3 ms·

I'd say there are very few people who are doing things so out there that they cannot explain what they are doing. We aren't that special. We aren't magical uni

I'd say there are very few people who are doing things so out there that they cannot explain what they are doing. We aren't that special. We aren't magical unicors who work in all corners of our galaxy. No we are programmers, PhDs, mechanics, engineers. Our jobs share a lot in common with each other and can be easily broken down to their base components. That's the job of a programmer like Carmack. To break things down into their simplest form and represent them clearly and consisely. Frankly for him it's exactly what he's paid to do. > There is no law of nature stating that the maximum complexity anyone can invent is smaller than the complexity the average person can understand. Unless you think you, or this hypothetical person, is so unbelievably smart that they can operate on a plane so far above that of the common man that they can't explain what they do up there. Then I can see how you can think that way. I personally don't belive that is the case. We as programmers don't do that and very few people do. For example, lets ask this question: How do you advocate for funding for the job you do? You need to market yourself. How do the people you market to market to their bosses? It's not PhDs all the way down the same way it's not turtles all the way down. There is a simple, consise, and understandable explination for every activity that humanity has done since the beginning of time. Wheather I know how to formulate it is one thing but one must exist.

marcosdumay 10y ago
> I'd say there are very few people who are doing things so out there that they cannot explain what they are doing. So, those very few people out there are basically fucked up with no recourse? How few are they? Anyway, yep, we are normal people. Normal people that spent a decade or four learning something very specialized. Now, why should we take for granted that people that never learned anything in our specialty can understand anything we know in a week?
abraae 10y ago
Spoken like an intellectual (not an insult - an observation). Could John Carmack explain binary space partitioning in an interesting way that made Joe Average feel like he got it? Most likely yes. But would Joe Average really "get it"? Could he make a meaningful judgement as to whether some other piece of software implementing BSP was more or less a copy of some other piece? (just using BSP as an example - not a good one) I sincerely doubt it. As the saying goes, a little knowledge is dangerous. Look at the number of people who claim that global warming is a myth because they saw a snowstorm last week. Its fashionable, and feels liberal, to claim that we're all much the same, but we're not. John Carmack is what many would call a genius, who can conceive of things that are simply beyond the ken of the bulk of society. Sure those people have other skills and virtues - but epecting them to make rational decisions about complex technical matters while two opposing sets of attorneys bark at them is...ridiculous.
- bshimmin 10y ago
  a little knowledge is dangerous The quote from Alexander Pope is actually "A little learning is a dangerous thing". People often get this wrong, which is probably a tiny bit ironic.
  - abraae 10y ago
    Touché! I stand mortified and corrected.
ctdonath 10y ago
That's the job of a programmer like Carmack. To break things down into their simplest form and represent them clearly and consisely. To machines and to fellow experts, yes - but not to the general population, and not to people who may be persuaded to jail him, or fine him at tremendous expense, if he doesn't adequately explain it while someone else is being paid big money to obfuscate his explanation. This isn't advertising or fundraising. This isn't helping people understand who have an ulterior motive for understanding. This is a hostile environment trying to convict him, an environment that leverages ignorance. And the complex topic at hand isn't sufficiently explained to ordinary people in a matter of hours, he'd have to put them thru a complete "high school thru post-doctoral education" process and bring them up in the vicinity of his level as an "among the greatest of all time in the field". When learning about a particular topic, I was directed to keep notes in a safe-deposit box precisely so if I was ever criminally charged for related reasons, I could literally put the entire jury thru the full education on the subject - and that would take about a month of courtroom time. I can't imagine Carmack's lawyer arranging to put a jury of commoners thru arithmetic-to-post-doctoral-rendering to "explain what he's doing". I've heard it said that the simple (in rules) game of Go can be so complex that it takes mastery of seventh-level skills just to comprehend what ninth-level skills are doing. You can't just take someone who barely learned the rules and explain, in a short time, what the grandmasters are doing to win.
dagss 10y ago
The point is not whether it can be understood. The point is that if two sides both make convincing arguments, how do you pick which one is correct? Without knowing the subject matter yourself, as a human one will usually fall back to confidence and charisma. Your brain unnoticeably substitutes "which one is correct" (which is hard / impossible to answer at a dumbed down level) with "who do I trust" which is ruled by charisma and confidence more than anything else. The wrong projects get funded all the time. In the legal system one should do better.
deong 10y ago
The problem is that there's an adversary who, in practice, is not attempting to lead the jury to the correct conclusion, but lead them to a conviction. And that adversary will be equally able to explain in an understandable way why, despite your explanation to the contrary, you in fact are guilty of the charges. The system works when juries are able to differentiate between expert witnesses, and that may very well require more expertise than just applying everyday common knowledge to two competing expositions.
seanp2k2 10y ago
In this case, I do believe that Carmack is so intelligent and so far removed from a random citizen when it comes to graphics engines that it would take probably thousands of hours for a normal person to actually understand what was happening here, if they even cared, which they would not. "So let's start with a voxel"... [much time passes] "so for the sake of making the rendering pipeline efficient, its common to ..." Without any sort of grasp on the range of solutions for the piece of this system in question, I find it absurd that a normal citizen would be tasked with ruling on this.

marcosdumay 10y ago

> I'd say there are very few people who are doing things so out there that they cannot explain what they are doing. So, those very few people out there are basically fucked up with no recourse? How few are they? Anyway, yep, we are normal people. Normal people that spent a decade or four learning something very specialized. Now, why should we take for granted that people that never learned anything in our specialty can understand anything we know in a week?

abraae 10y ago

Spoken like an intellectual (not an insult - an observation). Could John Carmack explain binary space partitioning in an interesting way that made Joe Average feel like he got it? Most likely yes. But would Joe Average really "get it"? Could he make a meaningful judgement as to whether some other piece of software implementing BSP was more or less a copy of some other piece? (just using BSP as an example - not a good one) I sincerely doubt it. As the saying goes, a little knowledge is dangerous. Look at the number of people who claim that global warming is a myth because they saw a snowstorm last week. Its fashionable, and feels liberal, to claim that we're all much the same, but we're not. John Carmack is what many would call a genius, who can conceive of things that are simply beyond the ken of the bulk of society. Sure those people have other skills and virtues - but epecting them to make rational decisions about complex technical matters while two opposing sets of attorneys bark at them is...ridiculous.

bshimmin 10y ago
a little knowledge is dangerous The quote from Alexander Pope is actually "A little learning is a dangerous thing". People often get this wrong, which is probably a tiny bit ironic.
- abraae 10y ago
  Touché! I stand mortified and corrected.

ctdonath 10y ago

That's the job of a programmer like Carmack. To break things down into their simplest form and represent them clearly and consisely. To machines and to fellow experts, yes - but not to the general population, and not to people who may be persuaded to jail him, or fine him at tremendous expense, if he doesn't adequately explain it while someone else is being paid big money to obfuscate his explanation. This isn't advertising or fundraising. This isn't helping people understand who have an ulterior motive for understanding. This is a hostile environment trying to convict him, an environment that leverages ignorance. And the complex topic at hand isn't sufficiently explained to ordinary people in a matter of hours, he'd have to put them thru a complete "high school thru post-doctoral education" process and bring them up in the vicinity of his level as an "among the greatest of all time in the field". When learning about a particular topic, I was directed to keep notes in a safe-deposit box precisely so if I was ever criminally charged for related reasons, I could literally put the entire jury thru the full education on the subject - and that would take about a month of courtroom time. I can't imagine Carmack's lawyer arranging to put a jury of commoners thru arithmetic-to-post-doctoral-rendering to "explain what he's doing". I've heard it said that the simple (in rules) game of Go can be so complex that it takes mastery of seventh-level skills just to comprehend what ninth-level skills are doing. You can't just take someone who barely learned the rules and explain, in a short time, what the grandmasters are doing to win.

dagss 10y ago

The point is not whether it can be understood. The point is that if two sides both make convincing arguments, how do you pick which one is correct? Without knowing the subject matter yourself, as a human one will usually fall back to confidence and charisma. Your brain unnoticeably substitutes "which one is correct" (which is hard / impossible to answer at a dumbed down level) with "who do I trust" which is ruled by charisma and confidence more than anything else. The wrong projects get funded all the time. In the legal system one should do better.

deong 10y ago

The problem is that there's an adversary who, in practice, is not attempting to lead the jury to the correct conclusion, but lead them to a conviction. And that adversary will be equally able to explain in an understandable way why, despite your explanation to the contrary, you in fact are guilty of the charges. The system works when juries are able to differentiate between expert witnesses, and that may very well require more expertise than just applying everyday common knowledge to two competing expositions.

seanp2k2 10y ago

In this case, I do believe that Carmack is so intelligent and so far removed from a random citizen when it comes to graphics engines that it would take probably thousands of hours for a normal person to actually understand what was happening here, if they even cared, which they would not. "So let's start with a voxel"... [much time passes] "so for the sake of making the rendering pipeline efficient, its common to ..." Without any sort of grasp on the range of solutions for the piece of this system in question, I find it absurd that a normal citizen would be tasked with ruling on this.