1.5M ratings
277k ratings

See, that’s what the app is perfect for.

Sounds perfect Wahhhh, I don’t wanna
blog.someben.com

Hashing Language – Some Ben?

Statistical NLP bag-of-words models that use the hashing trick have roughly the same accuracy as models that operate on the full bag-of-words dimensionality. There are two reasons why hash collisions in the low-dimensional space of the hash function’s output range do not trash our models. First any collisions that do occur, probably occur between two rare words. In many models, rare words do not improve the model’s regression / classification accuracy and robustness. Rare words and ngrams are said to be non-discriminatory. Now even if rare words are discriminatory in your problem domain, probability suggests the rare words do not co-occur in the same document. For this reason, the two rare words can be thought of as “sharing” the same representation in the model, whether this is decision tree sub-trees or a coefficient in a linear model. The Forman & Kirshenbaum paper says “a colliding hash is either unlikely to be selected as a feature (since both words are infrequent) or will almost always represent the word that led the classifier to select it.”

Well, I doubt I’m going to be fitting n-gram models again any time soon, but I’ll keep this in mind if I do!

math programming
mailadreapta
argumate

nice destructor you’ve got there sneered Rust, shame if it was never called thanks to cyclic references

poipoipoi-2016

Isn’t this true of any language that does reference counting and not that smart garbage collection?

argumate

also true of other garbage collection strategies, as the null gc which never collects anything is perfectly legal and in some cases even optimal, such as for short-lived processes which do not do much allocation.

destructors are bad.

rocketgarden91

I thought it was pretty common knowledge to never trust destructors, and expect that they’ll never be executed

argumate

a proper linear type system would help with resource tracking issues, and personally I wish Rust required explicit frees.

mailadreapta

I wish Rust required explicit frees.

Of all the bad opinions you’ve ever uttered here on this hellsite

Source: argumate programming

I once read “Cat and Mouse” by Günter Grass. I didn’t particularly like it, but I really hated this bit near the beginning:

Since then gull droppings had dried on the rust. In all kinds of weather the gulls flew sleek and smooth, with eyes like glass beads on the sides of their heads, grazing the remains of the pilothouse, then wildly up again, according to some indecipherable plan, squirting their slimy droppings in full flight – and they never fell into the soft sea but always on the rusty superstructure. Hard, dense, calcareous, the droppings clung fast, side by side in innumerable spots, or heaped up in mounds. And always when we sat on the barge, fingernails and toenails tried to chip off the droppings. That’s why our nails cracked and not because we bit our fingernails – except for Schilling, who was always chewing at them and had nails like rivets. Only Mahlke had long nails, though they were yellow from all his diving, and he kept them long by neither biting them nor scratching at the gull droppings. And he was the only one who never ate the chips we broke loose – the rest of us, because it was there, chewed the stony, shell-like mess into a foaming slime, which we spat overboard. The stuff tasted like nothing at all or like plaster or like fish meal or like everything imaginable: happiness, girls, God in His heaven.

Startup idea: a new ice cream flavor that tastes like everything imaginable. I’m keeping the recipe a secret though.

bad humor
argumate
argumate

rationalist-adjacent by transitive closure

hpgross

rationalist-adjacent-adjacent = rationalist-adjacent = rationalist-[adjacent]^n

argumate

sounds like the kind of thing a rationalist would post :|

stumpyjoepete

Well, if “adjacent” is a linear transform of dimension k, then by the Cayley-Hamilton theorem, any analytic function of “adjacent” can be represented by a matrix polynomial of degree at most k. So, at a certain point higher of values of “n” stop adding anything.

math bad humor
stumpyjoepete
t.co

The Next Big Blue-Collar Job Is Coding
therainstheyaredropping

> Politicians routinely bemoan the loss of good blue-collar jobs. Work like that is correctly seen as a pillar of civil middle-class society. And it may yet be again. What if the next big blue-collar job category is already here—and it’s programming? What if we regarded code not as a high-stakes, sexy affair, but the equivalent of skilled work at a Chrysler plant?

> Among other things, it would change training for programming jobs—and who gets encouraged to pursue them. As my friend Anil Dash, a technology thinker and entrepreneur, notes, teachers and businesses would spend less time urging kids to do expensive four-year computer-­science degrees and instead introduce more code at the vocational level in high school. You could learn how to do it at a community college; midcareer folks would attend intense months-long programs like Dev Bootcamp. There’d be less focus on the wunderkinds and more on the proletariat.

> These sorts of coders won’t have the deep knowledge to craft wild new algorithms for flash trading or neural networks. Why would they need to? That level of expertise is rarely necessary at a job. But any blue-collar coder will be plenty qualified to sling Java­Script for their local bank. That’s a solidly middle-class job, and middle-class jobs are growing: The national average salary for IT jobs is about $81,000 (more than double the national average for all jobs), and the field is set to expand by 12 percent from 2014 to 2024, faster than most other occupations.

argumate

But any blue-collar coder will be plenty qualified to sling Java­Script for their local bank.

sling JavaScript

for their local bank

plenty qualified

hell world, etc.

stumpyjoepete

Ok, as much as I also would like to complain about how idiots rewriting all banking software in php will break the seventh seal and usher in the apocalypse, that wouldn’t really be engaging with the central thesis of the article. The real question is this: Will coding become an accessible, reasonably well paying job option in the way that many blue collar jobs were in the post-war period?

No, no it won’t. What the hell are they even thinking?

Quoting Scott:

A lot of people [at the Asilomar Conference on AI] were really optimistic that the solution to technological unemployment was to teach unemployed West Virginia truck drivers to code so they could participate in the AI revolution. I used to think this was a weird straw man occasionally trotted out by Freddie deBoer, but all these top economists were super enthusiastic about old white guys whose mill has fallen on hard times founding the next generation of nimble tech startups. I’m tempted to mock this, but maybe I shouldn’t – this From Coal To Code article says that the program has successfully rehabilitated Kentucky coal miners into Web developers. And I can’t think of a good argument why not – even from a biodeterminist perspective, nobody’s ever found that coal mining areas have lower IQ than anywhere else, so some of them ought to be potential web developers just like everywhere else. I still wanted to ask the panel “Given that 30-50% of kids fail high school algebra, how do you expect them to learn computer science?”, but by the time I had finished finding that statistic they had moved on to a different topic.

Computer programming is about context-free symbol manipulation. (There are probably other jobs that are like this too.) In order to get a machine to do something, you have to be able to simulate that machine in your head. I don’t mean in some super low-level way; I just mean that you need to be able to run a program in your head to see if it does what you intend it to do (or, as is often the case, see why it’s doing something unintended). Machines don’t do what you want them to do. They do what you tell them to do.

This is not a trivial barrier to entry. Having TA’d programming classes before, I can confidently say that some people do not get this at all. When someone would ask what a particular program did, many students would attempt to divine the intent of the writer of the program. That is, they would answer what they thought the program was supposed to do (presumably based on the comments and variable names). This is not a useless skill. It gives you the right answers when a program is correct. As a working developer, it’s useful when trying to reverse engineer requirements from existing code (that may or may not actually fulfill those requirements). It’s not very useful in debugging though. And that’s mostly what programming is–taking incomplete or buggy attempts at solving a problem and refining them into a working solution.

In the class itself, people who didn’t really get it could skate along for a while, since they were mostly looking at code that worked. Eventually they would crash and burn, though, as the complexity of the assignments would outstrip their ability to make a working program on the first attempt. This is shitty from a pedagogical perspective, so a lot of my review sessions focused on this skill of simulating programs. One type of exercise that worked really well (both for breaking through this conceptual barrier as well as providing good practice) was to generate nonsense programs and then ask what they did. There’s really no way around simulating a program when the variable names are all single letters and the code snippet does not actually have any intelligible intent. Some people would eventually get it. I’d like to think that with enough time and effort that everyone could get it, but that’s probably not realistic even when we’re just talking about college students at selective institutions.

So, whenever I read something about the loss of well-paying blue-collar jobs that talks about retraining and the tech industry, I feel baffled that they don’t consider the obvious objections. Actually, scratch that. I feel insulted!

urpriest

Question: isn’t this true of mechanical jobs as well? That is, if you’re fixing a car or a sink or the like, don’t you have to model it in your head, and ignore the designer’s intent?

stumpyjoepete

For reasons that are completely mysterious to me, these abilities seem to be somewhat independent. FWIW, I don’t think I would be a very good auto mechanic. I have a vague feeling that mental rotation might somehow be involved in an explanation of what cognitive skills auto mechanics have vs those of computer programmers.

I’m seriously very interested in an answer to this question, both because I really can’t think of a good explanation and also because it would seem to be pretty relevant to finding future employment for current blue collar workers.

stumpyjoepete

collapsedsquid:

Could be that it’s because those are physically manipulated and observed

Maybe? I guess that would suggest that we have to not only teach programming with Logo but also reframe all real-life programming problems in terms of turtles painting?

Source: therainstheyaredropping programming economics ai
enye-word
t.co

The Next Big Blue-Collar Job Is Coding
therainstheyaredropping

> Politicians routinely bemoan the loss of good blue-collar jobs. Work like that is correctly seen as a pillar of civil middle-class society. And it may yet be again. What if the next big blue-collar job category is already here—and it’s programming? What if we regarded code not as a high-stakes, sexy affair, but the equivalent of skilled work at a Chrysler plant?

> Among other things, it would change training for programming jobs—and who gets encouraged to pursue them. As my friend Anil Dash, a technology thinker and entrepreneur, notes, teachers and businesses would spend less time urging kids to do expensive four-year computer-­science degrees and instead introduce more code at the vocational level in high school. You could learn how to do it at a community college; midcareer folks would attend intense months-long programs like Dev Bootcamp. There’d be less focus on the wunderkinds and more on the proletariat.

> These sorts of coders won’t have the deep knowledge to craft wild new algorithms for flash trading or neural networks. Why would they need to? That level of expertise is rarely necessary at a job. But any blue-collar coder will be plenty qualified to sling Java­Script for their local bank. That’s a solidly middle-class job, and middle-class jobs are growing: The national average salary for IT jobs is about $81,000 (more than double the national average for all jobs), and the field is set to expand by 12 percent from 2014 to 2024, faster than most other occupations.

argumate

But any blue-collar coder will be plenty qualified to sling Java­Script for their local bank.

sling JavaScript

for their local bank

plenty qualified

hell world, etc.

stumpyjoepete

Ok, as much as I also would like to complain about how idiots rewriting all banking software in php will break the seventh seal and usher in the apocalypse, that wouldn’t really be engaging with the central thesis of the article. The real question is this: Will coding become an accessible, reasonably well paying job option in the way that many blue collar jobs were in the post-war period?

No, no it won’t. What the hell are they even thinking?

Quoting Scott:

A lot of people [at the Asilomar Conference on AI] were really optimistic that the solution to technological unemployment was to teach unemployed West Virginia truck drivers to code so they could participate in the AI revolution. I used to think this was a weird straw man occasionally trotted out by Freddie deBoer, but all these top economists were super enthusiastic about old white guys whose mill has fallen on hard times founding the next generation of nimble tech startups. I’m tempted to mock this, but maybe I shouldn’t – this From Coal To Code article says that the program has successfully rehabilitated Kentucky coal miners into Web developers. And I can’t think of a good argument why not – even from a biodeterminist perspective, nobody’s ever found that coal mining areas have lower IQ than anywhere else, so some of them ought to be potential web developers just like everywhere else. I still wanted to ask the panel “Given that 30-50% of kids fail high school algebra, how do you expect them to learn computer science?”, but by the time I had finished finding that statistic they had moved on to a different topic.

Computer programming is about context-free symbol manipulation. (There are probably other jobs that are like this too.) In order to get a machine to do something, you have to be able to simulate that machine in your head. I don’t mean in some super low-level way; I just mean that you need to be able to run a program in your head to see if it does what you intend it to do (or, as is often the case, see why it’s doing something unintended). Machines don’t do what you want them to do. They do what you tell them to do.

This is not a trivial barrier to entry. Having TA’d programming classes before, I can confidently say that some people do not get this at all. When someone would ask what a particular program did, many students would attempt to divine the intent of the writer of the program. That is, they would answer what they thought the program was supposed to do (presumably based on the comments and variable names). This is not a useless skill. It gives you the right answers when a program is correct. As a working developer, it’s useful when trying to reverse engineer requirements from existing code (that may or may not actually fulfill those requirements). It’s not very useful in debugging though. And that’s mostly what programming is–taking incomplete or buggy attempts at solving a problem and refining them into a working solution.

In the class itself, people who didn’t really get it could skate along for a while, since they were mostly looking at code that worked. Eventually they would crash and burn, though, as the complexity of the assignments would outstrip their ability to make a working program on the first attempt. This is shitty from a pedagogical perspective, so a lot of my review sessions focused on this skill of simulating programs. One type of exercise that worked really well (both for breaking through this conceptual barrier as well as providing good practice) was to generate nonsense programs and then ask what they did. There’s really no way around simulating a program when the variable names are all single letters and the code snippet does not actually have any intelligible intent. Some people would eventually get it. I’d like to think that with enough time and effort that everyone could get it, but that’s probably not realistic even when we’re just talking about college students at selective institutions.

So, whenever I read something about the loss of well-paying blue-collar jobs that talks about retraining and the tech industry, I feel baffled that they don’t consider the obvious objections. Actually, scratch that. I feel insulted!

enye-word

I never even considered telling my bad-at-programming friends that they had to model the program in their heads! Might be worth a shot!

stumpyjoepete

Highly recommended. In addition to the nonsense-program exercises I recommended, there’s the “make a table where columns are variables and rows are time steps and walk through things step by tedious step” technique, which, while obviously overkill in most situations, is very helpful when people aren’t to the point where that’s superfluous.

Source: therainstheyaredropping programming
urpriest
t.co

The Next Big Blue-Collar Job Is Coding
therainstheyaredropping

> Politicians routinely bemoan the loss of good blue-collar jobs. Work like that is correctly seen as a pillar of civil middle-class society. And it may yet be again. What if the next big blue-collar job category is already here—and it’s programming? What if we regarded code not as a high-stakes, sexy affair, but the equivalent of skilled work at a Chrysler plant?

> Among other things, it would change training for programming jobs—and who gets encouraged to pursue them. As my friend Anil Dash, a technology thinker and entrepreneur, notes, teachers and businesses would spend less time urging kids to do expensive four-year computer-­science degrees and instead introduce more code at the vocational level in high school. You could learn how to do it at a community college; midcareer folks would attend intense months-long programs like Dev Bootcamp. There’d be less focus on the wunderkinds and more on the proletariat.

> These sorts of coders won’t have the deep knowledge to craft wild new algorithms for flash trading or neural networks. Why would they need to? That level of expertise is rarely necessary at a job. But any blue-collar coder will be plenty qualified to sling Java­Script for their local bank. That’s a solidly middle-class job, and middle-class jobs are growing: The national average salary for IT jobs is about $81,000 (more than double the national average for all jobs), and the field is set to expand by 12 percent from 2014 to 2024, faster than most other occupations.

argumate

But any blue-collar coder will be plenty qualified to sling Java­Script for their local bank.

sling JavaScript

for their local bank

plenty qualified

hell world, etc.

stumpyjoepete

Ok, as much as I also would like to complain about how idiots rewriting all banking software in php will break the seventh seal and usher in the apocalypse, that wouldn’t really be engaging with the central thesis of the article. The real question is this: Will coding become an accessible, reasonably well paying job option in the way that many blue collar jobs were in the post-war period?

No, no it won’t. What the hell are they even thinking?

Quoting Scott:

A lot of people [at the Asilomar Conference on AI] were really optimistic that the solution to technological unemployment was to teach unemployed West Virginia truck drivers to code so they could participate in the AI revolution. I used to think this was a weird straw man occasionally trotted out by Freddie deBoer, but all these top economists were super enthusiastic about old white guys whose mill has fallen on hard times founding the next generation of nimble tech startups. I’m tempted to mock this, but maybe I shouldn’t – this From Coal To Code article says that the program has successfully rehabilitated Kentucky coal miners into Web developers. And I can’t think of a good argument why not – even from a biodeterminist perspective, nobody’s ever found that coal mining areas have lower IQ than anywhere else, so some of them ought to be potential web developers just like everywhere else. I still wanted to ask the panel “Given that 30-50% of kids fail high school algebra, how do you expect them to learn computer science?”, but by the time I had finished finding that statistic they had moved on to a different topic.

Computer programming is about context-free symbol manipulation. (There are probably other jobs that are like this too.) In order to get a machine to do something, you have to be able to simulate that machine in your head. I don’t mean in some super low-level way; I just mean that you need to be able to run a program in your head to see if it does what you intend it to do (or, as is often the case, see why it’s doing something unintended). Machines don’t do what you want them to do. They do what you tell them to do.

This is not a trivial barrier to entry. Having TA’d programming classes before, I can confidently say that some people do not get this at all. When someone would ask what a particular program did, many students would attempt to divine the intent of the writer of the program. That is, they would answer what they thought the program was supposed to do (presumably based on the comments and variable names). This is not a useless skill. It gives you the right answers when a program is correct. As a working developer, it’s useful when trying to reverse engineer requirements from existing code (that may or may not actually fulfill those requirements). It’s not very useful in debugging though. And that’s mostly what programming is–taking incomplete or buggy attempts at solving a problem and refining them into a working solution.

In the class itself, people who didn’t really get it could skate along for a while, since they were mostly looking at code that worked. Eventually they would crash and burn, though, as the complexity of the assignments would outstrip their ability to make a working program on the first attempt. This is shitty from a pedagogical perspective, so a lot of my review sessions focused on this skill of simulating programs. One type of exercise that worked really well (both for breaking through this conceptual barrier as well as providing good practice) was to generate nonsense programs and then ask what they did. There’s really no way around simulating a program when the variable names are all single letters and the code snippet does not actually have any intelligible intent. Some people would eventually get it. I’d like to think that with enough time and effort that everyone could get it, but that’s probably not realistic even when we’re just talking about college students at selective institutions.

So, whenever I read something about the loss of well-paying blue-collar jobs that talks about retraining and the tech industry, I feel baffled that they don’t consider the obvious objections. Actually, scratch that. I feel insulted!

urpriest

Question: isn’t this true of mechanical jobs as well? That is, if you’re fixing a car or a sink or the like, don’t you have to model it in your head, and ignore the designer’s intent?

stumpyjoepete

For reasons that are completely mysterious to me, these abilities seem to be somewhat independent. FWIW, I don’t think I would be a very good auto mechanic. I have a vague feeling that mental rotation might somehow be involved in an explanation of what cognitive skills auto mechanics have vs those of computer programmers.

I’m seriously very interested in an answer to this question, both because I really can’t think of a good explanation and also because it would seem to be pretty relevant to finding future employment for current blue collar workers.

Source: therainstheyaredropping programming economics ai
collapsedsquid
t.co

The Next Big Blue-Collar Job Is Coding
therainstheyaredropping

> Politicians routinely bemoan the loss of good blue-collar jobs. Work like that is correctly seen as a pillar of civil middle-class society. And it may yet be again. What if the next big blue-collar job category is already here—and it’s programming? What if we regarded code not as a high-stakes, sexy affair, but the equivalent of skilled work at a Chrysler plant?

> Among other things, it would change training for programming jobs—and who gets encouraged to pursue them. As my friend Anil Dash, a technology thinker and entrepreneur, notes, teachers and businesses would spend less time urging kids to do expensive four-year computer-­science degrees and instead introduce more code at the vocational level in high school. You could learn how to do it at a community college; midcareer folks would attend intense months-long programs like Dev Bootcamp. There’d be less focus on the wunderkinds and more on the proletariat.

> These sorts of coders won’t have the deep knowledge to craft wild new algorithms for flash trading or neural networks. Why would they need to? That level of expertise is rarely necessary at a job. But any blue-collar coder will be plenty qualified to sling Java­Script for their local bank. That’s a solidly middle-class job, and middle-class jobs are growing: The national average salary for IT jobs is about $81,000 (more than double the national average for all jobs), and the field is set to expand by 12 percent from 2014 to 2024, faster than most other occupations.

argumate

But any blue-collar coder will be plenty qualified to sling Java­Script for their local bank.

sling JavaScript

for their local bank

plenty qualified

hell world, etc.

stumpyjoepete

Ok, as much as I also would like to complain about how idiots rewriting all banking software in php will break the seventh seal and usher in the apocalypse, that wouldn’t really be engaging with the central thesis of the article. The real question is this: Will coding become an accessible, reasonably well paying job option in the way that many blue collar jobs were in the post-war period?

No, no it won’t. What the hell are they even thinking?

Quoting Scott:

A lot of people [at the Asilomar Conference on AI] were really optimistic that the solution to technological unemployment was to teach unemployed West Virginia truck drivers to code so they could participate in the AI revolution. I used to think this was a weird straw man occasionally trotted out by Freddie deBoer, but all these top economists were super enthusiastic about old white guys whose mill has fallen on hard times founding the next generation of nimble tech startups. I’m tempted to mock this, but maybe I shouldn’t – this From Coal To Code article says that the program has successfully rehabilitated Kentucky coal miners into Web developers. And I can’t think of a good argument why not – even from a biodeterminist perspective, nobody’s ever found that coal mining areas have lower IQ than anywhere else, so some of them ought to be potential web developers just like everywhere else. I still wanted to ask the panel “Given that 30-50% of kids fail high school algebra, how do you expect them to learn computer science?”, but by the time I had finished finding that statistic they had moved on to a different topic.

Computer programming is about context-free symbol manipulation. (There are probably other jobs that are like this too.) In order to get a machine to do something, you have to be able to simulate that machine in your head. I don’t mean in some super low-level way; I just mean that you need to be able to run a program in your head to see if it does what you intend it to do (or, as is often the case, see why it’s doing something unintended). Machines don’t do what you want them to do. They do what you tell them to do.

This is not a trivial barrier to entry. Having TA’d programming classes before, I can confidently say that some people do not get this at all. When someone would ask what a particular program did, many students would attempt to divine the intent of the writer of the program. That is, they would answer what they thought the program was supposed to do (presumably based on the comments and variable names). This is not a useless skill. It gives you the right answers when a program is correct. As a working developer, it’s useful when trying to reverse engineer requirements from existing code (that may or may not actually fulfill those requirements). It’s not very useful in debugging though. And that’s mostly what programming is–taking incomplete or buggy attempts at solving a problem and refining them into a working solution.

In the class itself, people who didn’t really get it could skate along for a while, since they were mostly looking at code that worked. Eventually they would crash and burn, though, as the complexity of the assignments would outstrip their ability to make a working program on the first attempt. This is shitty from a pedagogical perspective, so a lot of my review sessions focused on this skill of simulating programs. One type of exercise that worked really well (both for breaking through this conceptual barrier as well as providing good practice) was to generate nonsense programs and then ask what they did. There’s really no way around simulating a program when the variable names are all single letters and the code snippet does not actually have any intelligible intent. Some people would eventually get it. I’d like to think that with enough time and effort that everyone could get it, but that’s probably not realistic even when we’re just talking about college students at selective institutions.

So, whenever I read something about the loss of well-paying blue-collar jobs that talks about retraining and the tech industry, I feel baffled that they don’t consider the obvious objections. Actually, scratch that. I feel insulted!

collapsedsquid

If you think computing jobs are the future, then what you should probably do is train these people in QA.  Better choice for everyone involved.

stumpyjoepete

See, this is just depressing enough to actually be realistic.

Source: therainstheyaredropping programming ai economics
Top Photos