It will never be economical to send people into space until we start doing it regularly. The only way to make something like that economical is to keep on fixing and fiddling things to make them cheaper. And that won't happen if you don't have anything to fix and fiddle.
The 'eggs in one basket' problem is the biggest reason I want us to get off the planet sooner rather than later.
The grandparent's experience mirrors my own. Some people just couldn't get it. They were really bright, wrote well, were articulate and capable. But they just couldn't get programming.
I used to spend a lot of time in the college computer lab trying to help people learn to program. Unlike many programmers, I actually kind of enjoy that sort of thing.
And there were some who understood, and many who didn't. No matter how you worked the angles or found different ways of explaining it, they just couldn't build an effective mental model of what the computer was doing in their head.
I would be really curious to watch the person in the article who claims that he can teach anybody to program and see his technique involving getting them away from a computer. Maybe he's found a way that I haven't.
But really, my personal feeling at this point is that no, not everybody can learn to program.
Thank you for such a carefully written and rational response. That's a rare thing on Slashdot. I didn't reply because I don't really have anything similarly useful to say in response, but I felt like I should at least compliment you.:-)
Knowing the basics of what a program is makes it easier to understand that these things and devices don't have to work the way they do. It also lets you understand their limitations and their strengths better.
For the purposes of having a basic understanding of what programming is, learning any turing complete language is sufficient. It's not that I think everybody should prepare for a career in which programming figure prominently. It's that I think that intelligently participating in modern society requires a basic understanding of how things work.
There's a huge problem with this, though. A little bit of knowledge is often far more dangerous than no knowledge at all. With every person a self-proclaimed "expert" on software engineering/coding because they mastered the basics of if/elses and basic for loop in highschool, we'll have a repeat of Apple v Samsung's foreman claiming "this code can't possibly run on that processor, so it's not prior art" on a massive scale. Have you ever had a boss/manager who knew a little about what you do and that made them micromanage, argue and inject patently false ideology or claims, etc? It's terrible.
This might be true, but smacks of an elitism that makes me profoundly uncomfortable. I would rather people fail in life because they try to pretend they know way more than they do than to keep people in general in the dark so they know nothing and look up to those godlike beings who are the masters of their domain.
For example, the medical profession currently has the same problem now because the Internet has made it very easy for people to learn things that previously were things only doctors knew. I prefer that doctors be forced to defend their knowledge and decisions to their patients than having their patients blindly accept the possibly flawed advice of the doctor essentially on faith.
Except that increasingly, understanding the workings of an internal combustion engine rely on understanding code. I do not think you can find one part of the economy, one device in which code is not an important piece. Perhaps silverware, maybe. Though how is it made?
Besides, fewer people that know programming means more job security and less competition for me.:P
I welcome the competition and the new ideas it will bring. I'm confident of my own ability to learn and stay on top of my field.
And people are often so incredibly surprised when they realize the level of 'own' someone has over their life if they have control of their computer. I've had the "Your computer is infected." discussion with more than one person and had them stare at me incredulously when I describe just how bad their life could be made by someone who had complete control of their computer.
It's an increasingly vital part of how absolutely everything in the world works. It's the battleground for various political factions (everything from stuxnet to DRM to Anonymous hacks). It increasingly determines what you can and cannot do with the stuff you think you own.
Not knowing anything about programming or how it works is something I consider nearly as bad as illiteracy in our society.
You realize, of course, that the first people to come out with 64-bit processors that were fully backward compatible with the 32-bit x86 instruction set were AMD. When Intel finally came out with their own a year or so later, they quietly made them completely instruction-set compatible with AMD's 64-bit offering. But their manual offered no hint that this was the case. You had to go through the entire instruction set and compare and see that all the same instructions existed with all the same numeric values. They wouldn't even tell people who asked in any direct sort of way.
For this reason, Linus was very peeved at Intel. And because Intel was being such a jerk about it all, he threatened to call the new platform 'AMD64' instead of 'x86_64'. And a few people agreed with him. The legacy of that debacle is lying around in a few places, and apparently one of them is Debian's platform naming convention.
There is a really excellent reason to migrate. The reason is the extra registers you get with the 64-bit instruction set cause a lot of programs to become significantly more efficient. Like 10-30% faster. But otherwise, you are completely correct.
And, of course, there's the mitigating effect that pointers are all twice as big and so they take more memory, more cache space, reduce locality of reference and so on. From what I've seen, the positive effect of the extra registers usually outweighs the negative effect of the pointers being larger.
The culprit is not X11, I believe, because xterm is really quick to start and responsive and highly usable over the same connection. So it must be the graphical toolkits that play high-latency ping-pong with the X server. Instead of "redefining the desktop", the focus should be on coming up with an asynchronous toolkit that could pipeline the requests properly. I understand that would make writing applications much more cumbersome, but that's where the cumbersomeness should go, not on the user interface.
I've noticed, much to my great disappointment, that most windowing toolkits tend to end up doing any number of stupid things. For example, they do not make use of the sophisticated X event system to minimize the number of events returned, and instead simply paint the entire dialog box by hand in a single X window and ask for all possible UI events within that box.
Xt, Xaw, and Motif were ugly and painful in many ways, but they at least got this right. Xterm uses Xaw, and that's why it's so very snappy.
Doing it the right way means you have to be smart about creating lots of little tiny X windows and setting up events and things just right, I suppose once you start having complex UI interactions like tooltips or tear-off menus that it starts getting to be a pain to make all that work.
He's absolutely request. GNOME's compatibility breaking is all GNOME. It's not a cultural norm set by the kernel developers.
Of course, it's much harder to define a good, stable API for upper layer stuff. It's closer to things that need to change frequently. Though X has done a remarkably good job of that.
Maybe, if that's what GNOME wants, they should sit down and think really hard about how to do it. And ignore all the current 'hot' technologies and buzzwords. That's what led them to.NET and CORBA, and those were complete dead ends.
Windows has, more or less, done it. I suspect though that it costs them a great deal. The Windows API has always been an insane mess, and IMHO a great source of the reason it was originally so very unstable.
I would like to add that I started working with Unix on University computers. And by the time I'd stopped hanging around there U of MN I had my own x86 system with a Unix on it. First it was SCO, then it was UnixWare, then Linux. Since about 1993, it's been the main platform I've used for just about anything on any of my own systems.
I've been programming since I was 8. I started with BASIC on Apple IIs, then BASIC on Atari systems, then BASIC on a Timex Sinclair 1000, then assembly on a variety of platforms, then Pascal on Atari ST and then C...
So, I'd been programming a long time before I could even really think about the job market.
My first real job was something my HS career counselor pointed me at. It was a small business who'd had an HS kid handling all of their on-site computer needs. I ended up being hired by them at just barely above minimum wage. They got way more than they bargained for.
I ended up writing my own smal windowing system, my own b-tree indexer, and a variety of other things. I sped up some of the programs their accounting department ran by 20 times (and that only because it was still limited by the speed the printer could print at). I instituted a backup policy. I made their accounting programs produce output that Quattro Pro could read so they could use Quattro Pro to create reports. I added data entry verification (which was previously completely missing). All kinds of stuff.
Eventually, a very small consulting company who was looking for the cheapest possible people noticed me and hired me. Their idea was 'manufacturing software' like you would manufacture an automobile. It was kind of a silly idea, but it got me work that was very clearly programming.
I also ended up meeting programmers online on MUDs. And talking to them, I landed a consulting gig in OKC that I wouldn't have gotten. That got Oracle on my resume.
After that, things got a lot easier. When I got back from that gig, a business the previous consulting company had hired wanted me to work for them because they were impressed with me when I'd come on-site. And a different small start-up liked my resume....
I didn't really start being chiefly focused on Unix/Linux though until Amazon paid me to move to Seattle. And even now, one of the things people really value is that there is almost no widely used computing platform that I haven't touched and learned something about.
I think there is a mix of political and non-political things going on here. I do not think Sweden is being honest (or at least being very misleading) when they say they have no plans to hand Assange over to the US. The way the UK is treating the case has everything to do with the fact we're angry with him, and very little to do with the actual possible criminal charges he's facing.
But I think there is a legitimate question of whether or not his conduct with those women was criminal or not. I would like that question to be answered. I would like Sweden to promise (a promise with teeth, something where they have to give up something really important to people or an organization with the power to enforce that and Assange's interests at heart if they break it) to not extradite Assange to the US under any circumstances, and to not base any of his sentencing (should that come to pass) on the opinion of any government about the activities of Wikileaks.
For example, if a cadre of the top candidates for a Nobel prize in the various sciences stated their intention to boycott the Nobel prize ceremony if they broke their promise, that would be enough for me.
*chuckle* Much like nobody got in trouble for leaking the fact that Valerie Plame was a CIA agent, thereby endangering her.
Yes, your comment deserves to be modded up because you're right. No matter who it is, when the leaks serve those in power, it doesn't matter if anybody is hurt, nobody is prosecuted.
But it's good that someone who was involved in the early Internet realizes that it's a good one.
And no, it doesn't mean throwing TCP/IP away.
But really, Slashdotting should be impossible. To me, the fact that it is possible indicates a fundamental problem with the current structure of the Internet. If you can come up with someone other than using content-addressing that solves the Slashdotting problem for everybody (even someone serving up content from a dialup) then it doesn't really solve the problem.
There is no decent teleconfering app that doesn't go through Google+. There has been strong pushback on the idea of using Google accounts for anything at work. And using a social networking site like that for a conference call seems broken.
Also, no good support for the kinds of instant messaging that's used inside of companies (which, unfortunately, tends to be Microsoft stuff).
As an email client, the IMAP client leaves much to be desired. And K-9 isn't much better.
Yes, this is exactly true. They could've taken the Android platform and ran in their own direction with it. They lacked imagination, or perhaps misunderstood the purpose and power of Open Source. That's what they should've done.
And you're right, when it comes to apps for doing stuff at work, Android isn't that great right now.
Yeah.:-) Those are easy to get. When I lived in MN I had a very forward thinking ISP and had an IPv6 address from the experimental (I think 3ffe::/16) range that was tunneled to a server they ran.
Slashdot Mirror
It will never be economical to send people into space until we start doing it regularly. The only way to make something like that economical is to keep on fixing and fiddling things to make them cheaper. And that won't happen if you don't have anything to fix and fiddle.
The 'eggs in one basket' problem is the biggest reason I want us to get off the planet sooner rather than later.
The grandparent's experience mirrors my own. Some people just couldn't get it. They were really bright, wrote well, were articulate and capable. But they just couldn't get programming.
This is my general feeling on the matter too.
I used to spend a lot of time in the college computer lab trying to help people learn to program. Unlike many programmers, I actually kind of enjoy that sort of thing.
And there were some who understood, and many who didn't. No matter how you worked the angles or found different ways of explaining it, they just couldn't build an effective mental model of what the computer was doing in their head.
I would be really curious to watch the person in the article who claims that he can teach anybody to program and see his technique involving getting them away from a computer. Maybe he's found a way that I haven't.
But really, my personal feeling at this point is that no, not everybody can learn to program.
Thank you for such a carefully written and rational response. That's a rare thing on Slashdot. I didn't reply because I don't really have anything similarly useful to say in response, but I felt like I should at least compliment you. :-)
Knowing the basics of what a program is makes it easier to understand that these things and devices don't have to work the way they do. It also lets you understand their limitations and their strengths better.
For the purposes of having a basic understanding of what programming is, learning any turing complete language is sufficient. It's not that I think everybody should prepare for a career in which programming figure prominently. It's that I think that intelligently participating in modern society requires a basic understanding of how things work.
There's a huge problem with this, though. A little bit of knowledge is often far more dangerous than no knowledge at all. With every person a self-proclaimed "expert" on software engineering/coding because they mastered the basics of if/elses and basic for loop in highschool, we'll have a repeat of Apple v Samsung's foreman claiming "this code can't possibly run on that processor, so it's not prior art" on a massive scale. Have you ever had a boss/manager who knew a little about what you do and that made them micromanage, argue and inject patently false ideology or claims, etc? It's terrible.
This might be true, but smacks of an elitism that makes me profoundly uncomfortable. I would rather people fail in life because they try to pretend they know way more than they do than to keep people in general in the dark so they know nothing and look up to those godlike beings who are the masters of their domain.
For example, the medical profession currently has the same problem now because the Internet has made it very easy for people to learn things that previously were things only doctors knew. I prefer that doctors be forced to defend their knowledge and decisions to their patients than having their patients blindly accept the possibly flawed advice of the doctor essentially on faith.
Except that increasingly, understanding the workings of an internal combustion engine rely on understanding code. I do not think you can find one part of the economy, one device in which code is not an important piece. Perhaps silverware, maybe. Though how is it made?
Besides, fewer people that know programming means more job security and less competition for me. :P
I welcome the competition and the new ideas it will bring. I'm confident of my own ability to learn and stay on top of my field.
And people are often so incredibly surprised when they realize the level of 'own' someone has over their life if they have control of their computer. I've had the "Your computer is infected." discussion with more than one person and had them stare at me incredulously when I describe just how bad their life could be made by someone who had complete control of their computer.
It's an increasingly vital part of how absolutely everything in the world works. It's the battleground for various political factions (everything from stuxnet to DRM to Anonymous hacks). It increasingly determines what you can and cannot do with the stuff you think you own.
Not knowing anything about programming or how it works is something I consider nearly as bad as illiteracy in our society.
You realize, of course, that the first people to come out with 64-bit processors that were fully backward compatible with the 32-bit x86 instruction set were AMD. When Intel finally came out with their own a year or so later, they quietly made them completely instruction-set compatible with AMD's 64-bit offering. But their manual offered no hint that this was the case. You had to go through the entire instruction set and compare and see that all the same instructions existed with all the same numeric values. They wouldn't even tell people who asked in any direct sort of way.
For this reason, Linus was very peeved at Intel. And because Intel was being such a jerk about it all, he threatened to call the new platform 'AMD64' instead of 'x86_64'. And a few people agreed with him. The legacy of that debacle is lying around in a few places, and apparently one of them is Debian's platform naming convention.
There is a really excellent reason to migrate. The reason is the extra registers you get with the 64-bit instruction set cause a lot of programs to become significantly more efficient. Like 10-30% faster. But otherwise, you are completely correct.
And, of course, there's the mitigating effect that pointers are all twice as big and so they take more memory, more cache space, reduce locality of reference and so on. From what I've seen, the positive effect of the extra registers usually outweighs the negative effect of the pointers being larger.
The culprit is not X11, I believe, because xterm is really quick to start and responsive and highly usable over the same connection. So it must be the graphical toolkits that play high-latency ping-pong with the X server. Instead of "redefining the desktop", the focus should be on coming up with an asynchronous toolkit that could pipeline the requests properly. I understand that would make writing applications much more cumbersome, but that's where the cumbersomeness should go, not on the user interface.
I've noticed, much to my great disappointment, that most windowing toolkits tend to end up doing any number of stupid things. For example, they do not make use of the sophisticated X event system to minimize the number of events returned, and instead simply paint the entire dialog box by hand in a single X window and ask for all possible UI events within that box.
Xt, Xaw, and Motif were ugly and painful in many ways, but they at least got this right. Xterm uses Xaw, and that's why it's so very snappy.
Doing it the right way means you have to be smart about creating lots of little tiny X windows and setting up events and things just right, I suppose once you start having complex UI interactions like tooltips or tear-off menus that it starts getting to be a pain to make all that work.
He's absolutely request. GNOME's compatibility breaking is all GNOME. It's not a cultural norm set by the kernel developers.
Of course, it's much harder to define a good, stable API for upper layer stuff. It's closer to things that need to change frequently. Though X has done a remarkably good job of that.
Maybe, if that's what GNOME wants, they should sit down and think really hard about how to do it. And ignore all the current 'hot' technologies and buzzwords. That's what led them to .NET and CORBA, and those were complete dead ends.
Windows has, more or less, done it. I suspect though that it costs them a great deal. The Windows API has always been an insane mess, and IMHO a great source of the reason it was originally so very unstable.
I would like to add that I started working with Unix on University computers. And by the time I'd stopped hanging around there U of MN I had my own x86 system with a Unix on it. First it was SCO, then it was UnixWare, then Linux. Since about 1993, it's been the main platform I've used for just about anything on any of my own systems.
I've been programming since I was 8. I started with BASIC on Apple IIs, then BASIC on Atari systems, then BASIC on a Timex Sinclair 1000, then assembly on a variety of platforms, then Pascal on Atari ST and then C...
So, I'd been programming a long time before I could even really think about the job market.
My first real job was something my HS career counselor pointed me at. It was a small business who'd had an HS kid handling all of their on-site computer needs. I ended up being hired by them at just barely above minimum wage. They got way more than they bargained for.
I ended up writing my own smal windowing system, my own b-tree indexer, and a variety of other things. I sped up some of the programs their accounting department ran by 20 times (and that only because it was still limited by the speed the printer could print at). I instituted a backup policy. I made their accounting programs produce output that Quattro Pro could read so they could use Quattro Pro to create reports. I added data entry verification (which was previously completely missing). All kinds of stuff.
Eventually, a very small consulting company who was looking for the cheapest possible people noticed me and hired me. Their idea was 'manufacturing software' like you would manufacture an automobile. It was kind of a silly idea, but it got me work that was very clearly programming.
I also ended up meeting programmers online on MUDs. And talking to them, I landed a consulting gig in OKC that I wouldn't have gotten. That got Oracle on my resume.
After that, things got a lot easier. When I got back from that gig, a business the previous consulting company had hired wanted me to work for them because they were impressed with me when I'd come on-site. And a different small start-up liked my resume....
I didn't really start being chiefly focused on Unix/Linux though until Amazon paid me to move to Seattle. And even now, one of the things people really value is that there is almost no widely used computing platform that I haven't touched and learned something about.
I think there is a mix of political and non-political things going on here. I do not think Sweden is being honest (or at least being very misleading) when they say they have no plans to hand Assange over to the US. The way the UK is treating the case has everything to do with the fact we're angry with him, and very little to do with the actual possible criminal charges he's facing.
But I think there is a legitimate question of whether or not his conduct with those women was criminal or not. I would like that question to be answered. I would like Sweden to promise (a promise with teeth, something where they have to give up something really important to people or an organization with the power to enforce that and Assange's interests at heart if they break it) to not extradite Assange to the US under any circumstances, and to not base any of his sentencing (should that come to pass) on the opinion of any government about the activities of Wikileaks.
For example, if a cadre of the top candidates for a Nobel prize in the various sciences stated their intention to boycott the Nobel prize ceremony if they broke their promise, that would be enough for me.
*chuckle* Much like nobody got in trouble for leaking the fact that Valerie Plame was a CIA agent, thereby endangering her.
Yes, your comment deserves to be modded up because you're right. No matter who it is, when the leaks serve those in power, it doesn't matter if anybody is hurt, nobody is prosecuted.
It's like a child's game of 'let's pretend' for grownups who like lots of pieces of paper, dice and rules. :-)
But it's good that someone who was involved in the early Internet realizes that it's a good one.
And no, it doesn't mean throwing TCP/IP away.
But really, Slashdotting should be impossible. To me, the fact that it is possible indicates a fundamental problem with the current structure of the Internet. If you can come up with someone other than using content-addressing that solves the Slashdotting problem for everybody (even someone serving up content from a dialup) then it doesn't really solve the problem.
I could only read it and helplessly chuckle to myself thinking "Why, of course it is!".
There is no decent teleconfering app that doesn't go through Google+. There has been strong pushback on the idea of using Google accounts for anything at work. And using a social networking site like that for a conference call seems broken.
Also, no good support for the kinds of instant messaging that's used inside of companies (which, unfortunately, tends to be Microsoft stuff).
As an email client, the IMAP client leaves much to be desired. And K-9 isn't much better.
That's true. But I bet that might be a lot easier than creating a whole new OS for yourself.
Yes, this is exactly true. They could've taken the Android platform and ran in their own direction with it. They lacked imagination, or perhaps misunderstood the purpose and power of Open Source. That's what they should've done.
And you're right, when it comes to apps for doing stuff at work, Android isn't that great right now.
Yeah. :-) Those are easy to get. When I lived in MN I had a very forward thinking ISP and had an IPv6 address from the experimental (I think 3ffe::/16) range that was tunneled to a server they ran.