Forgot your password?
typodupeerror

Comment: Re:Opus flawed test results (Score 1) 62

by atamido (#45645899) Attached to: Opus 1.1 Released

>Pretty pictures might work on most people, but a page full of "hear say" does nothing to help Opus.
Hi, I'm one of the authors of the codec. I wrote detailed pages explaining how aspects of the codec work, which you could not be bothered to read before lecturing us on what's supposedly wrong with Opus. Which one of us is indulging in 'hearsay' again?

Oh, snap! That was awesome.

Comment: Re:Already has good adoption (Score 1) 62

by atamido (#45645851) Attached to: Opus 1.1 Released

Not quite. It's true that for the majority of western music it performs just as well as AAC and Vorbis, however there are certain classes of audio that it does poorly with, in particular polyphonic music. This is an inherent limitation (steming from the pre/post comb filter), that cannot be overcome in future encoders.

This is not actually "true" now. I remember reading a while back that this was one of the major goals of the 1.1 release, and it looks like they largely met that goal.

Look for the section labeled "Tonality Estimation".
http://xiph.org/~xiphmont/demo/opus/demo3.shtml

The short of it is that they have additional code to detect when there are many tones, and when "we consider the frame to be tonal and increase its bitrate." The samples they have of the page seem to show a 25-50% increase in bitrate when it does detect. So, you could easily use it to transparently encode your music library with the caveat that some samples will encode at a significantly higher bitrate. Really though, unless your library consists of a lot of harpsichord music, you're unlikely to see a real impact from this.

Comment: Re:Great... (Score 1) 520

by atamido (#45318043) Attached to: Gunman Opens Fire At LAX

That's my point Airports are full of armed guards and police... At least UK ones are. Men and women, with guns, loaded, maintained, trained in their use and ready to act. Didn't seem to help here.

I'm going to add my experience to that of TWX's. Flying in and out of London in 2008-2009 was a little surreal relative to airports in the US. US airports are staffed by glorified security workers. In a literal sense. I had a roommate when the TSA was formed, who worked airport security. In his words, the only things that changed were his uniform, and getting a pay raise. I liked the guy, he was nice and honorable, but I'm not entirely sure he even graduated from high school. And he was, IMHO, one of the more qualified people I'm seen working for TSA. As far as weaponry, I've never even seen a rifle in a US airport. I'm sure there are some somewhere, but they aren't out in the open. A few TSA agents have handguns, but those are few and far between, and I'm not sure I'd trust them to actually shoot something with them. They occupy some state between rent-a-cop and actual-cop.

Compare that to the London security that was clearly military personnel, carrying fully automatic rifles, and wearing camouflage uniforms. They were scattered all over the place, and some were in fortified positions. These were not slack-jawed security guards, these were professionals putting off an air of menace. I recall at least most of them standing, being physically fit, paying attention to their surroundings, and properly holding their weapon. They were the exact opposite of what one sees in the US.

I imagine the type and effectiveness of a response to armed attack is completely different in the London airport than in the US, with the London response being far more effective. That US airport response was poor is not a surprise, but it has nothing to do with gun control.

Oddly, in London, law enforcement either had no firearm, or a fully automatic. There was no middle ground. No street police were armed. Compared to the US, people expect all public law enforcement to be armed with only a pistol. I found the wild extremes in London to be odd. Although I suspect the extra firepower is necessary to cover more extreme situations that regular law enforcement is totally unable to handle due to being unarmed (at least with firearms).

Comment: Re:So.... (Score 1) 169

by atamido (#44859871) Attached to: The Post-Lecture Classroom

I used to teach, and the method I used was "Here is what I told you I talked about from last weeks lecture, and here is today's lecture. And this is what I am going to talk about for the next lecture", ". Not always in that order, but it works. Three repetitions and the material is understood and learned. Home flipping is applying a similar principal.

As a student, I also find this type of teaching most effective. It can be extremely difficult when a topic is discussed, isn't really understood, and then nobody has any time to discuss it again.

Comment: Re:Huh? (Score 1) 241

by atamido (#44304575) Attached to: Ask Slashdot: Enterprise Level Network Devices For Home Use?

Source? It's true for disk cache because you're limited to random seek and read speeds, but shouldn't be for SSDs.

Just go and ask. I would have to post long and complicated chat logs which present criteria much more complicated than I mentioned.

irc://chat.freenode.net/squid

You should also read the FAQ to get an idea of the RAM limitation associated with a large cache.

http://wiki.squid-cache.org/SquidFaq/SquidMemory

Comment: Re:Huh? (Score 1) 241

by atamido (#44291309) Attached to: Ask Slashdot: Enterprise Level Network Devices For Home Use?

As per the folks over at #squid, this is only true if the primary goal is to reduce data usage on an internet connection, and it would be part of a larger tiered storage methodology.

In most use cases, organizations seem to be trying to reduce latency on small files. For example, caching all 200KB files to RAM would significantly improve browsing "snappiness" for a group of users on Facebook where there was a significant overlap in friends lists. Similarly for tiles on Google Maps, or other common sites. Storing those same files onto a typical USB thumb drive would likely decrease the performance. You can store an incredible amount of "small" files in 1GB.

Larger files would be cached to disk, but as mentioned previously this is usually just done to reduce data usage. For example, a small ISP on a limited data pipe might perform aggressive caching of larger files. A large SSD can store a lot of YouTube videos, and as it's not uncommon for certain videos to surge in popularity, you can actually get some cache hits off of these, so it'd make sense to make the investments in hardware.

But as pointed out previously, none of that is going to be particularly useful on an embedded device with 16-32MB of RAM. I haven't used the DIR-632, but I have used several other routers with WRT based firmwares, and they've been mostly good (I have seen instability that were likely hardware related). For most things I think they're great, but they are going to be CPU/RAM constrained for many larger tasks. When you try to QoS 10,000 Bittorrent connections at 50Mbps along with transparent caching of normal browsing traffic for several users, all on an underclocked ARM with 32MB of RAM, you're going to be disappointed.

Comment: Re:Huh? (Score 1) 241

by atamido (#44280525) Attached to: Ask Slashdot: Enterprise Level Network Devices For Home Use?

Bad advice... You're wasting a LOT of power, and you're spending a lot more money, for a device with lesser capabilities.

It's possible that GP has needs that aren't met by a cheap router. Maybe he does filtering, or operates a proxy, or transparent caching? Maybe he does fancy traffic analysis? Maybe he does complicated multi-level QoS? Maybe he has a landing page for open wireless that is separate from his encrypted wireless/LAN? Maybe he wants to be able to handle a large number of connection states that crashes lesser devices with not enough RAM? A least that's why I use a small netbook instead of one of the many cheap routers I've purchased over the years...

Comment: Re:Does anyone know (Score 1) 1737

by atamido (#44280467) Attached to: George Zimmerman Acquitted In Death of Trayvon Martin

I put a similar gun in a similar holster and had my six-foot 190-pound son straddle me, and I was easily able to draw. In fact, the only way my son could stop me from drawing was to put his knee on my arm and pin it hard. Even then I was able to work my arm free, though it was hard when he put his full weight on it -- and that was on carpet. On grass it would have been easier.

Out of curiosity, did you test fighting over the gun in that position? And for a point of reference, how large are you?

Comment: Re:Also (Score 1) 1737

by atamido (#44280395) Attached to: George Zimmerman Acquitted In Death of Trayvon Martin

Basically as a civilian in a self defense situation don't draw your gun unless to shoot and don't shoot except to kill. If the situation isn't serious enough to warrant that, then a gun isn't the answer.

It's worth noting that it's not this cut and dry in reality. If you don't have a gun drawn already, then someone could easily cover 20ft in the time it takes to draw a gun, aim, and fire. Even with the gun drawn and unaimed, an attacker could cover 10ft before the gun could be aimed and fired. Pulling your firearm early may result in legal hot water, but not pulling out your gun because you're worried about a lawsuit and getting killed as a result is not the better option.

Example situation: It's dark, you're an out of shape (or small) girl, and walking a street alone. A big sketchy character comes out from the shadows and appears to be following you about 10ft away. They might just be going your way, or they might be getting ready to knife your for your wallet/phone. If you run, the other person could easily catch you in a couple of seconds and you'd be poorly prepared as you're being slammed face first into the ground. If you just show your gun, the other person could easily cover the distance and overpower you before you could do anything. If you pull out your gun and the guy just happens to be going your way, you could have charges pressed. (It also makes you kind of a jerk.) What do you do?

This is one of those situations that non-lethal force (such as tasers) do a good job of filling. The law is much more lenient about drawing a taser early, and it also has significant stopping power. But at the same time it will only temporarily stop someone, and it's worthless against multiple attackers. Pepper spray works against multiple attackers, but has varying degrees of effectiveness against different people (I've seen it be totally ineffective).

Comment: Re:Man the FL state attornies just want to fuck up (Score 1) 569

by atamido (#44272869) Attached to: Whistleblowing IT Director Fired By FL State Attorney

Had Martin used his fists alone I would absolutely want to see Zimmerman punished

I also haven't followed the case at all, but from the comments it appears that some of the missing evidence indicated that Martin was involved in illegal street fighting. As such, he had special training/experience with his fists that could cause them to be classified as deadly weapons.

Comment: Re:So sue 'em. (Score 4, Insightful) 569

by atamido (#44272753) Attached to: Whistleblowing IT Director Fired By FL State Attorney

He can't sue.

Florida is a "At Will Employment" state. The only thing you can sue for here is Discrimination. In Florida, you can be fired for anything, with or without reason, and you can quit, with or without reason.

Emphasis mine. What you said is not strictly true. You can be fired without reason, but if you provide a reason then it can't violate the numerous federal laws on the matter. For example, you can't fire someone for race, religion, sex, etc.

In this case they appear to have given a reason, and the reasons would seem to violate federal whistle blower laws, so he can sue.

Comment: Re:Hmmm (Score 1) 325

by atamido (#43816805) Attached to: Predicting IQ With a Simple Visual Test

Then look at the graph (the one in the article with blue and red dots). That is a TERRIBLE correlation. It might be significant from a purely statistical argument, but the correlation is so weak that it would be difficult to eliminate other factors.

I think you might be looking at the graph the wrong way. The correlation isn't how long it takes to spot the direction. The correlation is the difference in time that it takes to see the small vs large. They certainly need a lot more data points, but it looks like they could nail down an IQ score to about 20 points within a couple of minutes with a high degree of confidence. That's pretty impressive. It might give a better starting point for tests to give a more accurate measurement.

Of course, then you have the question as to if an IQ score correlates to anything useful in life, or provides any useful information.

Comment: Re:Blackberry Enterprise (Score 1) 125

by atamido (#43759385) Attached to: How BlackBerry Is Riding iOS and Android To Power Its Comeback

Blackberry Enterprise is one of those products that I really just have to scratch my head at. It has always seemed to me that encouraging users to treat as secure something which is easily lost, stolen, or damaged is a fundamentally flawed concept for a business model.

Are you insane? Or you just have no idea what a blackberry enterprise server (BES) does?

The BES manages strong encryption (AES by default) on the devices. The encryption keys are found only in two places: one the BES, and on the blackberry itself.

The mobile carrier doesn't have the keys, and RIM doesn't have the keys. So if a government comes calling with a warrant, RIM doesn't have anything to give them. It's a very elegant design.

The BES can force mandatory policies onto the blackberries, such as strong full-disk encryption, strong passwords, remote tracking, remote wiping, remote locking, wiping if the phone doesn't check in regularly, restricting what apps can access, and many, many other things.

All of the encryption stuff is exactly like ActiveSync, which comes on every device worth it's salt, and every organization that has Exchange (is there an enterprise that doesn't?). All of the policies are included in ActiveSync, except for "remote tracking, remote locking, wiping if the phone doesn't check in regularly, restricting what apps can access", and most of those are easily implemented with a mobile management program. That said, most enterprises I've seen really only care about the features ActiveSync comes with.

We have a BES that we are actively trying to get the last few stragglers off of so we can decommission it. Really, once the first iPhone supported ActiveSync (2008?) BES stopped being relevant to any sysadmin I've interacted with.

Comment: Re:Cost Per Lumen? BS! (Score 1) 308

by atamido (#43548799) Attached to: Cause of LED Efficiency Droop Finally Revealed

The only things I see holding back LED bulbs are misinformation and lack of availability

Not true. You only give lifetime values for current prices. If the price of an LED bulb will be a quarter of the current cost in five years (a likely scenario), you could keep using incandescents for the next five years and then switch to LED. Your total cost will still be half of what it would have been if you had purchased LED bulbs now.

For me, it makes far more sense to replace heavily used bulbs with CFL now, and then switch to LED once they drop to a reasonable price.

16.5 feet in the Twilight Zone = 1 Rod Serling

Working...