Score 1) 244

by going_the_2Rpi_way (#26897133) Attached to: Researchers Hack Biometric Faces
Once they have your password, you choose another one and that's it. I'd like to see you do that with your face.

I take your point, but I don't understand the either/or philosophy of security. Besides, in most cases that matter, once they have your 'password', they have you. Period.

To me, security is all about layering anyways. Adding a biometric layer that works well for the user (i.e. effortless) and typically involves a brute force attack to defeat? Why not?

Comment: Re:Ok then... (Score 1) 244

by going_the_2Rpi_way (#26897059) Attached to: Researchers Hack Biometric Faces
It was a bit of a joke. But I don't think your scenario would work anyways given their need to adjust lighting conditions as they mentioned.

More to the point, you could use something like an Iphone with a DB of randomly generated photos until it cracked. This is what the researchers here did. This is the real vulnerability. But it's brute force attack, and on any proper 'secured' system it would have to be one of several.

Comment: Ok then... (Score 4, Interesting) 244

by going_the_2Rpi_way (#26896881) Attached to: Researchers Hack Biometric Faces
He says the laptop makers should remove the facial biometrics feature from their products because the vulnerability of this technology can't be fixed.

If that's the standard, all security features should be removed. Everything is somewhat vulnerable, and a determined intruder with infinite resource will almost always find a way in. The object is to make this unreasonably hard for most applications.

If you get your laptop lifted at the coffee shop, they better lift your wallet too I guess.

Comment: Re:Would this have widespread use? (Score 1) 462

by going_the_2Rpi_way (#26892457) Attached to: Casinos Warn iPhone Card-Counting App is Illegal
My first thought upon reading this article is that Vegas wants to get lots of people to think they can count cards and come on down to empty their pockets. Even (maybe especially?) in relatively desperate times.

Most won't dare use the IPhone in a Casino anyways, though they might kid themselves long enough to book a trip and hit the floor.

Some might decide to do it themselves once they figure out the mechanics of card counting. The casinos love people like this.

It's basically a big commercial for Vegas, Blackjack, and deluding yourself that you can maybe win too. Just like everything else these days, it seems.

Comment: Re:Government should not compete (Score 1) 269

by going_the_2Rpi_way (#26892169) Attached to: CRTC Mulls Canadian Content On the Internet
I don't really agree.

Canada's system of government is (and probably needs to be) Parliamentary system, Westminster-style. The nonsense with the Newfoundland delegates being allowed to vote for their constituent's interest, and the whole debacle about the coalition government a few months back illustrate how far we've gotten from this -- most people are apparently not even aware of what constitutes a government in Canada.

CRTC's biggest challenge is in mirroring this representation domestically, not branding at all. Anyone who watches Hockey Night in Canada knows that even with the amount of regional protection rules we have enforced now, we still get what seems like 80-90% coverage of a terrible Toronto-based team preferentially over much better teams in playoff races. Contrary to what LEaf fans think, this is not because it's 'what the people want to see'.

With all that said, I don't think any Canadian content as envision by the parent is currently needed or really, technologically viable.

I DO think that networks, websites, and shows that want to have regional IP blocking and a 'Canadian viewers go here' set up (as many now do) may reasonably be required to have a mandatory amount of Canadian content available though. Since they are the ones enforcing the distinction, it would be both reasonable and feasible for the CRTC to concern itself with 'airwaves' targeted purely at Canadians.

Make headway at work. Continue to let things deteriorate at home.

Working...