Among the many erroneous assumptions this makes is that the demand for software is not elastic or limited by the cost of development. In reality what would happen is that if code becomes cheaper companies will get more than they could before to automate increasingly larger aspects of their infrastructure, improve the quality of their current software, and make custom software more accesible to small businesses.
As the very same article says, custom software is more than 90% of all software out there. The cost of licenses for proprietary enterprise software like Oracle, AIX, SAP, etc. is dwarfed by the costs of consulting and customization to make those fit into existing infrastructures.
Programmers aren't going anywhere. If anything more people will need training, more computers will need fixing, and more experts will need to be called to fix the newbie's mistakes. Just because there are more programmers in general doesn't mean that the number of competent programmers will increase.
Without getting into statistically relevant data, of the hundreds of programmers I've met in my lifetime, at most a dozen worked on proprietary packaged software (game devs, more precisely). The rest did web sites, consultancy, customizations, everything that meant going to a customer and creating or changing something that was there to better fit a set of requirements.
If programmers are in short supply become one and make tons of money. If they are over-supplied become an entrepreneur, hire lots of cheap programmers and change the world.
I work with open source technology in my job all the time, and I don't think there's any danger of someone freely implementing the custom systems we need to support our clients' needs. Just because everyone knows how a wrench works doesn't mean my mechanic is fixing my car for free.
The point is that the arrangement of bits might not be appropriate to copy; people have been writing poetry for millennia but when you need to write copy of a webpage or lay out a spreadsheet you have to write out new 'bits', you can't just tear out a page of Hamlet and hope that it does the job.
"Now computer literacy is widespread enough, and basic programming tools have gotten easy enough to use, that the ability "to write a program" is no longer in short supply, and with the Internet at our fingertips we all have the combined output of the whole world's supply of programming talent available to us, not just programmers in our own city, state or country."
> "Now computer literacy is widespread enough, and basic programming tools have gotten easy enough to use, that the ability "to write a program" is no longer in short supply, and with the Internet at our fingertips we all have the combined output of the whole world's supply of programming talent available to us, not just programmers in our own city, state or country."
On the contrary. I wish it were. Notice Apple's efforts to streamline the UI on all of its devices so that any idiot could use them without a problem. The computer literacy goes only so far as operating a GUI application, not anywhere toward the more important underlying theories of computer programming.
Yes, computers are more widespread these days, but I don't see the amount of kids being interested in programming change anytime soon. When I was a kid in the 80s, not everyone in my class or school had access to a computer, but some did. Of those, not everyone cared to learn how to program, only a few did.
Today, everyone has access to computers, and the programming tools are better and more varied, but I still think that most kids actually thinks it's boring as hell to learn how to program, there's still only a small amount of people who have a passion for it, and the focus and aptitude needed to become good at it.
Yeah, the analogy breaks down at the point where a computer program solves someone's concrete problem. The most a musical piece will do is make you feel something. People are always going to pay for solutions to concrete problems, because everybody's problem is slightly different, and needs a slightly different solution.
As long as computer programming courses in school are as inadequate as they currently are in most of the world, many people with aptitude will not be exposed to it in a compelling fashion and thus most will not choose it as a career. If that is the case, demand for professionally developed software will likely outstrip supply for the foreseeable future.
People whom are not creative and can't adapt to change will get left behind as change happens. The rest will be fine. You only have to be able to run faster than X% of the competition, and most of those people you are outrunning hate what they are doing anyways.
One fundamentally wrong thing he says is that programmers don't want competition.
This just isn't true. Competition is often good for the companies, not just the customers. Competition makes you innovative, and competition is often a great way to do R&D without spending money.
It may be that this comes to pass eventually, but soon? Not a chance. The majority of kids don't even have any access to a programming course until college. The students I see in my intro courses often don't even have a very solid understanding of what a (mathematical) function is, much less have the precision-of-thought skills one needs to successfully program. It's one of the main things we have to teach in CS1. And these are the kids that actually want to learn to program.
Historically, the kids who'll become really good programmers don't have to be exposed to it before college. They'll already have started programming before they're in high school. If they're really drawn to it, noone will be able to keep them away from it that long.
In my experience it's not a question of exposure, it's a question of the pay being offered. Most programmers that kids know probably don't make very much money. That could be changed if there was more exposure for those who greatly enjoy and profit from the craft.
I thought about that; the issue is not that one must have a college course to get started on programming, or that a course in secondary or primary school is the only pre-college exposure people could have. It's that of all students in college, a very small percentage of them have any exposure to programming. And I think we can safely assume that the percentage would be an order of magnitude lower among students that never make it to college (which still describes a majority of people---attending college for even a day puts you in a minority).
By contrast, even the most NCLB-starved schools tend to have some art and music programs, even if they're cut way back, and virtually everyone has some (perhaps inadequate) PE classes and access to pickup games of basketball or stickball or whatever. There's really no imminent chance of the number of people who have done "some" programming coming within two orders of magnitude of the number of people who have done "some" art or played "some" sports.
Starving artist is not a good analogy. Software is a trade that provides economic benefit to the recipient, rather than purely entertainment.
Better analogies would be to an architect, general contractor, plumber, or mechanic. Anyone can do these jobs with proper training. They don't live off repeated sales of single pieces of intellectual property. And they all make reliable livings (though few of them become millionaires).
I think the tradesman analogy is good, but the future programming career may have elements of the starving artist in it as well. Working for free to build a portfolio is a feature typical of creative trades that would be foreign to an architect or plumber.
Well, I don't know about plumbers, but architects know all about working for free to build a portfolio... For example, if an architect wants to do something a bit more challenging than yet another McMansion, they are going to have to enter competitions where only the winner gets paid. They do it anyhow, because placing well in a competition also helps to build your reputation as a designer.
Nonsense. Programmers aren't artists, they're engineers. The difference is progress. Whereas art hasn't come a long way since Rembrandt and Mozart, engineering has. Open Source is just the stuff that's easy and has already been done. Progress won't happen if everyone has to develop it over and over again or spend millions on licenses for trivial products.
Whereas art hasn't come a long way since Rembrandt and Mozart, engineering has.
Hahahahaha... Sorry, but that is just false. It's the same oversimplification an (ignorant) artist would make : "Technology surely has not come a long way since books and horse buggies - we can just read books without books and ride horse buggies without horses."
I'm sure it was just a lapse of attention on your part, though.
Programmers regularly create jobs for themselves everytime they write code because the code they write create bugs.
If they write practically 100% bug-free code, it is probably because the program is small or they did it so slow and methodical. Which mean, a whole bunch of other stuff not getting coded.
Trust me, there's enough programming jobs for everyone for the forseeable future.
If the demand for programmers keeps growing, no one needs to starve.
For most of its life, computer adoption has been growing exponentially - each time they got cheaper, more businesses could afford them. This meant a constantly growing demand for programmers, and therefore higher wages. Eventually, every business that wants a computer will have one, and then demand will stabilize to grow at the rate of population growth - bye-bye supernormal profits for programmers. I saw a fascinating presentation that claimed that this happened at around the year 2000.
Of course, now computer adoption is greater than 1 per capita, with it being common to have at least two of a desktop, a laptop, an iPhone and a gaming console - and greater demand for software. Also, some consolidation of software would have seemed likely (and that happened within the categories of e.g. word-processors and OSes) - yet, the number of categories themselves seems to have grown, with a webapp for every occasion, and even add-ons to websites like facebook etc.
So maybe the exponential growth in demand will continue.
I like to think that software embodies knowledge, and like knowledge, is therefore be limitless - it's just that that hasn't been the driver of industry so far
I already see it when trying to hire programmers and designers.
Because a programmer knows a mish-mash of PHP, Java and a little Javascript, they think that they are immediately qualified to write a scaleable web app.
Designers are even worse. Just because they know photoshop or dreamweaver, they think they are a designer.
I find that my programming candidates from the open-source world, though, tend to be better then those using the latest and greatest editor with some kind of intellisense on Windows.
As the very same article says, custom software is more than 90% of all software out there. The cost of licenses for proprietary enterprise software like Oracle, AIX, SAP, etc. is dwarfed by the costs of consulting and customization to make those fit into existing infrastructures.
Programmers aren't going anywhere. If anything more people will need training, more computers will need fixing, and more experts will need to be called to fix the newbie's mistakes. Just because there are more programmers in general doesn't mean that the number of competent programmers will increase.