Unless you think that historians work in the same way as priests, you know that historical knowledge is also supposed to be corroborated by evidence, and so on. History is not a because-I-say-so discipline, at all.
Creationist believe that a single supernatural Omnipotent being creating everything is a simpler than everything happening through happenstance.
But that is only due to their poor imagination at asking questions... Once you've reduced everything to a single supernatural Omnipotent, you are left with the huge problem of understanding *that*, and the last few thousand years have shown that it is extraordinarily difficult to make sense of it.
Your question implied that the only alternative to teaching $X to ALL was to teach them to NONE.
But I was responding to "Not everyone is interested in programming, or any sort of engineering. Get over it. Forcing every kid to take programming is like forcing every kid to learn how to build an engine for their car" and that comment only makes sense if what he had in mind was mandatorily teaching programming to all kids, in a non-elective class, for otherwise there would be no forcing... So the all or none dichotomy was implied in the comment I replied to.
Well... there is not one factor to explain everything. Just read up on history of politics and economics. It's a long, complicated, series of interrelated factors all over the place.
If a programmer is using a file API with POSIX semantics in any non-trivial way, and is not aware of the fact that POSIX does not specify any assurances that data will be written to the device unless fsync is called or another similar action is taken, then that programmer is *not* competent.
This is not a minor detail. POSIX file system semantics have *never* implied that writing to a file includes that the data is actually on the device. Are you saying that API's should be resilient to developers using them under whatever unfounded expectations they may have about them?
No, not really. Journalling is done so that after a crash the filesystem is in a consistent state, and that does *not* include the no-data-loss requirement you are talking about.
As an application developer, the last thing I want to worry about is whether or not the fraking filesystem is going to persist my data to disk.
As an application developer, you are expected to know what the API does, in order to use it correctly. What Ext4 is doing is 100% respectful of the spec.
I was responding to a post which started with: "Not everyone is interested in programming, or any sort of engineering" and proceeded to use that as an argument to support not making programming a mandatory subject: I observed that just as well the claim can be made that "Not everyone is interested in $X" for all values of X, including programming, and I asked if he thought that his argument also extended to justify that kids not be taught any such X.
I cannot see *what* possible connection your comment has on anything. And I am a professional mathematician: I am quite aware of how negating a universal quantifier works, by the way...
challenge! I also write in assembly and occasionally microcode.
Writing in lower level languages has certain advantages. For example, the optimal C++ code for doing something, when compared to assembly, is slower, a LOT larger, and requires more memory.
In 99.999999999% of the cases, those extra advantages you speak of are simply imaginary, more related to the rush of appearing to be among the mythical REAL PROGRAMMERS than to any requirements of real life.
Bash isn't a proper low level language because it's interpreted, but it does keep a polish on one's skills to build up complex things from simple things. You can always do a complex task using simple tools, but you cannot always do a simple task using complex tools. (they are not general enough to do everything you could possibly want to do)
This is not a matter of simple or complex tools, but of correct or incorrect tools. If when presented with the problem of doing boring filtering and manipulations with server logs the first thing you can think of is writing an assembler program to handle it, in the immense majority of cases you are making a serious mistake.
There is absolutely no need to teach about bits and bytes: they are only incidental in actual programming. Programming is reasoning about actions, evaluation and transformation. HTML and CSS simply do not help with that.
There is no need to fully parse natural languages (or to substitute them with made up languages you can parse...) in order to answer questions posed in natural languages. Indeed, one does not need to *understand* a question (in whatever AI meaning you want) in order to find its answer.
Standards at their best are forward looking and interesting because they are stated without much thought as to how they would actually be implemented and part of the problem is figuring out how too implement them.
Slashdot Mirror
Unless you think that historians work in the same way as priests, you know that historical knowledge is also supposed to be corroborated by evidence, and so on. History is not a because-I-say-so discipline, at all.
Can you name a piece of knowledge acquired by any other method?
Why are you repeating this over and over?
Creationist believe that a single supernatural Omnipotent being creating everything is a simpler than everything happening through happenstance.
But that is only due to their poor imagination at asking questions... Once you've reduced everything to a single supernatural Omnipotent, you are left with the huge problem of understanding *that*, and the last few thousand years have shown that it is extraordinarily difficult to make sense of it.
If Creationism can come up with something that can be tested with the scientific method, who are you to decide it's inherently wrong?
When that happens, get back to us...
Your question implied that the only alternative to teaching $X to ALL was to teach them to NONE.
But I was responding to "Not everyone is interested in programming, or any sort of engineering. Get over it. Forcing every kid to take programming is like forcing every kid to learn how to build an engine for their car" and that comment only makes sense if what he had in mind was mandatorily teaching programming to all kids, in a non-elective class, for otherwise there would be no forcing... So the all or none dichotomy was implied in the comment I replied to.
That, without knowing what their lives were like, conveys zero information.
Well... there is not one factor to explain everything. Just read up on history of politics and economics. It's a long, complicated, series of interrelated factors all over the place.
If a programmer is using a file API with POSIX semantics in any non-trivial way, and is not aware of the fact that POSIX does not specify any assurances that data will be written to the device unless fsync is called or another similar action is taken, then that programmer is *not* competent.
This is not a minor detail. POSIX file system semantics have *never* implied that writing to a file includes that the data is actually on the device. Are you saying that API's should be resilient to developers using them under whatever unfounded expectations they may have about them?
No, not really. Journalling is done so that after a crash the filesystem is in a consistent state, and that does *not* include the no-data-loss requirement you are talking about.
As an application developer, the last thing I want to worry about is whether or not the fraking filesystem is going to persist my data to disk.
As an application developer, you are expected to know what the API does, in order to use it correctly. What Ext4 is doing is 100% respectful of the spec.
Huh?!
I was responding to a post which started with: "Not everyone is interested in programming, or any sort of engineering" and proceeded to use that as an argument to support not making programming a mandatory subject: I observed that just as well the claim can be made that "Not everyone is interested in $X" for all values of X, including programming, and I asked if he thought that his argument also extended to justify that kids not be taught any such X.
I cannot see *what* possible connection your comment has on anything. And I am a professional mathematician: I am quite aware of how negating a universal quantifier works, by the way...
challenge! I also write in assembly and occasionally microcode.
Writing in lower level languages has certain advantages. For example, the optimal C++ code for doing something, when compared to assembly, is slower, a LOT larger, and requires more memory.
In 99.999999999% of the cases, those extra advantages you speak of are simply imaginary, more related to the rush of appearing to be among the mythical REAL PROGRAMMERS than to any requirements of real life.
Bash isn't a proper low level language because it's interpreted, but it does keep a polish on one's skills to build up complex things from simple things. You can always do a complex task using simple tools, but you cannot always do a simple task using complex tools. (they are not general enough to do everything you could possibly want to do)
This is not a matter of simple or complex tools, but of correct or incorrect tools. If when presented with the problem of doing boring filtering and manipulations with server logs the first thing you can think of is writing an assembler program to handle it, in the immense majority of cases you are making a serious mistake.
There is absolutely no need to teach about bits and bytes: they are only incidental in actual programming. Programming is reasoning about actions, evaluation and transformation. HTML and CSS simply do not help with that.
Actually, "Not everyone is interested in $X" is a true statement for all values of X. Are you proposing not to teach kids anything?
Who came up with the requirement of using bash to do anything?
HTML and CSS do not teach you *anything* about how a computer works. At all.
What's with the underscores? You can write ass, you know. You are not a bit more "polite" by only implying it: you only show your sense of guilt...
And how long does it take you you restart all apps that were running and put them in the state they were?
How do you know that Stephen Wolfram's last name comes from wolframite?
There is no need to fully parse natural languages (or to substitute them with made up languages you can parse...) in order to answer questions posed in natural languages. Indeed, one does not need to *understand* a question (in whatever AI meaning you want) in order to find its answer.
Standards at their best are forward looking and interesting because they are stated without much thought as to how they would actually be implemented and part of the problem is figuring out how too implement them.
That's absolutely false, of any good standard.
Something is wrong with your reading comprehension.
googling for the coreutils source repo, searching for rm.c and reading it would have taken less time...