And another AC that does not even have a basic understanding of what is going on. Talk about Dunning-Kruger. The fact of the matter is that a modern industrial country at this time needs significantly more talented people that its population can supply. Thi sis not a question of "training". If the potential is not there, no amount of training will fix that.
The US and UK used to understand that, and they had a hiring advantage because many people globally learn English. But in many European countries you now run into pretty top-notch IT teams that also have English as a team language, because they come from all over the globe and hence that advantage is lost. Now, making immigration difficult will give you even less of that limited pool and that will long-term kill any competitive advantage you have. May take 10 or 20 years to become obvious, but what we see is an act of self-destruction, fueled by arrogance, stupidity and xenophobia. Of course, the UK is also going into Fascism at this time (apparently they do not believe the experiences others made with it and what to try it out themselves), so it will be hard to divide the reasons for coming economic catastrophe there apart afterwards.
Actually, they are not "sloppy" and "lazy". They are the cheapest "coders" the MBA-morons in charge could find. They could do a better job if their life depended on it. Alternatively, coders that do have it and can do it (a minority) are not given enough time to clean up and fix remaining issues, because said MBA-morons think "it works". I have learned to not give them anything that has the complete functionality before all other aspects are fine. Otherwise they declare the prototype "ready for production" and that is not good at all.
It serves to establish and maintain closer relationships between users and IT security people, so that, you know, if a user has a suspicion of a security problem, they feel more confident and approach IT security staff earlier. But that idea flays wayyyyy above the heads of MBA morons.
A risk is something that has an element of potentiality. In the case of climate change, the catastrophe is already assured. The only question is whether it will be severe, very severe or "collapse of civilization"-level. Calling a "risk" is, once again, making it sound a lot more harmless than it is.
For "most" computing tasks, it may appear so but rarely is so (performance also includes memory footprint and network footprint). But this is an invalid measure. The question changes completely when you start assessing by importance. And there, you suddenly find a significant number of really important things that cannot be done in things other than C.
Of course, most "learn to code" victims will just become technician-level coders and they basically have their time wasted. Better to become a carpenter, plumber or the like, then at least you have some real skills.
Do you not know that you're being something of a jerk when saying this, or not care?
If you think an invalid "argument" AdHominem makes a case for your POV, go right ahead. I do care very much because I think it is wrong to give people false hopes. What I do see is that most people are unable to learn to code well, regardless of education effort and experience. These will never get decent or secure jobs with a primary qualification of "coder". They may well get really bad jobs for a few years and then none at all. I think it is deeply wrong to do this to them.
Indeed. The Snoopers Charter is extremely dangerous to individual freedoms of ordinary citizens, but it will do absolutely nothing to reduce terrorism. It may encourage terrorism though, because the terrorists must think they are winning.
While I still tend towards "incompetence" and not "intent", it is getting harder. Obviously, May does regard this as an excellent opportunity to push stronger for her anti-freedom agenda (which will do exactly nothing to curb terrorism, but may encourage it).
And as an aside, I think you disrespect junior/novice programmers unduly. There's nothing wrong with being green, or even just being a hobbyist programmer. Where else are our greybeards going to come from? And don't we want other people to become involved in this activity that we enjoy? (We do enjoy programming, right? That's why we do this, right? I'm sure I'm debugging this misbegotten legacy junk because I enjoy it...) Anyway, I think it's worthwhile just for more people to learn to think programmatically.
I don't. When I say "technician-level coders", I do not mean novices. I mean people that sometimes have decades of experience. It may have contributed that a few hours before I wrote this I had to (again) explain to some web-application "developers" at a customer how their application works and how a HTTP request works. And then I had to do it again for a second team. And then I had to explain how a HTTP/302 works. All to supposed domain experts with around (my estimation) 10 years of relevant experience. These people have no clue how the technology they use works and it causes real problems, sometimes pretty bad ones.
Of course, most "learn to code" victims will just become technician-level coders and they basically have their time wasted. Better to become a carpenter, plumber or the like, then at least you have some real skills.
Incidentally, Go or Rust will never replace C, unless we get so much computing power that speed does not matter. As we seem to have hit a plateau a few years back in per-core speeds, that seems quite unlike to happen. When you need performance and control, nothing else will serve except assembler, and that is one order of magnitude more difficult.
A scientist is someone who uses the tool of science, who thinks like a scientist. You don't need a degree to be a scientist.
About the same as you can be a brain surgeon without the relevant degree. In theory possible, in practice it does basically never happen. In fact, you need that degree and quite a bit more to be a scientist.
And yes, she clearly turned off whatever science brain she has while writing that post.
If she ever had one. For a BA you can fake it. For an MA, you can still fake it, just takes more effort and potentially some help.
For both there are options to do without. They are not cost-effective at the moment, because in both trains and planes the engineer/pilot is a lot less of a cost factor than in a car or truck.
Undergrad degrees at best make you a technician, not a scientist. For that you need a PhD or equivalent. So she is a scientist in education science (apparently the dumbest and most non-scientific discipline imaginable) and going this career path rather strongly suggests something is wrong with her Math and CS skills.
Rutkowska is a scientist, not an SJW. That she is a woman matters very little for her work, which is beyond reproach. There are quite a few women like that in the sciences, and these do not lament any "imbalance" or fantasize that "things need to be cleaned up". In fact, some have a pretty dim view of the woman that do.
Not 100%, but likely the most meritocratic part of society by far. That also means those that have what it takes can get a second and even third chance. There is no "act" to "clean up". Those that fail do lack what it takes and need to fail in order to keep science effective. This is not some social club where merit does not matter. This is the future of humankind.
Inertia is surely a factor even so. There is no technical reason you couldn't have a language that retained the low-level control and flexibility of C while still being significantly safer, having much cleaner syntax, and including useful features that made the language more powerful but that could be implemented with no runtime overhead.
I used to think that as well. I do not anymore, and I think there _are_ technical reasons for this. For one thing, C is exceptionally well defined and has a very clean syntax (C++ does not qualify for either though). You can sees this by how little the core language has changed. Of course, compilers have gotten better at finding some classes of mistakes, and I would really not want to use an original K&R compiler. But these changes, useful as they are, have not changed much in the core language. That is one indicator that there may not be much that can be done.
The other is just my intuition for having worked with a lot of languages (functional and logic only a few, admittedly) and whenever somebody adds "safety" or "cleaner syntax", they actually make it more complex and by KISS actually make it worse overall. An extreme case of that is Java, where you often have much more non-functional code (declarations) than functional code and much practically produced code is basically unreadable without use of tools. That is not a good state of affairs at all and basically violates all cleanliness, safety, simplicity and turns the language into a design failure. (The "advantage" of Java is libraries for everything, allowing even the most incompetent coder to produce something...)
Bottom line is that I have seen these attempts to improve on C fail time and again, and it may just be time to admit that C is pretty optimal as a language, provided the coder has what it takes to understand the execution model. The execution model is also pretty simple, but requires understanding how a CPU and RAM actually works and many coders are failing at that these days. The execution models of most languages are much more complex than the one of C, but you can produce some code not understanding them. That is also a reason why some people consider C coders to be superior, because incompetent coders cannot produce anything at all in C, while they often can produce somewhat working pretty bad code in other languages.
As to average coder competence, I think the best take on it is still this one: https://blog.codinghorror.com/... You cannot by any means turn these people into good coders and language can only help to make their incompetence obvious. C does that well, because of the requirement that you understand the execution model.
But even if you did invent such a language, you'd still have to convince developers working on these kinds of applications to use that new language in preference to the global lingua franca of programming they all know already, and you'd still have to provide a way to link code written in your new language with the vast ecosystem of existing C-based libraries and APIs. That's a tall order, no matter how brilliant your new language itself might be.
Assuming such a language could be created, you are certainly right about this.
Of course. I might be blind as well, or we may just have different approaches to the same problem space, equally valid, but mine is far more language agnostic than yours. Whether that is better or not depends, as usual.
Being blind to more abstract views and approaches is not a question of experience and age. Some people have it, some do not. I have seen it countless times. The point is that most people cannot advance enough. You may just be in that group, sorry.
You think? You cannot have much experience in any of them (or with coding in general), if you think that similarity in syntax makes for similarity in language.
I fully agree. The thing is, to borrow terminology from other technological disciplines, most programmers these days are mere technicians and usually bad ones. You are describing what a good engineer does. Bad technicians will never be paid well, never have good job prospects and often get replaced long-term entirely. Good technicians (which is a small subset of them) and good engineers do not have that problem. Good technicians stay on to maintain the tech they know forever and learn a bit in addition over time. Good engineers learn new stuff constantly and over time build up an impressive array of tools they can wield competently.
That is why 90% of all people writing software are not in any good position job-wise and prospects-wise: They are just not good at it. The rest has a pretty golden future.
On the server, Linux is replacing Solaris and other "old" Unixes. On the client, it may become pretty popular, because some large enterprises are preparing to go web-application only when Win7 becomes non-viable. They will not do this on Windows. All your comment shows is that you have no clue what is going on.
Ultimately, the big problem is that I don't see a real universal contender for high-performance native code taking over from C/C++. There are a lot of promising languages, but at the moment, nothing is really taking off. Simple inertia is pretty hard to overcome, as it turns out.
I do not see this being inertia at all. The aspect that gives you speed in C (and less so in C++) is control. You decide about memory management, details of calculations, etc. You decide which safety-features can be left out. You do this understanding the problem you solve. No compiler will ever be able to come close, unless we get strong AI and that is looking very unlikely today. Hence all those "promising" languages will not be able to compete either. They will just allow more bad coders to produce more bad code because a whole army of stupid "managers" does not understand that tools cannot make bad coders turn out good code.
Actually, it is C that is king. In C++ it is just to hard to write really efficient code. C++ is for all those things were efficiency is relevant, but not so critical. Of course, many C++ programs are just C with some extensions and never really to any significant OO.
Slashdot Mirror
Apple is making excellent business. And it does so with high-level government support. Maybe you voted for the wrong people?
And another AC that does not even have a basic understanding of what is going on. Talk about Dunning-Kruger. The fact of the matter is that a modern industrial country at this time needs significantly more talented people that its population can supply. Thi sis not a question of "training". If the potential is not there, no amount of training will fix that.
The US and UK used to understand that, and they had a hiring advantage because many people globally learn English. But in many European countries you now run into pretty top-notch IT teams that also have English as a team language, because they come from all over the globe and hence that advantage is lost. Now, making immigration difficult will give you even less of that limited pool and that will long-term kill any competitive advantage you have. May take 10 or 20 years to become obvious, but what we see is an act of self-destruction, fueled by arrogance, stupidity and xenophobia. Of course, the UK is also going into Fascism at this time (apparently they do not believe the experiences others made with it and what to try it out themselves), so it will be hard to divide the reasons for coming economic catastrophe there apart afterwards.
Actually, they are not "sloppy" and "lazy". They are the cheapest "coders" the MBA-morons in charge could find. They could do a better job if their life depended on it. Alternatively, coders that do have it and can do it (a minority) are not given enough time to clean up and fix remaining issues, because said MBA-morons think "it works". I have learned to not give them anything that has the complete functionality before all other aspects are fine. Otherwise they declare the prototype "ready for production" and that is not good at all.
It serves to establish and maintain closer relationships between users and IT security people, so that, you know, if a user has a suspicion of a security problem, they feel more confident and approach IT security staff earlier. But that idea flays wayyyyy above the heads of MBA morons.
A risk is something that has an element of potentiality. In the case of climate change, the catastrophe is already assured. The only question is whether it will be severe, very severe or "collapse of civilization"-level. Calling a "risk" is, once again, making it sound a lot more harmless than it is.
For "most" computing tasks, it may appear so but rarely is so (performance also includes memory footprint and network footprint). But this is an invalid measure. The question changes completely when you start assessing by importance. And there, you suddenly find a significant number of really important things that cannot be done in things other than C.
Of course, most "learn to code" victims will just become technician-level coders and they basically have their time wasted. Better to become a carpenter, plumber or the like, then at least you have some real skills.
Do you not know that you're being something of a jerk when saying this, or not care?
If you think an invalid "argument" AdHominem makes a case for your POV, go right ahead. I do care very much because I think it is wrong to give people false hopes. What I do see is that most people are unable to learn to code well, regardless of education effort and experience. These will never get decent or secure jobs with a primary qualification of "coder". They may well get really bad jobs for a few years and then none at all. I think it is deeply wrong to do this to them.
Indeed. The Snoopers Charter is extremely dangerous to individual freedoms of ordinary citizens, but it will do absolutely nothing to reduce terrorism. It may encourage terrorism though, because the terrorists must think they are winning.
While I still tend towards "incompetence" and not "intent", it is getting harder. Obviously, May does regard this as an excellent opportunity to push stronger for her anti-freedom agenda (which will do exactly nothing to curb terrorism, but may encourage it).
And as an aside, I think you disrespect junior/novice programmers unduly. There's nothing wrong with being green, or even just being a hobbyist programmer. Where else are our greybeards going to come from? And don't we want other people to become involved in this activity that we enjoy? (We do enjoy programming, right? That's why we do this, right? I'm sure I'm debugging this misbegotten legacy junk because I enjoy it...) Anyway, I think it's worthwhile just for more people to learn to think programmatically.
I don't. When I say "technician-level coders", I do not mean novices. I mean people that sometimes have decades of experience. It may have contributed that a few hours before I wrote this I had to (again) explain to some web-application "developers" at a customer how their application works and how a HTTP request works. And then I had to do it again for a second team. And then I had to explain how a HTTP/302 works. All to supposed domain experts with around (my estimation) 10 years of relevant experience. These people have no clue how the technology they use works and it causes real problems, sometimes pretty bad ones.
Of course, most "learn to code" victims will just become technician-level coders and they basically have their time wasted. Better to become a carpenter, plumber or the like, then at least you have some real skills.
Incidentally, Go or Rust will never replace C, unless we get so much computing power that speed does not matter. As we seem to have hit a plateau a few years back in per-core speeds, that seems quite unlike to happen. When you need performance and control, nothing else will serve except assembler, and that is one order of magnitude more difficult.
A scientist is someone who uses the tool of science, who thinks like a scientist. You don't need a degree to be a scientist.
About the same as you can be a brain surgeon without the relevant degree. In theory possible, in practice it does basically never happen. In fact, you need that degree and quite a bit more to be a scientist.
And yes, she clearly turned off whatever science brain she has while writing that post.
If she ever had one. For a BA you can fake it. For an MA, you can still fake it, just takes more effort and potentially some help.
For both there are options to do without. They are not cost-effective at the moment, because in both trains and planes the engineer/pilot is a lot less of a cost factor than in a car or truck.
Undergrad degrees at best make you a technician, not a scientist. For that you need a PhD or equivalent. So she is a scientist in education science (apparently the dumbest and most non-scientific discipline imaginable) and going this career path rather strongly suggests something is wrong with her Math and CS skills.
Rutkowska is a scientist, not an SJW. That she is a woman matters very little for her work, which is beyond reproach. There are quite a few women like that in the sciences, and these do not lament any "imbalance" or fantasize that "things need to be cleaned up". In fact, some have a pretty dim view of the woman that do.
Yeah, rampart and pretty repulsive sexism. It is not worthwhile to even listen to these people, they have nothing to contribute.
Not 100%, but likely the most meritocratic part of society by far. That also means those that have what it takes can get a second and even third chance. There is no "act" to "clean up". Those that fail do lack what it takes and need to fail in order to keep science effective. This is not some social club where merit does not matter. This is the future of humankind.
Inertia is surely a factor even so. There is no technical reason you couldn't have a language that retained the low-level control and flexibility of C while still being significantly safer, having much cleaner syntax, and including useful features that made the language more powerful but that could be implemented with no runtime overhead.
I used to think that as well. I do not anymore, and I think there _are_ technical reasons for this. For one thing, C is exceptionally well defined and has a very clean syntax (C++ does not qualify for either though). You can sees this by how little the core language has changed. Of course, compilers have gotten better at finding some classes of mistakes, and I would really not want to use an original K&R compiler. But these changes, useful as they are, have not changed much in the core language. That is one indicator that there may not be much that can be done.
The other is just my intuition for having worked with a lot of languages (functional and logic only a few, admittedly) and whenever somebody adds "safety" or "cleaner syntax", they actually make it more complex and by KISS actually make it worse overall. An extreme case of that is Java, where you often have much more non-functional code (declarations) than functional code and much practically produced code is basically unreadable without use of tools. That is not a good state of affairs at all and basically violates all cleanliness, safety, simplicity and turns the language into a design failure. (The "advantage" of Java is libraries for everything, allowing even the most incompetent coder to produce something...)
Bottom line is that I have seen these attempts to improve on C fail time and again, and it may just be time to admit that C is pretty optimal as a language, provided the coder has what it takes to understand the execution model. The execution model is also pretty simple, but requires understanding how a CPU and RAM actually works and many coders are failing at that these days. The execution models of most languages are much more complex than the one of C, but you can produce some code not understanding them. That is also a reason why some people consider C coders to be superior, because incompetent coders cannot produce anything at all in C, while they often can produce somewhat working pretty bad code in other languages.
As to average coder competence, I think the best take on it is still this one:
https://blog.codinghorror.com/...
You cannot by any means turn these people into good coders and language can only help to make their incompetence obvious. C does that well, because of the requirement that you understand the execution model.
But even if you did invent such a language, you'd still have to convince developers working on these kinds of applications to use that new language in preference to the global lingua franca of programming they all know already, and you'd still have to provide a way to link code written in your new language with the vast ecosystem of existing C-based libraries and APIs. That's a tall order, no matter how brilliant your new language itself might be.
Assuming such a language could be created, you are certainly right about this.
Of course. I might be blind as well, or we may just have different approaches to the same problem space, equally valid, but mine is far more language agnostic than yours. Whether that is better or not depends, as usual.
Being blind to more abstract views and approaches is not a question of experience and age. Some people have it, some do not. I have seen it countless times. The point is that most people cannot advance enough. You may just be in that group, sorry.
You think? You cannot have much experience in any of them (or with coding in general), if you think that similarity in syntax makes for similarity in language.
I disagree. It may seem so on your level of skill, but advance enough and that distinction vanishes.
I fully agree. The thing is, to borrow terminology from other technological disciplines, most programmers these days are mere technicians and usually bad ones. You are describing what a good engineer does. Bad technicians will never be paid well, never have good job prospects and often get replaced long-term entirely. Good technicians (which is a small subset of them) and good engineers do not have that problem. Good technicians stay on to maintain the tech they know forever and learn a bit in addition over time. Good engineers learn new stuff constantly and over time build up an impressive array of tools they can wield competently.
That is why 90% of all people writing software are not in any good position job-wise and prospects-wise: They are just not good at it. The rest has a pretty golden future.
Politicians are usually a few decades behind. This is a symptom of that.
On the server, Linux is replacing Solaris and other "old" Unixes. On the client, it may become pretty popular, because some large enterprises are preparing to go web-application only when Win7 becomes non-viable. They will not do this on Windows. All your comment shows is that you have no clue what is going on.
Ultimately, the big problem is that I don't see a real universal contender for high-performance native code taking over from C/C++. There are a lot of promising languages, but at the moment, nothing is really taking off. Simple inertia is pretty hard to overcome, as it turns out.
I do not see this being inertia at all. The aspect that gives you speed in C (and less so in C++) is control. You decide about memory management, details of calculations, etc. You decide which safety-features can be left out. You do this understanding the problem you solve. No compiler will ever be able to come close, unless we get strong AI and that is looking very unlikely today. Hence all those "promising" languages will not be able to compete either. They will just allow more bad coders to produce more bad code because a whole army of stupid "managers" does not understand that tools cannot make bad coders turn out good code.
Actually, it is C that is king. In C++ it is just to hard to write really efficient code. C++ is for all those things were efficiency is relevant, but not so critical. Of course, many C++ programs are just C with some extensions and never really to any significant OO.