Stubborn attempts to think clearly

Narrative as data compression

2009-01-18T12:30:00.002+02:00

The ability of our working memory to hold different pieces of information is limited. We can usually remember something like six random digits, but not much beyond that. Some people manage to remember much more but they usually do it by using some kind of memory trick. Rather than remembering six-three-six-five-six-seven, they remember sixty three - sixty five - sixty seven. They compress the data from six separate pieces to three. I believe we do this with almost all data that we have. We combine it into narratives.

E. M. Forester pointed out the difference between "The king died and then the queen died.", where we just have a sequence of events, and "The king died and then the queen died of grief.", where we have have given the sequence of events a casual connection and thus have a plot. We add these casual connections to our data to gather them together into a narrative to compress them. For it is much easier to remember the narrative than to remember the separate pieces of information.

The problem arises because this data compression isn't lossless. Things that then don't fit our narrative explanation can get left out. This gives us the confirmation bias. We can also go in the other direction and start adding other information to the narrative just because the fit, even though there is nothing connecting them to the events in question. Do this enough and you get conspiracy theories.

Paul Thurrott plans on dying young.

2007-10-12T17:16:00.000+03:00

Paul Thurrott, of WinSuperSite, has started a new series he calls Digital Media Core, and has started by explaining why MP3 is the one audio format to rule them all, and why using any other is irrational.

Not surprisingly the article leaves one wanting. There is simple silliness there, like he claims that WMA offered: "better compression, better fidelity, and smaller file sizes" than MP3. This is true, but better compression, better fidelity, and smaller file sizes is saying the same thing three times over. Better compression means better fidelity at the same file size, or a smaller file size at the same fidelity. If the compression is good enough it means better fidelity at smaller file size. In the end they are all the same thing. He also offers this brilliant comparison between AAC and WMA: "very similar [...] file sizes at identical bit rates". What makes this such a brilliant observation is that bit rate pretty much is file size. Bit rate times length gives you the stream size (VBR can throw it off). Then as overhead you add the container and whatever meta data it contains and you got file size. The fact that the file size are similar at identical bit rates is evident if you know what you are talking about.

The gist of Thurrott's article is that you should use MP3 when ripping music because of its wide compatibility and that is flaw, bad compression, isn't an issue with todays storage. Generally I agree with him. MP3 is a great format to choose because of its almost universal compatibility. What makes his article such a catastrophe is that MP3 has two other, worse flaws which he doesn't even mention. Recommending someone to go MP3 without giving them an adequate understanding of the consequences seems wrong to me.

MP3's two other faults, besides bad compression, is the fact that it is lossy, and the transcoding related issues, and its licensing uncertainness. The legal environment surrounding MP3 remains uncertain and every once in a while a lawsuit pops up. There remains the possibility that at some future point in time the widespread compatibility of MP3 decreases sharply because of some legal issue.

The legal issues aggravates the second issue. Transcoding with lossy formats. When you transcode one lossy format into another you loose quality because you end up with two different compressions on top of each other. So when you go MP3 you are stuck with MP3 unless you want to transcode and live with the quality loss. This I see as the biggest advantage of lossless formats. The lossless compression is completely removed when the file is decoded so there is no quality loss from double compression. This is why I rip my CDs in a lossless format.

The potential legal pit holes isn't the only reason why you might want to change the format of your music files. With the rapid technological advancement who knows what formats with what features we might have ten of twenty years from now. The ability to move you music to a new format without the quality loss, or needing to re-rip it, may be essential to deal with the future.

As I mentioned earlier I rip into a lossless format. What I chose was FLAC, and not because of the reasons Thurrott claims I have: "The people who do use (and advocate) these formats are generally more concerned with religious issues surrounding their disliking of proprietary technologies or products made by companies like Apple and Microsoft." Although the proprietary nature of Microsoft's format was why I chose the open source FLAC, it has nothing to do with religious issues of any kind of principles, and everything to do with prudence. FLAC is well documented which means if punch comes to shove I can write my own FLAC decoder, which isn't nearly as easy to do with proprietary formats.

He makes a point about how his life is too short for him to bother to manage two sets of your music collection. I presume he plans on dying before he has to rip a new collection as well. This is a valid point though I wonder how much trouble it really would be to set your system up so that it makes two copies of everything you rip, one, in MP3, for your daily usage, and one, in lossless, that it back up somewhere.

Lastly I should add a side note on AAC. Thurrott dismisses AAC due to its inferior compatibility, especially in Microsoft software. At the moment this remains true, although adding AAC support to WMP and WMC is easy with the lighweight and sleek CCCP, which is free. Also, seeing as AAC is part of MPEG-4 along with H.264/AVC, which is Thurrott's favorite video format, which is sweeping the net, I would guess that it is only a matter of time before Microsoft does MPEG-4.

The gates have opened

2007-09-21T10:33:00.000+03:00

ThoughtVent, the sister site to Bleeping Computer, has opened its gate. It, as the name implies, is a general chat away forum. I you feel you need to vent your thought drop by.

Robert Jordan dies

2007-09-15T20:33:00.000+03:00

Robert Jordan has passed away.

Author Robert Jordan, whose was best known for the Wheel of Time series of fantasy novels has died of a rare blood disease aged 58.
Jordan - whose real name was James Oliver Rigney Junior - died at the Medical University of South Carolina.
His personal assistant, Maria Simons, said the disease caused the walls of his heart to thicken.
Jordan's first fantasy book, The Eye of the World, was published in 1990 and went on to sell millions of copies.
He also wrote a trilogy of historical novels set in Charleston under the pen name Reagan O'Neal in the early 1980s.
After turning his hand to the Wheel of Time series, he went on to publish 11 books and had been working on the 12th at the time of his death.
Jordan's books tell of Rand al'Thor, also known as The Dragon Reborn, who is destined to battle evil in a mythical land.
The author is survived by his wife, Harriet McDougal Rigney.

This proves it: There is no God.

Cylons versus United States

2007-08-10T17:41:00.000+03:00

Non of the computers on the Battlestar Galactica are integrated or networked to stop the Cylons from being able to hack into them. This makes a certain amount of sense. To be able to hack a computer you need to interface with it, have a channel of communication. Networking computer adds interfaces to them.

Now we hear about telephone wiretapping on a grand scale. The only way to do something like this is to automate it. Have computers listening i and the flagging calls with certain keywords. An ECHELON type thing. All these computers doing this tapping then need to report to NSA or whoever is behind it, which means it is all networked.

Now were back to the Cylon reference. This provides ample of surface to attack for an enterprising hacker. You need not only worry about what the government might be hearing, but the rest of the world as well.

We also hear discussions in different countries about how ISPs are supposed to add wiretapping APIs to their servers for police use. This is another hacking opportunity served up on an silver platter. Despite what you see in the movies a hacker can't make a computer do something it can't do. He is limited to perverting already exsisting functions to his needs. Adding sensitive abilities like the wiretapping stuff greatly increases the amount of damage a hacker can do.

And here I though I was going to get the last laugh

2007-05-23T15:45:00.000+03:00

HA! I should have known better. This is a continuation, for the lack of a better word, of my clash with the GeForce 8800GTX.

"So, what happened?" you ask. Windows wanted me to reactivate, after changing just the graphics card. And Microsoft claims it isn't so sensitive to hardware changes. Of to activating it then. Won't work: "Your product key is already in use" it tells me. What? Yeah, well I'm using it. Lets just try to enter it all over again. Nope, still won't accept it. Phone activation, here I come. After navigating through the automated system it asks me for an installation ID. Just one problem: My installation ID field is empty. Right about now Windows Update tells me it needs to restart the computer.

After the restart I figure I might try to activate it again. Product key is still in use. But now I have a installation ID. I call the automated system again. It won't accept my installation ID. Arrgh! Now it connects me to a real life person. He is so quiet I can barely hear him. After explaining my problems to him, he puts me back on the automated system, which read up a number for me to enter. It is quick about it, really putting my numpad skills through their paces. At least it accepts the number and I get Windows activated once again.

I must say this was way too many hops to jump through just to get Windows to accept that I upgraded my graphics card. Now I'm waiting on what will be The Revenge of the Graphics Card.

Mr Alpha VS GeForce 8800GTX

2007-05-21T14:59:00.000+03:00

I actually treated myself to a 8800GTX. Getting it installed turned into a bloody clash of titans.

I was worried that it wouldn't fit in my Antec P180 case. The card is huge; 27.9 cm to be exact. Taking it out of the box the first time was an awe inspiring experience. Turns out if fits! Had to reroute some cable, but it fit. Barely.

Seeing this as an good omen I shut down the computer. Then I remembered I might want to uninstall the old drivers first to make this switch as painless as possible. So I booted the machine back up, and it gave me a black screen instead of the welcome screen. Telling myself it was just a bad mistake I restart the computer hoping for everything to start as it should. No such luck. Restart again, this time to safe mode. In safe mode I uninstall the graphics drivers. Lo and behold, the machine now booted properly.

With that heart wrenching moment out of the way I set myself to mounting the new graphics card in the case. It won't fit! But it just did and I can't figure out what is wrong now. After some head-into-wall-pounding, world-in-general-cursing and other standard troubleshooting techniques I noticed that my northbridge heatsink was in the way. The motherboard (ASUS A8N-SLI Deluxe) orginally came with a deceptively small, but loud fan, which I had replaced with a Zalman northbridge heatsink. Some of the northbridge pins got in the way of the 8800GTX HSF. When I first checked the fit I hadn't actually tried to mount the card.

"I can fix this, I can fix this", I kept telling myself. If I break of a few pins and bend away a few others this should work. Just to prove the world was against me the northbridge heatsink was attached with plastic clips you had to remove the motherboard to get to. I wasn't going to let that stop me. Digging up some tools of destruction from my toolbox I set about doing some open computer surgery. I do not recommend trying to rip things loose from a heatsink while it is still mounted in the computer. You can easily damage the motherboard. But I have never been good at heading my own advice, so I set about demolishing the heatsink while it still was mounted. Partially destroyed heatsinks can be really sharp, let me tell you. Soon I was bleeding all over the place. Band-aid to the rescue!

When I had gotten the card into the computer it was time to boot up. Everything worked fine. (Had you there for a sec, didn't I?) Of to nVIDIA's site to get the latest drivers. After a few more reboots I was of to play System Shock 2.

Joost Invites

2007-05-04T00:19:00.000+03:00

Nice! Seems the Joost team delivered on their promise of unlimited invites for I now have 999 invites.

The Socratic Paradox

2007-05-01T17:37:00.000+03:00

Socrates claims to have no knowledge. This has traditionally been seen as a dialogue tactic, and that he isn't serious. It has even become known as Socratic irony.

Problem is that it isn't really consistent with what he actually says. As an famous example when Chaerephon asked the Oracle at Delphi: "Is there anyone wiser than Socrates?", and the oracle answered, "No". About this Socrates says: "When I heard this I kept thinking: 'What on earth does the god mean? What is he hinting at? For I am aware of being wise in nothing, great or small. What then could he mean by saying that I am wise?'". This is far from the only place where Socrates clearly denies any knowledge. One of his main rules of elenctic dialogue is to say what you believe.

So, Socrates really didn't have any knowledge, where's the paradox? For a guy claiming to have no knowledge he sure says he knows a lot: ". . . but that to do injustice and I know to be evil and disobey my superior, god or man, this base" and "I know well that if you will agree with me on those things which my soul believes, those things will be the very truth".

Can't make it too easy now, can he? Clearly stating that he knows nothing and the turning around five minutes later and claiming to know important ethical truths. For somebody who had spent his life digging through the statements of others it seems unlikely that he isn't aware of the contradiction he said.

He claims to know what makes life worth living, and then gives us a riddle.

Joost

2007-04-22T17:31:00.000+03:00

I've been using the Joost BETA now for little over a week. For those that don't know, Joost is a IPTV application from the makers of Skype, which combines P2P technology with traditional streaming to deliver (supposedly) DVD quality, full-screen, free TV.

It is ad-based, with one 30 second ad every 10-15 min, which is much less than on normal TV. One problem I'm hoping they can fix is that the ads start randomly, sometimes even in the middle of a word.

The promised DVD quality isn't there. Some show are almost DVD quality, like a documentary about snakes (did you know snakes come with built in optical zoom?), while other fast-paced shows like Indy car racing is unwatchable, due to the compression artifacts. Bandwidth wise it uses, if I got all the numbers and math correct, 700 kbps down and 230 kbps up.

It is nice, easy to use, no more complicated than a media player. Vector based interface, meaning it zooms bigger without any problems. At the moment channels work like simple playlists, which works fine when there still is only a little content, but they will have to do something about it if they get a lot more content, which hopefully they will.

What about the content, then, you ask? Mostly a bunch of smaller independent channels. A couple of better known ones like Comedy Central (not available here), National Geographic (not available here) and Gong. CBS has signed up to show CSI, CSI: NY and CSI: Miami, NCIS, Numb3rs, Jericho and Survivor.

Realist

2007-04-21T10:55:00.000+03:00

Somebody told me that pessimists always claim they are realists. This struck me as odd, not because it isn't true, but because there was a judgment in it, as if that makes a pessimist less than an optimist. Doesn't an optimist also consider himself an realist? In fact, I would go further and say it has to be so.

A realist is someone who sees the world as it is, without putting a positive nor a negative light on matters. An optimist put a positive light on the world, and sees things as better than they are. But what the optimist sees and believes about the world is what he believes the world to actually be, and thus sees himself as an realist. For what he sees as the world is what he believes the world to be. It is not like he could believe the world to be something other than what he believes the world to be.

Everybody must, deep down inside, believe that what they believe the world to be, is as close to what the world is as it is possible for them to believe, otherwise they would not believe it.

If the BIOS ain't broke, don't fix it

2007-04-19T21:49:00.000+03:00

It's good advice. Screwing up a BIOS update can make the computer unbootable.

But (and there's always a but, isn't there?) how do you know if it is broke? A buggy BIOS can cause practically any conceivable computer issues except pop-up storms. Any kind of performance issues, stability issue or compatibility issue could be caused by a faulty BIOS.

Practically every piece of data going through the system goes through the hands of the BIOS. There was the well know problem on the 680i motherboards when they came out that was corrupting data on SATA drives. There was a MSI motherboard where a bug in the BIOS was causing the southbridge to overheat. Lots of motherboard went haywire with the 7950GX2, not expecting to have to look for a graphics card behind a PCIe splitter.

So the question becomes: Sure, don't fix it unless it ain't broke, but how do you know it is broke?

Scientific truths

2007-04-19T21:35:00.000+03:00

There are many theories about truth. Scientific truth seems to fall into several of them.

It seems to fit with the coherence theory of truth, especially if you subscribe to Khun's paradigm idea. It also fits with consensus theory of truth, considering science is a institutionalized form of knowledge, with its authorities (experts), journals and peer review. Even the pragmatic theory of truth finds a foothold in scientific truth. There is Occam's razor, the aversion to ad hoc theories and the requirement for they theories to be ably to make measurable predictions.

The really interesting part is that scientific truth doesn't need the correspondence theory of truth, which is the one the comes most naturally. There is no need for the theory to actually correspond to the external world. If you had two theories that in different ways described something, and both were just as provable and stand up to all tests equally well. The one chosen would be the easier one, or the one that majority of the scientific community backs, or the one that fits into the prevailing paradigm. Which one actually corresponds to reality doesn't matter.