You might want to see other replies—the CompE thing has already been discussed. (My experience is that pure CS majors are even worse software developers than computer engineers.)
But, yes, I am being that pedantic, and furthermore, you should be, too. Why? Because "CS = theory" still shapes the curriculum and the educational goals of professors at universities. The only institutions I've seen that offered dedicated software development programs (not merely a stream within a theoretical framework) are community colleges. That's why people in business complain about unprepared CS grads; because while they've been heavily educated in the mathematics of computer science, the mentality found in a CS department is not what real-world software development expects.
I'm bringing all of this up not because I don't value software engineering and development, but because I think it exists in the shadow of theoretical CS. It should be given its due as a unique engineering program, but few if any institutions I know of actually do that.
I also said it was significant that they were popular—but the inaccessible prototypes need to be put on the highest pedestal. At the same time that products are entering the market and revolutionizing how the world does business and sees computers, new ideas are being found and created that go to power those changes. Without the ideas, the products would not exist; the same cannot be said in reverse. In every other field, priority and recognition is given to those who invent things first, followed by the people who popularized them. (Otherwise, people wouldn't foam at the mouth when it is claimed that Edison invented the lightbulb.) There is no reason why innovations in software should be treated differently.
To short-circuit the conversation: if your university's CS department considered that to be true, you probably would have had that course available to you. Software development just gets lumped in with CS because there's not enough demand to make a separate engineering department. This isn't something that became obvious to me until grad school.
Xerox Bravo (1974), Xerox Gypsy (1975), and Xerox Markup (not sure of exact year, in the vicinity). As a general rule, whatever you can think of, PARC had it ten years earlier. By the late eighties they were working on a PDA/tablet/smart surface, touch-driven ecosystem.
Point being—people disproportionate weight on programs that they experienced. It's the same story whenever an amateur writes a computer history article; a few pages of nostalgic bullshit without any real research. Yes, it's significant that the Mac programs (which, oh by the way, already existed on the Lisa, too!) were popular, but severely erroneous to give them all the scrutiny. As historians we should endeavour to look past our own biases and provide an accurate image of history, not play favourites with specific products.
The second point, okay. In my experience a lot of CompE students end up taking a large number of software dev courses, and are required to be competent C++ developers to graduate with honours. I guess that's a regional thing. Certainly engineering schools should offer a software stream, though.
And c'mon, product history? That's a community college thing. The words "computer science" literally indicate a science of computations; I know there was an argument about how it should be defined here a few months ago, but even accepting the applied component into the definition (i.e. the theory and practice of what constitutes good software development, like HCI and project management), that's several steps removed from trivia about various specific products and what they did to revolutionize other industries. The author's whole line of reasoning is ignorant of CS is, and why it's associated with the arts and sciences rather than an engineering department.
Actually, Wolf was preceded by HoverTank 3-D and Catacombs 3-D, both of which were made by the id Software team. Catacombs 3-D would best be classified as the direct progenitor of Wolfenstein 3-D; it had all of the core game features except multiple guns (you picked up different magic missile patterns instead). It even had powerups, including one that froze all of the enemies for a set period of time, a feature not often found in FPSes.
That being said, though, these were all circa 1991. Zork and Colossal Cave Adventure are from 1979 and 1976 respectively. When you realise that there were programs like MacPaint and MacWrite a full decade before the Mac came out, it really makes TFA sound bratty.
Which, it should be emphasized, we do study. While I'm a major advocate for the study of computer history, CS is not about software development, it is a branch of mathematics. The author of the article would be better off pestering computer engineers.
Social engineering? I'm pretty sure some part of "hiding a scraping box in a closet" involves playing on expectations. Good luck beating that in a large organization without DHS-grade paranoia.
They did—it was called the Crusades. Of course, that was before the fall of Constantinople, but it was the same idea. Every now and then things get interesting (the British, so moved by ancient literature, helped during the Greek war of independence, for example) but for the most part it's been mudslinging and mutual abuse.
As you can see, it doesn't take long to assess the miserable state of Greek-Middle Eastern relations during the last three thousand years. It's been a very, very, very long time since the regions were on good terms with each other. If that ever even happened.
'Many eyes' is a statistically valid principle, just over-trusted. You're right that it's not a guarantee that bugs will be found, understood, and fixed more quickly as staff are added, but as long as developers (and testers) aren't slacking off due to herd mentality effects, the rate of finding bugs cannot be any worse than it is with fewer people. It's a submodular function.
...also, if you have an infinite number of programmers reviewing the code at the same time, however, it is certain that all bugs will be diagnosed and fixed. Possibly instantaneously. So that's a theoretically nice result.
The second article makes it clear that this is specifically about mobile networks, so yes. It's a quintessential net neutrality issue. The user is already paying for the data.
I'm talking about a slightly higher level of conversation structure. The features you're describing are certainly significant in how they affect spoken language (modern French is very drastically different from written French, BTW; it's very weird), but are less likely to predominate in written text like Facebook posts. This is especially the case for other samples studied by the researchers, like comments left on CNN articles. Professionally-written text is just generally well-organized; concepts and events are introduced in an efficient manner, using consistent and correct word choice. Even if the sentence structure is more familiar, when people blunder through recounting an event, we have to do more work to reconstruct what they're saying. Professional writing is composed with the benefit of hindsight and more thoughtful analysis.
...however, the researchers believe that people are just natural gossips. With that in mind, it could easily be about the density of opinions and moods in the text that makes the snippets easier to remember; the emotions of the author provide another anchor to build an associative memory around.
Then shouldn't Orange have a contract with that 'other provider' to act as either (a) a customer or (b) a push-pull arrangement, like big top-tier American companies do?
There's another force working on the opposite direction that favours casual banter written by others: published text is often heavily massaged to use idiomatic language that fits in familiar patterns. The lack of novelty in the writing and the lack of effort required to read it makes it stick out less. As a general rule, you'll remember things better when you spend more effort in understanding them.
All programs should be exactly 100 lines long. No more, no less. If your program is fewer lines than that, you're using too high-level a language. If it's longer, you're using too low-level a language.
Slashdot Mirror
Yeah, well... in APL, that incomprehensible one-liner would be about three characters!
You might want to see other replies—the CompE thing has already been discussed. (My experience is that pure CS majors are even worse software developers than computer engineers.)
But, yes, I am being that pedantic, and furthermore, you should be, too. Why? Because "CS = theory" still shapes the curriculum and the educational goals of professors at universities. The only institutions I've seen that offered dedicated software development programs (not merely a stream within a theoretical framework) are community colleges. That's why people in business complain about unprepared CS grads; because while they've been heavily educated in the mathematics of computer science, the mentality found in a CS department is not what real-world software development expects.
I'm bringing all of this up not because I don't value software engineering and development, but because I think it exists in the shadow of theoretical CS. It should be given its due as a unique engineering program, but few if any institutions I know of actually do that.
I also said it was significant that they were popular—but the inaccessible prototypes need to be put on the highest pedestal. At the same time that products are entering the market and revolutionizing how the world does business and sees computers, new ideas are being found and created that go to power those changes. Without the ideas, the products would not exist; the same cannot be said in reverse. In every other field, priority and recognition is given to those who invent things first, followed by the people who popularized them. (Otherwise, people wouldn't foam at the mouth when it is claimed that Edison invented the lightbulb.) There is no reason why innovations in software should be treated differently.
To short-circuit the conversation: if your university's CS department considered that to be true, you probably would have had that course available to you. Software development just gets lumped in with CS because there's not enough demand to make a separate engineering department. This isn't something that became obvious to me until grad school.
Xerox Bravo (1974), Xerox Gypsy (1975), and Xerox Markup (not sure of exact year, in the vicinity). As a general rule, whatever you can think of, PARC had it ten years earlier. By the late eighties they were working on a PDA/tablet/smart surface, touch-driven ecosystem.
Point being—people disproportionate weight on programs that they experienced. It's the same story whenever an amateur writes a computer history article; a few pages of nostalgic bullshit without any real research. Yes, it's significant that the Mac programs (which, oh by the way, already existed on the Lisa, too!) were popular, but severely erroneous to give them all the scrutiny. As historians we should endeavour to look past our own biases and provide an accurate image of history, not play favourites with specific products.
The second point, okay. In my experience a lot of CompE students end up taking a large number of software dev courses, and are required to be competent C++ developers to graduate with honours. I guess that's a regional thing. Certainly engineering schools should offer a software stream, though.
And c'mon, product history? That's a community college thing. The words "computer science" literally indicate a science of computations; I know there was an argument about how it should be defined here a few months ago, but even accepting the applied component into the definition (i.e. the theory and practice of what constitutes good software development, like HCI and project management), that's several steps removed from trivia about various specific products and what they did to revolutionize other industries. The author's whole line of reasoning is ignorant of CS is, and why it's associated with the arts and sciences rather than an engineering department.
Actually, Wolf was preceded by HoverTank 3-D and Catacombs 3-D, both of which were made by the id Software team. Catacombs 3-D would best be classified as the direct progenitor of Wolfenstein 3-D; it had all of the core game features except multiple guns (you picked up different magic missile patterns instead). It even had powerups, including one that froze all of the enemies for a set period of time, a feature not often found in FPSes.
That being said, though, these were all circa 1991. Zork and Colossal Cave Adventure are from 1979 and 1976 respectively. When you realise that there were programs like MacPaint and MacWrite a full decade before the Mac came out, it really makes TFA sound bratty.
Wolf 3D, Blake Stone (by a week), ShadowCaster, Catacombs 3D... Close, but no cigar. :)
Which, it should be emphasized, we do study. While I'm a major advocate for the study of computer history, CS is not about software development, it is a branch of mathematics. The author of the article would be better off pestering computer engineers.
Social engineering? I'm pretty sure some part of "hiding a scraping box in a closet" involves playing on expectations. Good luck beating that in a large organization without DHS-grade paranoia.
They did—it was called the Crusades. Of course, that was before the fall of Constantinople, but it was the same idea. Every now and then things get interesting (the British, so moved by ancient literature, helped during the Greek war of independence, for example) but for the most part it's been mudslinging and mutual abuse.
As you can see, it doesn't take long to assess the miserable state of Greek-Middle Eastern relations during the last three thousand years. It's been a very, very, very long time since the regions were on good terms with each other. If that ever even happened.
"Bestest."
'Many eyes' is a statistically valid principle, just over-trusted. You're right that it's not a guarantee that bugs will be found, understood, and fixed more quickly as staff are added, but as long as developers (and testers) aren't slacking off due to herd mentality effects, the rate of finding bugs cannot be any worse than it is with fewer people. It's a submodular function.
...also, if you have an infinite number of programmers reviewing the code at the same time, however, it is certain that all bugs will be diagnosed and fixed. Possibly instantaneously. So that's a theoretically nice result.
In other other news, Google penalizes the rankings of spelling and grammatical errors. Cynically, I'm surprised this headline got posted.
Works for me.
The second article makes it clear that this is specifically about mobile networks, so yes. It's a quintessential net neutrality issue. The user is already paying for the data.
I'm talking about a slightly higher level of conversation structure. The features you're describing are certainly significant in how they affect spoken language (modern French is very drastically different from written French, BTW; it's very weird), but are less likely to predominate in written text like Facebook posts. This is especially the case for other samples studied by the researchers, like comments left on CNN articles. Professionally-written text is just generally well-organized; concepts and events are introduced in an efficient manner, using consistent and correct word choice. Even if the sentence structure is more familiar, when people blunder through recounting an event, we have to do more work to reconstruct what they're saying. Professional writing is composed with the benefit of hindsight and more thoughtful analysis.
...however, the researchers believe that people are just natural gossips. With that in mind, it could easily be about the density of opinions and moods in the text that makes the snippets easier to remember; the emotions of the author provide another anchor to build an associative memory around.
Then shouldn't Orange have a contract with that 'other provider' to act as either (a) a customer or (b) a push-pull arrangement, like big top-tier American companies do?
Pretty much. Awkward, unpolished language sticks in your mind. In fact, a lot of memes are that way.
There's another force working on the opposite direction that favours casual banter written by others: published text is often heavily massaged to use idiomatic language that fits in familiar patterns. The lack of novelty in the writing and the lack of effort required to read it makes it stick out less. As a general rule, you'll remember things better when you spend more effort in understanding them.
Isn't there a free implementation of that in Emacs?
All programs should be exactly 100 lines long. No more, no less. If your program is fewer lines than that, you're using too high-level a language. If it's longer, you're using too low-level a language.
Granted, it takes a bit more imagination, but that happens too.
Queue is acceptable as a synonym for cue if there are many people waiting to make the same snide remark.
Carbon. He's going carbon.
OR: we switched to a case-insensitive universe while you weren't looking. Something something no one ever got fired for buying Microsoft.