APIs, Not Apps: What the Future Will Be Like When Everyone Can Code

An anonymous reader writes: There's been a huge push over the last few years to make programming part of the core academic curriculum. Hype or not, software developer Al Sweigart takes a shot at predicting what this will be in a future where some degree of coding skill is commonplace and he has an interesting take on it: "More programmers doesn't just mean more apps in app stores or clones of existing websites. Universal coding literacy doesn't increase the supply of web services so much as increase the sophistication in how web services are used. Programming—by which I mean being able to direct a computer to access data, organize it, and then make decisions based on it— will open up not only a popular ability to make more of online services, but also to demand more.

Almost every major website has an Application Program Interface (API), a formal specification for software to retrieve data and make requests similar to human-directed browsers. ... The vast majority of users don't use these APIs—or even know what an API is—because programming is something that they've left to the professionals. But when coding becomes universal, so will the expectation that websites become accessible to more than just browsers."

Comments

"When everyone can code . . . "

[PolygamousRanchKid]

. . . everyone will just think that they can code.

Hey, I have a Black & Decker cordless drill! And a can of Spackle . . . I guess that makes me a dentist!

Re:"When everyone can code . . . "

[Krishnoid]

Sure, but it's better if you have some practical experience first.

Re:"When everyone can code . . . "

[mark-t]

Not to be confused with when people can make code just by thinking.

Re:"When everyone can code . . . "

[Dutch Gun]

The trend is actually reversing in some ways. Many kids are no longer interested in tinkering with PCs - which used to be a near requirement just to get them to work. They just see them as appliances now, about as interesting or exciting as their refrigerator, except for what it can DO for them. And it's often easier and more convenient for them to simply use their iPad.

So, no, I just don't see a world in which everyone is a programmer. There will certainly be a lot MORE programmers than ever, but it will still be viewed as something of a black art by the rest of the population. It's no different than an auto mechanic in that regard. The average person nowadays opens the hood of their cars and their eyes glaze over. They have no idea how to fix anything in there, and they don't want to know. There are people who do that for a living, and it's far more efficient to simply pay someone else to fix it.

Seriously, can you imagine an average person wanting to program something for themselves to scratch some itch, or do you think they'll just find a $5 app to take care of it for them? The second scenario sounds a hell of a lot more likely to me.

Re:"When everyone can code . . . "

[hawguy]

. . . everyone will just think that they can code.

Hey, I have a Black & Decker cordless drill! And a can of Spackle . . . I guess that makes me a dentist!

I don't understand the big push to get everyone to code -- not everyone *wants* to code, nor should they have to. Not everyone knows how to grow their own food, fix their own car, build their own house, and they don't need to - there are specialists for all of those that are better trained and more skilled at it and will do a much better job. Some people may grow a small garden or tinker with cars as a hobby, but few people are capable of effectively growing food for their family or overhauling an engine. Just some people may enjoy creating small (or even large) software projects for fun, but not everyone wants to.

Re:"When everyone can code . . . "

[Anonymous Coward]

Bingo. Relatively few people -want- to code. Look at how well iOS is doing which not just separates the user from any meaningful interactions with hardware, but forcefully locks them away from anything other than basic commands (turn on/off Wi-Fi, power on/off the device, etc.) Similar with Android (where it took a $20,000 bounty to just get root on a Samsung phone, much less have a bootloader exploit that doesn't blow up KNOX.)

The problem is that to management, developers are fungible. One coder can be tossed and another coder dropped in. With this in mind, someone who knows their stuff, but gets a good paycheck gets chucked for a bargain basement H-1B, who can code 10,000 lines of code a day. Bugs and crappy code are accounted for in this process. This is why deadlines slip, and "hey, it builds, package it" is the norm for the industry. The day of shipping a game once without updates (which was the norm for consoles all the way to the PS/2) is impossible now, with the only exception being malware.

Coding may be an interesting hobby, but so is bottle cap collecting or battery swallowing. Neither makes money. It is a lot like textiles these days -- completely replaced by commodity workers.

It would be nice to see software companies go back to actual quality... but they will be turfed by the guys who promise the world and ship pure crap.

Re:"When everyone can code . . . "

[Anonymous Coward]

The government is pushing this because they realize now what a horrible idea it was to off-shore the majority of our manufacturing jobs. They are trying to find something for everyone to do instead of 9-5 jobs. Coding is not it. 80% of the people in this area (upstate NY) are not capable of putting together a coherent sentence. There is no way the ADD riddled population can maintain a clear head long enough to make x do y.

Re:"When everyone can code . . . "

[Junta]

Exactly, *everyone* being able to code flies in the face of pretty much every other industry.

Decades ago, everyone knew how to work on cars. In part because cars were more straightforward, but *mostly* because cars were so unreliable, you pretty much had to know how to work on it to be able to own one without going nuts. Nowadays a typical car maintenance schedule has a a recommended oil change every 8,000 miles or so and otherwise very little maintenance that hits a *typical* car. Tires and brakes wear out over a long period of time, but generally there's plenty of warning. A car suddenly not starting or general misbehavior is no longer that commonplace, so most people just leave it to a mechanic. Even things as dead simple as changing oil or even the air filter is perceived as a black art by most folks.

Same for computing. If you had a system in the 70s or 80s, you essentially had to know how to program to be productive. As the software ecosystem has matured with canned applications targeting the use case of more and more and more usual situations, people just don't need to know that stuff anymore. As a portion of the computing population, programmers are less represented now than they were in the past. To imagine a reversal of that trend is silly.

Some view programming as a sort of 'literacy', but that's more of the misnomer of programming as 'languages'. It's really not a form of literacy at all.

Re:"When everyone can code . . . "

[Darinbob]

And this "coding" isn't what many people would call coding. There are people out there who seriously think that creating a web page is coding, or packaging up a URL inside some XML and calling it an app is coding.

Re:"When everyone can code . . . "

[Anonymous Coward]

Also, some of those manufacturing jobs left because there weren't enough people with the skills to do them. There are people who were directed to get some bullshit college degree (most likely with the word "studies" as a part of the title) that has little use in the real world. They could have been learned a useful trade at a vocational/technical school instead and actually do something productive. It's why some manufacturers or states are offering free vo-tech education for some trades: they need skilled workers and there aren't enough available.

IMHO, the "Everyone Should Code" push is being done by the new "Internet billionaires" who want a bigger (read cheaper) job applicant pool. Sadly, the press equates marking up a text document with HTML tags as the same as coding.

Re:"When everyone can code . . . "

[mjwx]

The trend is actually reversing in some ways. Many kids are no longer interested in tinkering with PCs - which used to be a near requirement just to get them to work. They just see them as appliances now, about as interesting or exciting as their refrigerator, except for what it can DO for them. And it's often easier and more convenient for them to simply use their iPad.

This.

Computers are seen as appliances these days. In the future there will be fewer computer users who can program, not more.

The same happened with everything we consider appliances now from refrigerators to cars. Not so long ago, being able to fix a washing machine or perform rudimentary maintenance on a car was almost mandatory for owning one, now these skills are rapidly disappearing as people need to call the AA (roadside assistance) to change a tyre.

Re:"When everyone can code . . . "

[Darinbob]

That's the problem with IT now, the IT workers are treated as fungible, and the workers help this attitude along by going out and getting certificates as proof of their qualifications ("look, I'm interchangeable too!").

But I do embedded systems. The API is often just the data sheet (sometimes because the chip maker's own library is utter crap). It won't ever be a fungible job. It's too difficult to offshore effectively. The best bet for a programmer or engineer is to not be just one of the masses. Even if you end up doing normal every day programming it's very good to have some other skill or experience that stands out (domain knowledge).

Re:"When everyone can code . . . "

[JustAnotherOldGuy]

Hey, I have a Black & Decker cordless drill! And a can of Spackle . . . I guess that makes me a dentist!

Or a gynecologist.

Re: "When everyone can code . . . "

[lucm]

Is our children learning the internets?

Re:"When everyone can code . . . "

[chihowa]

In part because cars were more straightforward, but *mostly* because cars were so unreliable...

Cars back then actually weren't all that straightforward, which is a huge reason why they were so unreliable. Much of it is due to advances in materials, but electronic fuel management, spark advance, etc are much simpler and less fidgety designs that the mechanical kludges that used to be used. The physics behind a carburetor, for example, are fairly simple but keeping one well tuned has sent many the amateur mechanic over the edge!

The tools you need to work on new cars are different, but well within the reach of an interested amateur. Try poking through an old (pre-1980s) mechanic's toolbox if you want to see a crazy collection of inaccessible tools.

Re:"When everyone can code . . . "

[ILongForDarkness]

That is just it. It isn't just a matter of can you understand what an if or for loop is. It IS a skill/psychological problem: you need be the type of person that wants to understand how things work all the way down to the metal, and stubborn enough to keep at it till you can get things to work. If anything the more easy to use computers become the more frustrating it will be to fix that 1% of remaining issues. It takes a special type of idiot to keep bashing their heads against the wall when things are already "good enough".

Re:"When everyone can code . . . "

[ILongForDarkness]

Amen to that. I both don't think it is possible: not everyone has the temperament to stick with a problem for several days till they figure something out, nor desirable (there are a lot of dumb people out there (or at least dumb when it comes to tech)) for everyone to code. I want my lawyer to focus on kicking ass at being a lawyer, by doctor to focus on kicking ass at making me healthier etc. We ain't the only profession that will matter in the future no matter how much tech is around, there is still going to be the need for good art, good music, good health and other advise, good bjs etc. Everyone has their place.

Re:"When everyone can code . . . "

[ILongForDarkness]

Part of the problem is the expectation that everyone contributes/carries their wait. I'm sorry but if robots truly replace all menial jobs the idiots out their will have no work that makes sense for them to do, unless you subscribe to the Calvinist work ethic: work itself has value. Drop the Jebus and all that is left is: keep the idiots out of the way while the remaining "useful" people work. It is only our sense of fairness and greed for "more" that makes us think that it is a useful use human life to have someone paid full time to flip a burger when a machine can do it just as well and cheaper.

Re:"When everyone can code . . . "

[ILongForDarkness]

Yeah that is scary. You can say because the cost of crap code is lost customers/law suits etc,not to mention moral: knowing you are selling crap but not caring enough to actually try to deliver on your promises. They could still factor in all those costs and say Tata is still cheaper sorry. My degrees are in physics and management so if programming doesn't work out I have other domains to move in. I think that is what we need generally in our profession and probably others: you are just a budget cutback away from your grant being denied, your department outsourced etc. If you are truly smart learn multiple useful professions and hope like hell that they don't all go to crap at the same time.

Re: "When everyone can code . . . "

[jhoger]

A programmer is to one who can code is as a mathematician is to one who can do algebra.

If we put enough emphasis there could be plenty of people who could tie a couple of glue a couple of Apis together with a script to do some work.

Re:"When everyone can code . . . "

[Dutch Gun]

In part because cars were more straightforward

I grew up with a Honda Civic and an old Ford van in my family. As a youngster, whenever I looked under the hood, I marveled that people used to be able to work on cars themselves. I eventually got a chance to peek under the hood of some older car from the 60's (I forget the make/model), and I finally understood. Wow, cars engines were a LOT simpler back then, and what's more, the engine compartment wasn't nearly so cluttered. There was actually enough room to get in and work on things, and you could clearly see all the critical components of the engine and figure out how they worked for the most part.

Early computers were a bit like this as well. Think about how simple and minimalist a command-line interface (when that's ALL you had) was compared to the layers and layers of abstraction you have in the OS today. There's no doubt both the hardware and software is vastly more powerful today, but I'd imagine it's actually quite a bit harder for a student today to get a real grasp on what the computer is actually doing at a fundamental level. There's simply a hell of a lot more to learn, so programmers are necessarily becoming specialists.

I don't see this as a bad thing, because it simply means computers are reaching an appropriate level of maturity as devices intended for use by lay-persons, rather than exclusively by and for specialists. We "specialists" occasionally grumble about attempts to "dumb down" computers, but by and large, I think it's a good thing that computing is now ubiquitous, as it's added a tremendous amount of convenience to everyone's lives. And besides, the broader that market is, the more opportunity we have to earn a living writing software for the benefit of the masses.

Re:"When everyone can code . . . "

[narcc]

You've elevated programming to a laughable level. It's more in line with basic literacy than it is to higher math or physics.

Programming is ridiculously simple. So simple, in fact, that young children can (and often do) teach themselves. I'd say that would put it below basic literacy if that weren't a necessary prerequisite.

Reading and writing are skills we teach everyone. Anyone, unless severely disabled, can learn to read an write. Yet very few go on to be professional writers. Fewer still are writers worth your time. Yet they're still valuable and, in this modern age, essential skills.

The problem with developers is that every hack out their wants to believe that they're Hemingway. They're not, of course. They're more like a newspaper reporter in charge of covering community events. I suspect every developer, save the hopelessly deluded, understands this. That's why we see so many posts insisting that programming requires a "special mind" or that most students will "just forget" after the class is over. They're modern scribes, lobbying to keep literacy out of the hands of the general public. They have no other skills, and they're terrified that they'll find themselves unemployed.

That's ridiculous, of course. Like writing, programming is a skill. Like all skills, it improves practice. Just as we don't hand out writing jobs to anyone who can hold a pencil, we won't hand out programming jobs to anyone who can bang on a keyboard. Though everyone who can program can benefit from that skill, just as anyone who can read and write can benefit. Just as any literate person can tell a story, leave a note, make a list, etc, so can someone with basic programming skills solve a problem, make a game, or whatever it is that's important to them.

Unless they discover Python. Then they're a lost cause.

Re:"When everyone can code . . . "

[narcc]

Relatively few people -want- to code.

I disagree. Lots of people want to do things that would require them to write code, but they don't know how.

What you want, unsurprisingly, is to keep competition out of the market. If you're afraid that you can't compete, perhaps you should develop other skills?

Programming is incredibly simple, after all. Anyone can learn. Few do, unfortunately, because people like you spread the myth that it's incredibly difficult.

Re:"When everyone can code . . . "

[Anonymous Coward]

> Programming is ridiculously simple. So simple, in fact, that young children can (and often do) teach themselves.

Learning to store values in variables, write "if else" and loops, and call subroutines is ridiculously simple.

Learning a dozen frameworks, a hundred "gotchas" in your language of choice, security best practices, database interfaces, GUI libraries, IDE quirks, platform incompatibilities, version control and arcane build systems & toolchains is NOT simple, and I would challenge anyone to learn a modern commercial software dev process in less than 5 years.

Re:"When everyone can code . . . "

[TheRaven64]

That depends a lot on what you mean by programming, and that's where this kind of discussion falls down. Anyone who has used an office suite has done, by hand, things that could be trivially automated. It's not a huge step to get from complete manual automation to writing some simple scripts that make life easier. It's a much bigger jump to get from there to being able to write complex and maintainable applications with sensible algorithms, but in a lot of cases that naive n^2 algorithm will do fine because you're only giving it small values of n and even a slow modern computer will do it a lot faster than the operator (which is the goal).

It's also worth remembering that most things that people will need to program are DSLs, not general-purpose languages. Before he retired, my stepfather was head groundskeeper at a golf course. The system that they used for controlling the irrigation system was a Turing-complete (visual) DSL. It had loops, conditionals, and maintained arbitrary state. It took input from a variety of sensors (motion sensors to track people, sunlight and moisture sensors around the place) and controlled a load of valves for the sprinklers. This is the kind of programming that someone in a blue-collar job has to do today. The programs are simple enough that maintainability is not a serious concern, but they do need debugging sometimes (get the conditions wrong and you accidentally get some quite damp people!).

Re:"When everyone can code . . . "

[Mr.CRC]

Reading is pattern recognition and mapping to innate experiential meaning primitives conditioned since birth. Writing is just transcription.

Programming is entirely different: decomposition, 1-N stages of abstract generalization, followed by iterative optimization of mapping to the thinking primitives of an alien life form.

All I know is that I've seen people of average intelligence who simply couldn't get the most basic programming concepts. Even if they could grasp programming primitives at some level, they were still miles from being able to know WHY to program. They simply couldn't see the connection between tasks they perform, and telling a computer to do it for them in an automated way.

It seems modern GUIs require us to manually perform many repetitive tasks with our computers. It's highly regressive.

Re:"When everyone can code . . . "

[TuringTest]

> I don't understand the big push to get everyone to code -- not everyone *wants* to code, nor should they have to.

Not everyone wants to write novels, why would they want to learn to write?

There's a sense of freedom in being able to solve your own problems without having to delegate the solution to a professional. There are lots of small, day-to-day tasks that could benefit from being automated, i.e. triggered under conditions set up by the users themselves (anything beyond "set an alarm off" is beyond the reach of average users with current tools).

There are some mobile apps that allow users to set such rules, but they are too difficult to use and little known. The alternative is letting Google build all such automations for you in the form of Google Now, at the cost of:

1) only having those automations that the Google guys put in the system,
2) being unable to alter their behavior if it doesn't fit your needs, and of course
3) revealing your whole life to Google so that they can make it work.

A really easy coding tool plus some coding literacy would allow people to reap all the benefits of Google Now without selling your soul to the company.

Re:"When everyone can code . . . "

[K. S. Kyosuke]

If ten-year-old children were able to write useful programs in the 1970's experiments at Xerox PARC, anything less than that today would make us having made zero progress during the past forty year. I think you're mistaking teaching ordinary people have some trivial woodworking skills into making the population a nation of professional carpenters. That is hardly the case here.

Re:"When everyone can code . . . "

[K. S. Kyosuke]

Exactly, *everyone* being able to code flies in the face of pretty much every other industry.

The problem with that statement is that computers are not like "pretty much every other industry". Otherwise Excel wouldn't be one of the most popular computer applications of all times.

Re:"When everyone can code . . . "

[K. S. Kyosuke]

Learning to store values in variables, write "if else" and loops, and call subroutines is ridiculously simple.

This is equivalent to learning the letters of the alphabet. That won't make you a writer, though. Writing classes are about something different.

Learning a dozen frameworks, a hundred "gotchas" in your language of choice, security best practices, database interfaces, GUI libraries, IDE quirks, platform incompatibilities, version control and arcane build systems & toolchains is NOT simple, and I would challenge anyone to learn a modern commercial software dev process in less than 5 years.

I think you're making two major mistakes here. First of all, you're assuming that these things are intrinsic rather than accidental complications of current computing. Some people believe them to be the latter rather than the former. Second, even if this were the case for core computing functionality, there's still a large class of useful programs that does not necessitate the involvement of any such complications into the process of programming. Nobody is calling for turning the population of this world into a nation of low-level programmers by trade. There's a middle ground between that and current casual computer users that multiplies the productivity of computer users considerably.

Re:"When everyone can code . . . "

[K. S. Kyosuke]

Modern GUIs have almost entirely abandoned the concepts of flexibility that were expected in the 1970s to take over the world in the future. One of the reasons for that was the succession of reinvention of GUI technologies by several waves of new computing devices that started small whose creators, having come from different backgrounds, had only a very faint grasp of what progress in computing had been made previously in larger computing systems. Macintoshes, Unix RISC workstations, Windows x86 machines are all parts of this sequence of knowledge decay.

Re:"When everyone can code . . . "

[K. S. Kyosuke]

This is exactly the argument that the people from VPRI and behind HtDP are making in favor of teaching everyone the basics of computing. Not everyone needs to be a novelist, but most people would benefit from being able to compose a simple letter without mistakes.

Re:"When everyone can code . . . "

[K. S. Kyosuke]

I've posted this above, a text which I think makes some compelling arguments.

Re:"When everyone can code . . . "

[Anonymous Coward]

I think many people underestimates maths and coding in everyday life. It always amazes me when people spend whole day or even more manually editing thousands of rows in a spreadsheet instead of spending half an hour to find appropriate function(s) for the job. Or people in soft science fields: "Oh, we dont need maths - we are not mathematicians" and then making absolutely wrong conclusions about some survey because statistics interpreted wrong. You dont need to be able to code whole programs or to solve differential equations, but some basics of coding and math will often make you much more productive

Re:"When everyone can code . . . "

[thegarbz]

I don't understand the big push to get everyone to code -- not everyone *wants* to code, nor should they have to.

I have met a lot of people who said they don't need or want to code in their lives. Those people are the same that fail to understand a complex sequence of logic.

I'm not a programmer. Yet something as basic as excel formulas can get complex enough to be considered "programming", and that's before you get into some of things that have made my life a LOT easier such as ill conceived but none the less working VBScript.

People could benefit a lot from not being full blown programmers, but at least taking a basic course on discrete mathematics and logic. That along with common sense is something that is woefully under-represented in today's society.

Re:"When everyone can code . . . "

[elgatozorbas]

I don't understand the big push to get everyone to code -- not everyone *wants* to code, nor should they have to.

It is just a /. thing: every two weeks there is an article along the line of "how can we teach children to code", "how can we teach poor people to code", "why everyone should be a coder" etc. I know this is "news for nerds", but imho the (recent) emphasis on coding is over the top.

Coke and Mentos rockets are also cool nerd stuff, but those headlines have become very scarce.

Re:"When everyone can code . . . "

[LaurenCates]

I'm not sure what you mean by this.

Most people that learn math. It doesn't mean that everyone who takes math thinks of him or herself as a mathematician.

Not that I think everyone should learn to code, but if everyone was taught to code, I'm pretty sure most of the people who learned wouldn't think of themselves as people who think they can code.

Re:"When everyone can code . . . "

[AmiMoJo]

There are many levels of programming, from hacking assembler to point-and-click languages. If you look at basic programming courses for kids they are mostly visual and tend to avoid written code, instead focusing on logic and combining things. All useful and transferable skills.

Re:"When everyone can code . . . "

[narcc]

In my experience, I've never had a student who just "didn't get" programming. I've had a few lose interest, but every last one mastered the basics, with little difficulty.

The literature seems to support my experience as well. This absurd belief that some people just can't code is not.

Perhaps you're just not a very good instructor?

Re:"When everyone can code . . . "

[RabidReindeer]

Myself, I look forward to the day when everybody can manufacture their own automobile transmission.

Re:"When everyone can code . . . "

[houghi]

And even if everybody could code, some will be better at it and that will be the coders. Just like some will be better at making bread and yet others better at banking, being a manager or being a nurse.

Just like now.

Everybody can run (using the generalisation of everybody) That does not mean everybody is joining the Olympic Team of their country.

Re:"When everyone can code . . . "

[mrchaotica]

That goes against everything I've heard anyone say about older vehicles: "They were easier to work with". I brought my newer vehicle in to get a headlight changed and it took 3 mechanics to figure it out.

I'm going to take a wild guess that your car is German. A lot of those cars are almost intentionally designed to be hard to work on. In contrast, the Korean car I used to own had the headlight assembly attached simply by a couple of Phillips-head machine screws.

Of course, in the particular example of headlights older cars were different due to regulation. Until the early '90s cars sold in America were required to use one of only a few varieties of sealed-beam bulbs, which is why all cars from that era used either circular or rectangular headlights, and why pop-up lights were so common. (Designers didn't like the look of the lights they were stuck with, so they hid them.)

Re: "When everyone can code . . . "

[Junta]

People are forced to do algebra in high school.

Most people never use it after their academic life. They might implicitly use some of the common concepts without thinking which is good, but they'll not sit down to write out an algebra problem.

Re:"When everyone can code . . . "

[frank_adrian314159]

Good luck at taking the "accidental" out of languages, let alone quirks out of frameworks. A marvelous goal, but the trick is in actually doing it. Make sure to have any fixes backwards compatible (or have a way to automatically upgrade existing code), too, because all of those amateur programmers still don't want you to fuck over their code when you fix what you think is an accident or a quirk.

Re:"When everyone can code . . . "

[Junta]

Excel is at it's most sophisticated use in most cases an integrated pocket calculator. Yes Excel can do very potent things for people who know more, but most of the time I see it used as just a tabular text editor with sporadic people knowing how to make a cell auto-add other cells.

It is hubris to think that this particular skillset has more universal appeal than any other industry that was transformative. Just because it's important and everywhere, doesn't mean everyone is willing and able to become masters of it. Cars are also massively popular, and yet only a relative few are mechanics. Indoor plumbing is very popular, yet folks still call for plumbers when things go south. Popularity does not imply mastery.

Re:"When everyone can code . . . "

[toddestan]

The thing is, those systems may be kludgey compared to modern electronically controlled systems, but to someone with some mechanical aptitude, basic electronics, and the fundamentals of engines worked, they were fairly easy to understand and figure out why the car wasn't running right when they inevitably went awry. Sure, they required constant adjustments and tinkering, but it's something you could do and was just part of owning a car. Nowadays, you don't have to think about that stuff - cars are pretty much turn the key and go. But if it doesn't start, you pretty much can't do much but open the hood at stare at it. Used to be you just had to make sure you had fuel, spark, and compression and the thing should run. Those days are gone.

Thought pre-1980 does include the cars of the 70's and their early emission control systems. All analog, usually involving a maze of vacuum lines. Some of those cars are totally bewildering to work on.

Re:"When everyone can code . . . "

[K. S. Kyosuke]

Well, it's been shown that it can be done. And if anyone is capable of pushing it further, I'd trust people that Alan Kay trusts.

Re:"When everyone can code . . . "

[K. S. Kyosuke]

They are essential to programming in the real world, not an academic ivory tower.

IDE quirks, platform incompatibilities, and arcane build systems are "essential"? I'm sorry, but I'm quite sure I could live just fine without them. That's like saying that complicated tax code is "essential" for running a country. In reality, it's most likely much more essential for accountant job security.

Where have I heard this before?

[Frnknstn]

So with this API stuff, what you are proposing is that all systems should be divided up into functional blocks, where every block has a single logical purpose, and that savvy users are able to chain these functions together however they see fit to suit their purposes?

Where have I heard this idea before? Oh yeah... it's called The Unix Philosophy.

Re:Where have I heard this before?

[ColdWetDog]

No, no, no. The Unix philosophy is old. Think guys with ponytails way longer than they should have them.

This is different because it will be SHINY. And patent pending, trademarked and protected by copyright until you can't do a damned thing without paying extra for the privilege.

Get with the program. You're probably a socialist.

Re:Where have I heard this before?

[CRCulver]

Where have I heard this idea before? Oh yeah... it's called The Unix Philosophy.

Indeed. I've been using Linux since the turn of the millennium, but in the last couple of years I've been trying to gain a more proficient command of Unix standard utils and piping commands with tutorials like O'Reilly's Classic Shell Scripting . I feel like a computing god, and friends and relatives are baffled at how I can so quickly solve computing needs that, they believed, would have to take minutes or hours of laborious pointing and clicking.

And that's why I find the premise of this article so odd. The average public does not seem to me on the cusp of a programming revolution. I might as well link here to Philip Guo's essay The Two Cultures of Computing, a.k.a. "How Ya Gonna Get 'Em Down on UNIX After They've Seen Spotify?". The interfaces ordinary people use so hide hackability that they generally forget it even exists.* Plus, with people in the developing world starting to do more and more of their computing on their phone, a device without a real keyboard, they are hardly able to do all the typing that coding requires.

(Perversely, this might be something that millions of people should be thankful for: that ignorance is why they still have jobs. So much time-consuming work could be done in a much shorter time were the Unix philosophy applied. If scripting were something that managers keen on every possible costsaving measure were strongly aware of, even more jobs would be automated away.)

Re:Where have I heard this before?

[Dcnjoe60]

Programming is like writing. Just because you know how to write, doesn't mean you have anything worthwhile to write about. Likewise, just because you know how to code, doesn't mean you have anything needing to be coded.

Re:Where have I heard this before?

[CRCulver]

In place of "people in the developing world", I actually meant to write "in the developed world", but in fact, it makes no difference. Android phones (even if they have to be bought on credit) are increasingly common even in the Third World that a decade or two ago would have seemed the antithesis of technological.

Re:Where have I heard this before?

[silas_moeckel]

My basic IT interview question is give them an example task and how often it will happen and see how they document it. The bad have notes that end up like click ok next next do this etc so that it's specific to a particular UI and minor changes invalidate it and do things via the UI. The good tend to make scripts or one liners. It's not even unix centric anymore. Take a basic task of adding a user to AD and assign a primary SMTP address and a couple aliases you can use AD's gui and then pop open ASDIedit or bang out a line or few of powershell. The best figure out that it's a list and bang out a quick loop.

Re:Where have I heard this before?

[ADRA]

Well, more relevantly, it was that Yahoo based web services mixin project that they shuddered recently (a year maybe)? Never used it, but it was an interesting idea that alas I heard about as they were murdering it.... =/

Re: Where have I heard this before?

[Anonymous Coward]

But one day all of Unix will be contained in systemd so programmers will no longer be required.

Re:Where have I heard this before?

[turbidostato]

"Just because you know how to write, doesn't mean you have anything worthwhile to write about."

Quite true, but still irrelevant in shaping a future. Geocities, Twitter, Facebook, Wordpress... all be my guest.

Re:Where have I heard this before?

[youngatheart]

You're quite right. The good news that instead of only the elite knowing how to read and write as was normal a century ago, now most people can communicate with the written word. The number of interesting things to read and the number of people capable of appreciating it now compared to just a few generations ago is amazing when you stop to think about it. Maybe only one in ten thousand is a great writer, but not that percentage is taken from a hugely larger portion of the population. Likewise with coding, perhaps only one in ten thousand will be a coder with impact, but if that is taken from 318 million instead of 2 million, that's a big impact.

Re:Where have I heard this before?

[thegarbz]

Where have I heard this idea before? Oh yeah... it's called The Unix Philosophy.

Ironically both the Unix Philosophy and the idea that everyone is going to be a programmer and everything will be an API are equally viable in the modern world. One is dead of old age and the other is a concept dead on arrival.

Re:Where have I heard this before?

[godefroi]

Except that this will be the Unix philosophy, with advertisements. So much better.

Re:Where have I heard this before?

[Dr_Barnowl]

Yegods yes.

I worked in a place that used a whitelisting program - the very antithesis of a useful computer is one that's been locked down to only run what has been deemed OK by the IT dept.

You couldn't even write a .BAT file.

Re:Where have I heard this before?

[Dcnjoe60]

"Just because you know how to write, doesn't mean you have anything worthwhile to write about."

Quite true, but still irrelevant in shaping a future. Geocities, Twitter, Facebook, Wordpress... all be my guest.

But that just goes to prove my point. Look at how those services are used by millions of people, the vast majority of which are not programmers. In the early days of the automobile, one had to pretty much be a mechanic to drive and keep a car running. Today, that is no longer the case. Likewise with programming. Just as driving a car doesn't require special skills anymore, neither will using online services.

No.

[nmpg]

"But when coding becomes universal, so will the expectation that websites become accessible to more than just browsers" That's just not true. Website want you to go to the ... you guessed it, the website! Do you really think everyone will offer access to content without making you see these pesky ads? Ain't gonna happen dude.

Re:No.

[Dcnjoe60]

"But when coding becomes universal, so will the expectation that websites become accessible to more than just browsers" That's just not true. Website want you to go to the ... you guessed it, the website!

Do you really think everyone will offer access to content without making you see these pesky ads?

Ain't gonna happen dude.

Yes, because in the future, where everybody codes and everything on the web is a service, you won't have ads. You'll just pay directly for the services you want to access.

Re:No.

[alexhs]

you won't have ads. You'll just pay directly for the services you want to access.

I thought that if American cable TV ever taught us anything, it is that you will end up paying for the service and have ads.

Re:No.

[Stan92057]

I am not a programmer and i know that ads are not just to cover the cost of a site those who think so are very wrong. You will be forced to buy the program AND you will also get ads with it because ads bring MORE money, MORE profits. Just look at any magazine subscription that's a service and in that magazine how many pages are ads? half or more. so you theory of service means no ads is a pipe dream. HBO has ads,magazines have ads,newspapers have ads,netflix WILL have ads.

There are scant barriers at the moment

[smittyoneeach]

You can't lower the bar much more without crushing a foot.
You can lead a horse to water, but you cannot make it drink.

Prima facie ridiculous

[argStyopa]

The assertion seems to be that since code will be ubiquitous, then everyone will learn to code is provably silly.
Cars have become universal, yet the ability to fix them decreases with every generation.
Literacy is nearly universal now, would anyone assert that people write better now than they did 100 years ago?
Let's use even a code-like example: mods for first person shooter...the modding community used to be robust for these games but as the level of quality and complexity of the games have improved, I'd posit that the modders have (largely) disappeared; it just takes to much time/effort for hobbyists to make something comparably decent that doesn't look like crap compared to professionally produced levels.

Re:Prima facie ridiculous

[ShanghaiBill]

Cars have become universal, yet the ability to fix them decreases with every generation.

A big reason for that, is there is less that can go wrong. A fuel injector is much more reliable than a carburetor. My wife has an electric car. There is no oil, no radiator, no transmission, etc. Even the brake pads don't need to be replaced, because of regenerative braking. It is nearly maintenance free, and there is almost nothing that can go wrong.

Literacy is nearly universal now, would anyone assert that people write better now than they did 100 years ago?

Yes, I would assert that. The average person today is almost certainly a better writer than the average person a century ago.

I'd posit that the modders have (largely) disappeared

We organized an after school session to learn Mindcraft modding using Python at my son's elementary school. We had seats for 20, and we had nearly 40 kids sign up. We even had a few girls sign up.

Re:Prima facie ridiculous

[ColdWetDog]

We've seen this a couple of times before. BASIC, the thing with the turtles, even HTML.

First off nobody cares. This is especially true for large values of nobody. If it requires much more mental effort than breathing, most people do not want to interact with anything or anybody. There are already APIs for much of everything. But, surprise, they're complicated. It takes study and smarts to use them.

And that is the big problem. The coding is left to professionals because it's too complex to leave to anyone else. All of those attempts to simplify programming have not made it much past simple look up forms. Remember Microsoft Access? VBA? Yes, it can be useful but it's not all that easy to do much with, especially if it regards real data because you have to understand a whole bunch of things about data before you can let the world have access to it.

Re:Prima facie ridiculous

[Moof123]

Agreed.

API's are just one piece necessary for a committed individual to be able to accomplish something. Even with well documented API's the average Joe just wants stuff to work. If I have to go learn a dozen API's just to make a sandwich, I won't.

Learning to code does not mean you are magically omnipotent. As a microwave design engineer I have coded a fair number of things for work along the way, but if I revert to writing code most of my daily work I will immediately be far behind schedule. Each API is another hunk of vocabulary to learn and become fluent in to be able to make use of it. Most of my design software has an API, but becoming an expert in that means I will not be one for my real work. We have dedicated CAD guys to do most of the heavy lifting for customization.

Might as well argue that once everyone learns to solder the plants in China will go out of business...

Re:Prima facie ridiculous

[vux984]

The assertion seems to be that since code will be ubiquitous, then everyone will learn to code is provably silly.

Silly yes. Provably silly? I don't know about provably.

But even if they TAUGHT everyone to code in school they still wouldn't know how. How many adults can still do high school math?

Cars have become universal, yet the ability to fix them decreases with every generation.

Decent example. But I'd have just pointed at pants. Those are ubiquitous too. How many people do you know that could make a pair? Or even repair them? Hem them? Replace a button?

I think the average persons ability to write code will be on par with their expertise with clothing. They'll be able to get dressed, and that's about it.

Literacy is nearly universal now, would anyone assert that people write better now than they did 100 years ago?

Um. Bad example. I'd say that is an unqualified: 'yes'.

it just takes to much time/effort for hobbyists to make something comparably decent that doesn't look like crap compared to professionally produced levels.

Well... Mario Maker launched last week.... I'd say the number of average people designing game levels has quite possibly spiked to an all time high. ;)

http://supermariomaker.nintend...

Re:Prima facie ridiculous

[exomondo]

Decent example. But I'd have just pointed at pants. Those are ubiquitous too. How many people do you know that could make a pair? Or even repair them? Hem them? Replace a button?

I think the average persons ability to write code will be on par with their expertise with clothing. They'll be able to get dressed, and that's about it.

That is exactly right! Lots of people love clothes and fashion, it's ubiquitous like technology...however very few people are interested in (or capable of) actually making clothes.

Re:Prima facie ridiculous

[turbidostato]

"Cars have become universal, yet the ability to fix them decreases with every generation."

I basically agree with you, but your comparation is wrong. The ability to fix cars decreases with every generation among other things, because cars are -on purpose, no less, more and more difficult to fix, not even by amateurs but for professionals too, the last iteration trying to lock you out not only technically but legally too.

This could potentially be different about programing:
1) The computer world could still be more and more open.
2) From a very low level, all people do is programing (aplying logics to data sets to get an intended output), so it would only take for everybody to program to first, understand this fact and, second, the will to do it.

Of course, it won't happen because the capitalism works agains point 1 above and lack of curiosity and will works against two.

Re:Prima facie ridiculous

[PRMan]

And most Access applications turned into absolute nightmares for the company as they outgrew the toy database.

Re:Prima facie ridiculous

[PRMan]

Nintendo just took down all the previous hobbyist videos where they were making levels with an engine unrelated to Nintendo.

you'll be the API

[tommeke100]

and code will learn you.

Re:you'll be the API

[ADRA]

Gold star sir/madem, you made my laugh for the day!

Writer's cramp

[Impy the Impiuos Imp]

I assume it will be like now, when everybody can write, only less.

What kind of future....

[Dcnjoe60]

If we have AI good enough to drive a car by itself, then surely, we will have AI good enough to access the online services that the article is talking about. Instead of a future where everybody knows how to code, wouldn't it be better to create one where nobody needs to code?

Re:What kind of future....

[ADRA]

Code is just an expression of thought, and regardless of how that is expressed, you'll have people chaining together multiple facets of information and processes to make something coherent. There's hundreds if not thousands of DSL's which express coding constructs into something a little more akin to a basic human comprehension. Maybe we'll just be able to talk to computers to have the information 'figured out' based on my needs in a universal way, but we're a long way off from that point IMHO.

Re:What kind of future....

[PopeRatzo]

If we have AI good enough to drive a car by itself, then surely, we will have AI good enough to access the online services that the article is talking about. Instead of a future where everybody knows how to code, wouldn't it be better to create one where nobody needs to code?

I think most of us will live to see a day when we need about as many computer programmers as blacksmiths.

Re:What kind of future....

[youngatheart]

I can't disagree with your observations, but I do have a positive perspective to add. Imagine that you wanted a way to measure wireless signal strength as you deployed your first wireless network in 1997. You'd need to buy special equipment where today, you just download an app. The 1997 person might want a calculator at the gas pump, requiring planning ahead to have the right equipment, but today you can just pull out your phone to use the built-in app.

Today the average end user doesn't buy a newspaper, consumer reviews magazine, music player, tape recorder, map or a camera. Few drive to the book store or the bank. Anyone can buy household goods while they wait in line at the DMV and catch up with family and friends across the world during their lunch break.

Today the average consumer solves thousands of needs by just knowing there is probably a bit of programming to be added to the ubiquitous pocket computer. It's probably not fair to refer to the end consumer's actions as coding or even programming, but the average consumer today changes what that pocket computer is capable of every day without even thinking about how it works.

Re:What kind of future....

[narcc]

What makes you think we're any closer to Kuzweil's delusional fantasy than were 60 years ago?

WRONG

[TomR teh Pirate]

I've been trying to teach my wife vlookup in Excel and it's futile. API coding? Puleeze

Re:WRONG

[lucm]

I've been working in IT since before most Apple store employees were born, and I have worked with the most obscure and unpleasant APIs IBM's nightmare factories were able to vomit. Yet, I'm baffled by vlookups in Excel.

Re:WRONG

[narcc]

I'd blame the instructor. A lot of people don't realize that teaching is a skill. Just because you understand something doesn't mean you're capable of teaching someone else.

Basic understanding doesn't equate to daily use

[ravenscar]

The fact that a primary education provides a basic understanding of a thing doesn't mean that your're suddenly proficient to the point that you use it, in depth, in your daily life. Even if you could, it doesn't mean you'd want to.

Maybe I'm wrong.
People get a basic understanding of Biology so they don't need doctors.
People get a basic understanding of Chemistry so they just purchase elements and make their own chemical compounds (who buys soap when you can make it?).
People (might) get a basic understanding of music so they simply put on their own performances.
Right?

Re:Basic understanding doesn't equate to daily use

[fermion]

I recall using the MS DOS and AppleDOS hooks and pokes to do some pretty fancy things at the time. Frankly, the ease of programming from about 1980 to 2000 increased so quickly that it became difficult for minimally educated people, even if they were talented, to make a living at it. There were simply too many people around willing to code for nothing, and the tools to make sure those less qualified people did a reasonable job became increasingly effective. For that matter, gates became so much cheaper than people, that is did not matter what crappy code was running, there were cycles and memory to spare. Which is to say we are already in a world where APIs have trumped coding. Being a software developer is more about understanding process, understanding how to take those bits that are the API and putting them together into a coherent package that will perform a customized function. This is a hard skill. This is why people mostly download a file for their 3D printer that is almost what they want instead of designing one themselves. After all, Sketchup Make is free. So what we are really talking about are the people who can start with a blank sheet of paper, study the available resources, and spec out a process that will transform data in a human usable form for a particular purpose. We tend to call these people engineers and easy APIs will definitely make them more efficient without having to learn how to interpret intelligible error messages. What I see is the problem are people without process skills using these things and thinking that what they get out is a reflection of reality. One can imagine an unskilled business person push data into some analysis program, making a subtle mistake, and bankrupting the company. We now that bussinesses have little analytical talent, that is why Enron was able to fool the so-called professionals in New York and California for some many year, while the engineers in Texas saw what was going on from day one and stayed away.

Re:Basic understanding doesn't equate to daily use

[TuringTest]

> Basic understanding doesn't equate to daily use

That's what medieval monks said with respect to everybody being able to read and write.

Re:Basic understanding doesn't equate to daily use

[thegarbz]

You don't become proficient but you gain a basic understanding which is critical in performing some functions.

People get a basic understanding of Biology so they know when they don't need to see a doctor, so they understand that anti-biotics won't cure their common cold, and that a homeopathic remedy is not a remedy at all.

People get a basic understanding of Chemistry so they know that dihydrogenmonoxide is not dangerous to ingest despite being a fracking compound.

People (might) get a basic understanding of music so they can appreciate more the fine art of music, why someone did a key change where they did, etc.

By extension People may learn to code so they can solve a simple logic problem, put together a funky excel formula (dare I say VBScript?). Not everyone is going to be a programmer capable of throwing together a device driver.

APIs and business models

[Anonymous Coward]

Website and services have been moving *away* from giving away API access for free. Twitter doesn't want third-party devs to write Twitter clients; they want to control the experience end-to-end. There's a few reasons off the top of my head:

• Companies can break private APIs at any time; public APIs are expected to remain stable and maintained
• Companies that rely on advertising want to be able to insert ads, or prevent clients from stripping out ads (e.g. promoted tweets)
• Companies don't want to dilute the value of their service. It'd be cool if I could build my own traffic alert system using the same data as Google Now, even if I have to pay for it, but Google would rather have you stick solely to Google devices, apps, and websites

In short, companies like Apple and Google and others would frequently rather build their own apps rather than allow third parties to "mash up" or build innovative new apps using their services and data (which in many cases is really the user's data).

Re:APIs and business models

[Oligonicella]

Nice ambiguous word, that "data". Very IT-centric. Now please describe how Amazon or JC Penny say, divides their products (their data) up a hundred ways and sells each item multiple times. Your definition of "data" is a very small subset of what 'everyone who can code' wants access to I do believe.

Maintainability is THE Bottleneck

[Tablizer]

Amen! The bottleneck of coding is not getting a program to "work", but making it maintainable: easy to add or change requirements and easy to study (read) to fix.

I've inherited "amateur" programs before, and they are often just not worth it. It was often less total resources to start from scratch.

Now, I'm fine with small personal automation utilities, but it shouldn't be expected that if the author (newbie programmer) moves on to another job, that a regular programmer should be expected to maintain it.

If it has shared utility or data, then it's best to have an experienced professional code it.

(Except, maybe in the few cases that the "professional" is so big of a jerk that the amateur is the better alternative.)

Re:Maintainability is THE Bottleneck

[plover]

Exactly. The difference between design and code is the difference between thermoplastic engineering and assembling a house out of Lego. And managing the dependencies means the difference between a proven clean API and code that has no chance of change short of total replacement.

A future where everyone is a skript kiddie is one where every damn product will become completely fragile. You won't be able to buy a reliable alarm clock or a refrigerator that keeps your food consistently cold, because some stupid interaction with a non-essential system will cause them to break.

You can have my cloud when you can blow it from my cold, damp fingers.

Re:Maintainability is THE Bottleneck

[Darinbob]

We had the problem back when Visual Basic came out, where every non-coding manager learned that coding was simple and that all the programmers were sandbagging and overpaid.

Re:Maintainability is THE Bottleneck

[ILongForDarkness]

It is the exact problem I have with some "free" solutions.There are 1000 people that ever downloaded the thing with maybe 100 current users and 3 contributors. It either does what you want or you need to take the time to understand their code, or you need to keep looking. Script kiddies aren't just caused by amateur coders, they are caused by side projects we make without a proper business need/case. We make it to fill such a niche need that you either have to live with the limitations or move on to another solution. That is what the future will be: a google image search effectively of "hot girl" and countless hours spent finding the one you meant, vs hiring someone with the skill to make yet another "script kiddie' that does exactly what you want, or spending the time to actually learn the market and hiring a professional to make something that is generally applicable and/or innovative.

Re:Maintainability is THE Bottleneck

[TuringTest]

If everyone could code, there would be no need for 1000 people to download a program made by someone else, they would buld their own from scratch. Therefore reusability would be not as important as it is now.

Re:Maintainability is THE Bottleneck

[K. S. Kyosuke]

Even worse, the computing industry is most likely not interested in the "democratization" of programming. Microsoft et al. have no reason to commit a business suicide and make themselves obsolete by simplifying anything related to everyday computing. After all, having people grown dependent on their current software offer has turned them into one of the largest software companies in the world. Clearly from the business perspective, having your users self-sufficient is a horrible idea.

Re:Maintainability is THE Bottleneck

[K. S. Kyosuke]

I've inherited "amateur" programs before, and they are often just not worth it. It was often less total resources to start from scratch.

As opposed to "professional" programs, which are...often the same!? ;)

Re:Maintainability is THE Bottleneck

[ILongForDarkness]

Well there is a time issue too. At my current work I make software for accountants, in the past I made it for doctors and scientists either group their time is better spent practicing their expertise not making software. Just because you "can" do something doesn't mean your time is well spent doing it, or that you can do it as well as someone that does it all the time. Sure I could install my new windows, but I'd rather pay a guy to do it because I don't like the work, it would take me at least 4X as long to do so and I'm more likely to screw something up than they are.

Re:Maintainability is THE Bottleneck

[TuringTest]

The kind of coding that a doctor may benefit from is akin to listening to an event in a data stream or setting an alarm, not building a fancy new application from scratch. You should not need to hire a professional clockmaker to wake up in the morning.

Re:Maintainability is THE Bottleneck

[ILongForDarkness]

I don't understand what your saying. Depends on the doctor/field. In my case oncology so needed tracking different stages of the patient workflow, reports for amount of dose delivered to different types of patients etc. No really good off the shelf solution handles all the parts needed since generally speaking you are using a collection of different vendors CT scanners, EMR, PACS, treatment planning, patent scheduling and radiation treatment machines. Heck even software from the same vendor often does things differently because the Phillips Imaging department doesn't talk to the Phillips Oncology department before designing their software.

Re:Maintainability is THE Bottleneck

[TuringTest]

Maybe the doctor just needs a small script that saves a copy of each mail received into a folder that matches the a keyword in the mail subject, and backup all them to an external hard disk every night.

Not all tasks that merit automating are related to the expert's knowledge domain - most are routine actions that the user needs to make by hand every single time. The kind of repetitive tasks that a developer would create a simple script for, an end user needs to ask the software vendor to include them as a feature.

You're agreeing with me with mentioning incompatible apps - users often need to painfully transfer data between them with Copy-Paste, the universal data exchange standard that glues the whole User space together. If end users knew scripting, they could create their own transfer subroutines that automated most of their repetitive steps that they must perform by hand now.

Re:Maintainability is THE Bottleneck

[Tablizer]

You have to give VB credit for taking GUI's from an arcane dark art to something intuitive and approachable. For "non-enterprise" or local applications, you didn't need the fine control over every pixel that say C++ gave you.

The same thing now requires 4 different languages (HTML, CSS, JS, server-side prog.) and layers of goofy API's, and everybody does it differently.

Re:Maintainability is THE Bottleneck

[Darinbob]

Things removed the dark art long before then. Maybe it was awful on Windows, but everything was awful on Windows.

Nonsense

[brunes69]

Sure, just like when they teached automotive repair in high school, every student left repairing their own cars!

And when they teached shop in middle school, every student left being able to build their own houses! Magnificent!

Just because something is force-taught to kids, does not mean the majority will pick it up after school. The vast majority will take their C+ or B-, move onto the next class, and forget everything they learned before the year is even over.

I agree that giving the kids the option to learn coding as part of school would be a step forward. So would bringing back shop and art class. But please let's not live in a dreamworld where every kid LOVES coding - it isn't going to happen.

Re:Nonsense

[Jawnn]

Sure, just like when they teached automotive repair in high school, every student left repairing their own cars!

And when they teached shop in middle school, every student left being able to build their own houses! Magnificent!

How about we bring back basic grammar lessons, and see that it is teached as well? Normally, I love such irony, but this just makes me sad.

Re:Nonsense

[Oligonicella]

Wonderful example of why this won't work. You point out a trivial typo, yet you understood what was meant. An API won't.

Want to know what it's going to be like?

[Opportunist]

Two words: Visual Basic.

That's what it's going to be like. Every shitty webpage, just like every shitty office suite in the past, will come with its tailor-made API that allows people to develop basic cargo-cult, copy/paste "programming" skills. People will just do that. And that's basically going to be it.

Nothing will change. People will look at something that does what they want to do, copy the code from there (usually without attributing the source or contributing anything to it) and slap together an ugly, frankenstein-esque monstrosity that will more by accident and chance produce something that, provided no special case happens, do more or less what they intend to do, with more or less accuracy.

In other words, nothing new to see here, please move along.

Re:Want to know what it's going to be like?

[TuringTest]

You said all that as if it was a bad thing.

When everyone can code?

[Jawnn]

When would that be, Al? Come on. Put down the pipe and back away. Most people can't even figure out how to use their fucking turn signals, and you think they'll somehow magically become programmers.

Apps and APIs

[Todd Knarr]

I think there'll still be apps, but things will evolve to having apps in parallel with APIs. For example, for an appointment-management service there'll be an app (or more than one, eg. a webapp for desktop use and a mobile app or mobile-optimized webapp for phone/tablet use) for customers who just want to let clients make appointments and want to be able to manage those appointments and don't need a lot of customization. There'll also be API access that would let customers get at the basic operations (with appropriate authentication and filtering for safety) that the apps themselves use so more sophisticated customers can use the service's functionality through a custom interface. It won't be an exclusive-or situation either, a customer may have their Web developer use the API to integrate setting up appointments into their own client web site while their staff use the standard webapp to manage appointments.

Neither the app nor the API will likely be free, either. The customer's clients won't pay in the above scenario, but aside from a trial period the customer would have to pay for access to the service (the amount probably depending on their usage and what features they want access to). We're about to hit a point where merely being able to get a huge userbase won't be considered valuable, you'll need to have a feasible roadmap for getting a revenue stream from your users. A thousand users each paying a monthly subscription will be more valuable than a million users not paying you anything. And frankly it's a lot easier to maintain paying users if you start them out as paying users.

What do you mean "when"?

[hack slash]

There are always going to be people who are, to put it bluntly, thick as shit and have absolutely no aspirations to increase their intelligence level.

Exactly when is "everyone" going to code?

[mariox19]

This isn't an exact analogy, but calculus is more than 250 years old, and it's not like everyone is doing calculus. In fact, never mind calculus: there are plenty of people who, though they have sat in an algebra class, don't get even rudimentary algebra. So, why are we imagining that someday everyone is going to code?

Re:Exactly when is "everyone" going to code?

[narcc]

Because programming is far simpler than even basic algebra.

We teach reading, writing, and arithmetic as well. An absurd majority of students manage to graduate both literate and numerate.

Re:Exactly when is "everyone" going to code?

[frank_adrian314159]

Because programming is far simpler than even basic algebra.

OK. I guess in your world, it is. Please tell us where this amazing world exists. In my world, programming is a slog. And there's enough trivia to know that anyone who could learn it all would do much better memorizing world capitals and becoming the next Ken Jennings on Jepoardy! But that's my world.

In case you didn't notice, this post contains sarcasm.

Predicting the future

[WaffleMonster]

When everyone can code there will be no such thing as a "web service" API because enough people would have been smart enough to spend their time developing useful transports rather than always relying on the lowest common denominator/lowest hanging fruit (e.g. http).

When everyone can code APIs will be designed properly as abstract interfaces that will not break every time something changes.

When everyone can code people will leverage the Internet to communicate directly among themselves as they see fit on their own terms. They will refuse to be limited, harassed, spied upon, sold and censored by massive content companies (Google, facebook, linkedin, twitter...etc) which currently are relied upon to facilitate most communication.

When everyone can code all of the mobile platforms will be completely open and free with no restrictions. They will maximally protect the interests of the user rather than interests of vendors and app vendors.

When everyone can code entire classes of vulnerabilities will become impossible due to sufficient effort having been spent on constraint based systems to avoid and detect problems with certainty.

When everyone can code DSLs will have displaced general purpose languages for most coders allowing them to operate more efficiently in their domains of interest. Coders will be grateful rather than whine about having to learn yet another language.

When everyone can code hell will have frozen completely over and re-purposed as a skating rink for angels.

Re:Predicting the future

[Lennie]

"When everyone can code there will be no such thing as a "web service" API because enough people would have been smart enough to spend their time developing useful transports rather than always relying on the lowest common denominator/lowest hanging fruit (e.g. http)."

I think this is an interesting discussion, but only if you are looking at the future. Because I'm much more optimistic about the future of http.

First off, yes it would be good to have certain applications use their own protocol.

But having a standardized protocol really helps. Especially if you combine it with something like REST and json and use standard libraries (like libcurl) for that.

http is stateless which means you can easily proxy it.

Now http 1.1 isn't a great protocol, it's old.

But http 2 solved a bunch of parts which needed to be solved like:
  less overhead (for example for the headers)
  no head of line blocking
  multiple streams in the same connection

It didn't get a websocket protocol though, so no tcp like behaviour. Obviously the websocket protocol itself can already do that.

Why would you want something like websocket ?

Because the people who make the standards are working on improving the underlying layers for https (tls) and tcp as well.

Look at what Google has done with quic:
  zero roundtrip connections (thus better than plain tcp, it can do that because it uses udp)
  multipath support
  always uses (datagram)tls so it uses security

The tls 1.3 specification which the ietf (which makes protocols like tcp and http) is adding the pieces needed for a quic like protocol.

I think in a few years a http 2 over quic like tls 1.3 will be in widespread use.

To the developer it looks like http 1.1 but underlying transport will be secure, low latency and fast and it will be part of the library you use.

When you look at what the dns workinggroup is doing, I wouldn't be surprised if they adopt tls 1.3 as a transport well. Because they want to get privacy for the dns protocol as well.

Now if you think we need a more peer2peer and distributed model then I agree. But if you look at the most popular peer2peer application, bittorrent, is even inspired and partly based on http.

If you combine namecoin and some of the dnssec protocol then you can have:
  no central root (for at least part of the namespace, namecoin uses .bit )
  distributed blockchain based dns names
  certificated tied to dns names
  certificate validation by the client library when connecting to a http 2 host
  local validation of dnssec information which can be used as an anchor to store your public keys

What is the result of all of the above ?:
  no use for Certificate Authorities other than maybe 'green bar' if you really want it
  no icann
  standardized protocols by people agreeing and writing things down
  possibility to use http semantics for web and distributed or peer2peer protocol without the overhead
  secure by default because of encryption
  support for multipath and low latency communication
  open source libraries

PS I wrote all the abbreviations in non-caps and removed all the dashes, because I encountered the lameness filter

Never going to happen....

[bobbied]

There will never be a time when specialized programmers are unnecessary.... NEVER...

Fredrick Brooks, "The Mythical Man Month" needs to be consulted by those who think otherwise... His sage advice is a true today as it was in the 70's when he wrote that book, all that's changed is the names on the technologies.

Yea, you might have a rash of folks who can drive the mouse around, but you will need the specialized skills working behind that UI to keep it all working...

Sure

[CanadianMacFan]

Because it worked out so well when we got everyone "programming" their own stuff in Excel and Access.

Re:Sure

[narcc]

The spreadsheet was empowering. It's what drove the microcomputer in to the business world. It's still an irreplaceable tool today. For every overgrown nightmare workbook, there are millions of others empowering users and driving the companies forward.

Access is what allowed many small businesses to succeed well enough to outgrow all those "awful" Access databases you hear horror stories about. I've replaced quite a few over the years. Some far worse than others. Still, without those ugly kludges, the companies that depended on them would have gone belly-up long before I came along.

Computing for the masses was the rallying cry for computer hobbyists back in the 70's. It's the reason that you have a home computer. Tools like the spreadsheet and simple languages like BASIC took the power away from the computing priesthood (with their over-priced time-sharing services) and put it in the hands of everyday users. That's a very positive thing.

Would you really rather cede control back to an elite few? Have all your computing needs handled by some third-party for hire? Why would you deny others the ability to become empowered and take control of their computers?

People aren't going to code

[CanadianMacFan]

So many people today have the ability to take videos. And yes they do a bit for personal use but they mostly stick with pictures. However we are still consumers of video content for most of the time. Why would this be different with computers? I'm not going to write every app I need from a set of APIs. How many of us do that now? The average person is still going to get a game with the great graphics and physics engine from some company. You aren't going to put that together yourself with a bunch of API calls.

Re:Lock-in and dependence

[youngatheart]

1) Each website will inevitably have its own API that is incompatible with every other similar service. With service APIs, no one will bother with standardization on anything but the base protocol, because no generic standard satisfies any individual service's full functionality.

That's pretty much a description of how web pages worked just fifteen years ago. Nobody misses the days when most websites were designed for IE or Netscape and you couldn't make a website that worked well in both without taking quite a few special behaviors into account. Today, Microsoft is pushing Edge, Chrome has a significant market share, and Firefox's trailblazing changes made it normal for a browser to use tabs and block pop-ups by default. Not only has standard behavior improved for the better, the industry leaders are all trying to adhere to standards.

Your instant messaging providers would still be incompatible with each other, for instance, because company A wants a whiteboard mode and company B wants custom emoticons.

Recognizing that is the state of things doesn't mean it will stay that way forever. Consider how many websites have realized that managing their own authentication process is a bad idea. There are still plenty of problems, but it is becoming normal to use an inter-operable system. Right now trapping people into a single messaging system is a business model, but I don't think it will stay that way.

2) Because it's easier/cheaper/faster, everything will depend on some company being in business and providing their service. If somebody makes a service that does text-to-speech extraordinarily well, for instance, and does it as an online API call instead of locally on the device, then if that company goes out of business the service dies with them and everything that depended on their particular text-to-speech engine would stop working.

Unless it's an open source service, then five more will spring up in it's place. If that text-to-speech service is widely desired, an open source version is guaranteed to spring up eventually.

Wait a moment... aren't these types of problems happening today? Business as usual then. Carry on.

Business as usual is widely panned, but we already live in a world where business as usual has radically changed in just a couple decades. I carry a computer in my pocket that is massively more powerful than I could have predicted twenty years ago. Yesterday I talked to it and got an answer. When I was a kid, I couldn't have gotten that answer without a trip to the library and possibly weeks of waiting for the library to get the book exchanged from another geographically distant library.

Business as usual for the average person is change, not measured in generations or even decades, but measured in months.

A presumption in search of an article

[Oligonicella]

But when coding becomes universal, so will the expectation that websites become accessible to more than just browsers.

Now tell me why, as a designer/owner of a website, I should give a shit if someone wants me to provide something other than the web page I've provided already. No one shopping on Amazon - or reading one of my fables - has any need (and probably much less desire to code) accesses to the data otherwise. Make the argument something more substantial than "Information wants to be free, man."

Re:Lock-in and dependence

[Oligonicella]

Unless it's an open source service, then five more will spring up in it's place. If that text-to-speech service is widely desired, an open source version is guaranteed to spring up eventually.

Which will then have their own twists to the API - and the service will still be down and gone (and probably discarded) in that "eventually" you mention. You didn't solve the problem, you just created an never ending series of speed bumps.

..."When everyone can code//"

[nult]

That day will never come. That's how the world works. People are getting progressively lazier and its pretty long stretch to say that there will ever be a point when "everyone can code"...

How about just teaching kids to reason?

[emorning]

Baby steps. Young people would be much better served by learning to brainstorm, reason, and think abstractly. And those basic skills will server them whatever they do. At most, teach coding as a way to obtain these basic skills, but not as an end to itself.

sample bias

[Georules]

I'm sure this future feels a lot closer in San Francisco.

I wanna be a minority

[tepples]

But I'd have just pointed at pants. Those are ubiquitous too. How many people do you know that could make a pair? Or even repair them? Hem them? Replace a button?

Then I guess I must be in a minority. I've replaced buttons on store-bought clothes. I've even made my own pants, but they have a drawstring closure instead of a button closure because I haven't learned buttonholes yet. And before I learned pants, I was making ankle-length shirts to wear.

Well... Mario Maker launched last week.... I'd say the number of average people designing game levels has quite possibly spiked to an all time high. ;)

For one thing, this is Nintendo catching up to where Sony was years ago with LittleBigPlanet. For another, around the launch of Super Mario Maker, Nintendo went on a DMCA takedown streak on YouTube, handing out copyright strikes to uploaders of TAS and Kaizo videos.

Re:I wanna be a minority

[vux984]

Then I guess I must be in a minority.

Yes. You can't possibly think you are not?!

For one thing, this is Nintendo catching up to where Sony was years ago with LittleBigPlanet

N may be playing catchup, but that doesn't take away from the fact that there were probably more people making game levels this last weekend than ever before. LBP was good, but I don't think it had the launch Mario Maker did.

For another, around the launch of Super Mario Maker, Nintendo went on a DMCA takedown streak on YouTube, handing out copyright strikes to uploaders of TAS and Kaizo videos.

It just points to N trying to steer any interest in making mario levels towards its new game. Its a dick move, but not surprising, especially for N.

I'm waiting for when every programmer can code

[raymorris]

Yeah I'll get excited when most paid programmers can code. I'd say about 25% are reasonably competent and 10% are actually good. I too have done a lot of rewriting.

One case was - the guy was a pretty good software architect, and a terrible coder. The overall system worked well and the structure made sense. It was extensible. But if you looked inside any function, the lines of code were all kinds of bad. They didn't know about 80% of the errors their code threw only because the other errors hid them. For example, an important hourly,cron job was full of error-ridden code, but they didn't get the error messages because the cron job hadn't run for five months.

Management has no way of knowing how much technical debt they've accumulated, so they don't know - can't know - that a certain programmer put them $150,000 further in the hole.

Re:I know how to change my oil.

[tepples]

Yet not everyone becomes a writer

Not everyone becomes a published writer, but a knowledge worker needs to write in order to communicate with co-workers.

Re:Will never happen

[tepples]

Shit, everyone has a browser that supports JavaScript and everyone could start programming TODAY with zero entry cost, if they wanted to.

Does Safari for iPad support user-entered JavaScript, or just JavaScript downloaded from an external web server?

iDon't measure wireless signal strength

[tepples]

Imagine that you wanted a way to measure wireless signal strength as you deployed your first wireless network in 1997. You'd need to buy special equipment where today, you just download an app.

Since when does iOS have a public API to determine signal strength? Last time I checked, it didn't, and Apple's App Store forbids use of non-public APIs. This is why WiFi-Where was removed. iPhone and iPad users end up needing to buy "special equipment" in the form of an Android phone because unlike iOS, Android has public Wi-Fi APIs.

Re:iDon't measure wireless signal strength

[youngatheart]

You're right. I actually mentioned it because it was on my mind after discussing the merits of iPhone vs Android at work. It's one of the things I appreciate after switching from iOS to Android.

Oh for christ's sake

[JustAnotherOldGuy]

"What the Future Will Be Like When Everyone Can Code"

This is NEVER going to happen. Stop trying to make it happen. It doesn't NEED to happen. FFS, not everyone needs to know how to "code".

Replace the word "Code" in the title with nearly any activity and you'll immediately see how fucking stupid it is. For example:

What the Future Will Be Like When Everyone Can Wire A House
What the Future Will Be Like When Everyone Can Ride A Unicycle
What the Future Will Be Like When Everyone Can Crochet
What the Future Will Be Like When Everyone Can Mow The Lawn
What the Future Will Be Like When Everyone Can Adjust A Carburetor
What the Future Will Be Like When Everyone Can Solder
What the Future Will Be Like When Everyone Can Read An X-Ray

Enough already. Please stop with this delusion that everyone needs to code or even wants to.

Re:Oh for christ's sake

[thegarbz]

All of your examples are very specific. The discussion around "coding" is not. For the record everyone should in theory be able to wire a house due to their knowledge of basic high-school physics. Coding is the same.

Not everyone will be able to throw together an operating system, but it would be nice to be in a world where someone could apply simple logic, and god forbid tie two things together to produce something usable. That does NOT mean we're cranking out proficient C# experts in highschool.

Re:Oh for christ's sake

[JustAnotherOldGuy]

For the record everyone should in theory be able to wire a house due to their knowledge of basic high-school physics.

This is utter bullshit, and only an imbecile would dare to make such a stupid claim.

Re:Oh for christ's sake

[thegarbz]

Not at all. The complete sum total of understanding of electrical principles are taught as V=IR, and P=IV. Physics classes then go into what happens when too much current flows, how current is divided among paths, how voltages work etc.

The rest of it is reading and complying with a set of wiring codes. The fundamental understanding is taught in school, the rest is following a very VERY prescriptive set of instructions, and by logical extension many of the electricians I know didn't even finish school, but just followed instructions from someone who has read a manual written to be understandable by a high-school dropout.

No really. I've seen electricians who don't know why they use circuit breakers of certain types, or how a safety switch works, just that they open up to a section of the rules that says for the lights use this cable put this circuit breaker on it and you're all sweet. I like even better the old and very experienced electrician who wired up my house, a guy who learnt from a guy, who learnt from a guy, and who somewhere along the line didn't realise that standards change, ovens don't need accessible isolators just stoves, and who when he finished wiring up the downstairs of my house I could tell him exactly which year he finished his apprenticeship and that he really ought to read the current edition of the codes.

That's not learning, that's monkey see monkey do. The learning stopped at school.

Re:Oh for christ's sake

[me3head]

I concur. It would be a better world if most people learned how to do those things.

Re:Oh for christ's sake

[JustAnotherOldGuy]

Not at all. The complete sum total of understanding of electrical principles are taught as V=IR, and P=IV.

The rest of it is reading and complying with a set of wiring codes. The fundamental understanding is taught in school, the rest is following a very VERY prescriptive set of instructions, and by logical extension many of the electricians I know didn't even finish school, but just followed instructions from someone who has read a manual written to be understandable by a high-school dropout.

That must be why there's no licensing of electricians, because anyone can do it as long as they had "basic high-school physics". And that's why there are no tests for the non-existent electrician's license!

Because there's no real knowledge involved, all ya gotta do is just look at all the purty lines and number-thingys on the schematical doohickey and you too can wire a sky-scraper! Easy peasy, baby!

By your logic, all you need to troubleshoot any electrical device is to know V=IR and P=IV. As someone with 30+ years in the electronics field, I can tell you that your idea is nonsense of the first order.

Re:Oh for christ's sake

[thegarbz]

Don't confuse capability and licensing. Licensing is an attempt at regulation and to ensure people follow the rules given. Yes electricians need to be licensed. I am not, yet I know more about how to wire up a house than most who are. Oh also the test are a joke with a very lame punchline, and you can see that quite clearly from the incredibly amount of crap and non-compliant work that generally goes on.

And you're part right. These is knowledge, just not very much. That's the wonderful thing about a prescriptive standard, you don't need to know how or why that 2.5mm^2 cable should not be run in the same duct as a ULV comms cable, you just need to read the page that has among other things, drawings for the really stupid people showing 100mm gap, barrier, etc.

Also as someone in the electronics field you should know that V=IR and P=IV is all there really is until you get to frequencies where things are better explained by Maxwell's equations (such as antennas). Inductors, Capacitors, HV, MOSFETs, everything boils down to V=IR and P=IV. The fact that we start introducing time varying integral terms and imaginary power doesn't change the fundamental equation. You may think things get more complicated like the "transistor man" model of a transistor, but that can be simplified to a hybrid-pi model and solved using ohms law.

Speaking of: none of this is at all relevant to wiring a house, which can be done via a glorified flowchart where the conditional statements get as complicated as: is there 3ph air-conditioning, is here more than x number lights, is the circuit longer than x, etc. Open up a set of wiring codes from any country and have a look. There's lovely complicated formulas for resistance and fault loop impedance of subcircuits (I=V/R by the way) for engineers, and right next to it a table or idiots guide listing the common wiring found in a house along with common maximal lengths so that the poor confused electrician doesn't need a calculator and can just think hmmm this cable is only 50m long, rule book says okay.

I stand by my statement. Residential electricians learn no new theory if they have done highschool physics.

I have to yeah/but

[ILongForDarkness]

There is bad code for sure. There is also bad architecture and feature specs. In my experience there are often very fuzzy business requirements, sometimes legitimately: the customer can't tell you what they need they just ask for something "better". Example something I'm currently dealing with: the customer asks to be able to add multiple addresses instead of one to each order. Well is there only one address of a given type allowed (postal, physical, fax, etc) or an arbitrary number of each type? If so how do you know which should be listed first? They aren't sure but they need the feature in a couple months so they can demo it to customers.

So I end up making something as general as possible "just in case" someone somewhere might want to do things that way. Then a few months later I expect someone to complain: we screwed up and added 5 postal addresses to one customer, that make no sense why didn't you warn us?

Anyways, sometimes bad architecture comes about by poorly nailed down design requirements. Sometimes that can be helped by asking the customer more, and sometimes they don't know so are no help All you can do as a dev is tell them it is likely going to be 5X more work to do it in a way that will support all scenarios "just in case" vs doing it the "one true way". If they say fine you need to live with the legacy junk until they feel free to free you up to do a refactor on the code and not spend the whole release trying to pound out new features (fat chance in hell). Sometimes the best you can do is write code that doesn't suck to bad in what the expected typical use case is, and drag it forward as long as you can to meet the current requirements. Once it breaks to the point of no return you have to pay all the tax of the lack of good specs you had all along. This doesn't excuse crappy coding at the method level though: that is a matter of skill and hopefully anyone that considers themselves a professional continues to learn from code reviews, training etc and gets better over time. If you don't you suck and need to go.

Re: I have to yeah/but

[prefec2]

The customer never knows what he wants and even more they do not know what they need. That is why they hire software developers to do that. If they really knew what they need, they could just write it themselves. That is why we need social skills, because we have to be able to understand other humans and translate that into software.

Re:I have to yeah/but

[Tablizer]

So I end up making something as general as possible "just in case" someone somewhere might want to do things that way

I've had similar situations, and often it backfires because the most general also tends to be rather abstract to users. Often one has to make the best guess based on experience, knowing that fuzzy requirements are fairly likely to bite everyone.

The future: when everyone can be a hairstylist

[bigsexyjoe]

Ugg. The future where it's all API's is already here. The fact is learning API's is often a pain in the ass. Maybe most people could code if you told them they would be killed in five years if they didn't learn to code, but no, we are not moving towards everyone being able to code, or everyone even being that computer savvy, or people knowing how to turn their problems into specific algorithms.

Not about coding, about productivity

[justhatched]

If you cannot automate tasks as part of your job, someone else who can will be more cost effective, or someone else will fully automate your job. Either way, I imagine being able to script some automation together will be an essential part of being able to gain work in the future, especially as so much of general work consists of repetitive tasks.

Sites like iftt and other introductions to scripting solutions, make this type of automation easy for the general population to get the building blocks of automation without the overheads of making a request to an IT specialist.

How many programmers work on mail merge or the many other office functions that have been solved many times over already? Yes, just as today, there will be some spreadsheets from hell and some vb messes, but the same productivity pressures will retain more of the capable ones than the mess generators.

Middle managers creating their own crap scripts can still have a huge impact on their productivity, and this is an area that some traditional IT departments have not supported very well other than to try to lock it down, albeit with some good reasons.

Holy Shit Will That Not Happen

[dcollins]

I hear some ludicrous predictions for the future all the time. This pretty much takes the cake. There is no concern whatsoever that all or a majority of people could possibly be proficient enough programmers to interface usefully with arbitrary API's from day to day.

Recall that the majority of people cannot pass a basic algebra class. Just the idea of a variable is beyond about half of people. I'd be happy if most people just got educated enough about programming to realize how friggin' ludicrous this prediction is.

High School Subject Literacy

[mtippett]

I doubt that we will be awash in coders as a result of high school literacy in code.

This is much the same that a high school curriculum of science, math, language arts does not make the world flush with chemists, engineers, theoretical mathematicians. What it does create is a community that has a level of appreciation, and the potential to specialize in that particular field.

Think of it as having a whole lot of engineers that have become managers. They can still code if they needed to, but generally won't. However, the engineering concepts are known even if they aren't being used.

This article should be rated +5 funny

[chthon]

Enough said.

Also

[Sycraft-fu]

Unless computers become AIs, they will never be able to deal with fuzzy logic and imprecise specifications. As XKCD said " No language will free you from the burden of having to clarify you ideas."

To make a computer program you have to be able to formalize your ideas and present them in a logical format to the computer. Even when you have lots of tools to make the details easier, the fundamental requirement remains. You have to get the flow of your ideas in to a logical format that can become something an imperative device can execute.

Well, some minds are better at that than others. Some people are not very good at the kind of logic and creativity that programming requires. So they aren't going to become programmers, even if there was some real reason for it.

Re:Also

[K. S. Kyosuke]

Now contrast this view with the following...

Many professions require some form of computer programming. Accountants program spreadsheets and word processors; photographers program photo editors; musicians program synthesizers; and professional programmers instruct plain computers. Programming has become a required skill. Yet programming is more than just a vocational skill. Indeed, good programming is a fun activity, a creative outlet, and a way to express abstract ideas in a tangible form. And designing programs teaches a variety of skills that are important in all kinds of professions: critical reading, analytical thinking, creative synthesis, and attention to detail. We therefore believe that the study of program design deserves the same central role in general education as mathematics and English. Or, put more succinctly, everyone should learn how to design programs.

How to Design Programs

Re:Also

[mrchaotica]

I see no conflict between the two perspectives. Sure, everyone should learn how to design programs... but some people fundamentally just aren't capable of clarifying their ideas and thus are doomed to fail at it.

Everyone can code?

[lolococo]

Yeah, right, just like everyone can cook.

What will be the future

[ruir]

...when anybody can service their car. ...when anybody can be a surgeon. ...when anybody can be a hooker...oh wait. Dumb article.

BINGO!

[denzacar]

How many people studied math in school?
Everyone. You can't get through elementary school without math.

How many people use anything more advanced than multiplication and division?
How many of them are actual mathematicians?

Now... How many people HATE math cause they either never had aptitude for it OR due to the way they were taught math?

Besides that... What's the use of a half the humanity of below average programmers?

watch end users to see what they need

[raymorris]

Oh indeed, asking representatives of customers about requirements sure isn't very effective. Even worse is trying to get requirements by asking someome in our department, who asked someone in the users' department, who may or may not have asked the users. I always try hard to actually watch real users actually do the task, and say "show me when might you need multiple addresses ". If it's not possible to get to the users desk, I try to talk to them on the phone get screenshots , etc. I want to understand their process better than they do, so I can build tools which improve their process.

We love to complain about poorly specified requirements, yet we often do a poor job of getting the requirements. Rarely have I been turned down when I asked "can I come see?" We fail to ask "what happens with the information my program gives you? What do you do with it after you run my program?" All too often the answer is "I paste it into a spreadsheet and ... so I can get the information I actually need." Then what do you do next? "I type those numbers into foo.exe." Of course foo.exe just puts the information into the same database that we got the initial data from. The entire process they do each day is a 30-minute way of accomplishing "insert into daily_summary select sum(blah) from items".

Re:watch end users to see what they need

[ILongForDarkness]

Well in my case we are very far from customers. Our software is sold in ~80 countries by distributors and ultimately used by accountants. So addresses might not be the clearest example of this but often we have to make things very generic because one countries tax code differs from another, or they have the same data but need it at different times in their process etc. Often when we ask a distributor they ask a few customers I think and we go with what they say but sometimes no one knows because it is a completely new government requirement/process (ex. recently Australia started requiring their version of a personal retirement savings plan to have yearly audits: not required before so no existing paper based system or whatever to automate, we had to guess what they would need when).

Agreed though, whenever you can actually see a customer working is the best. Even better if you are automating something you do as you know both your current process and what you'd ideally like to have (did that a lot when I worked more on the IT side, or as more of a scientist/technician playing around with accelerators).

That is bull shit propaganda

[Anonymous Coward]

I am so sick of this kind of propaganda that the industry is pumping out. It's complete BS. It's an attempt to drive down labor costs since most companies cringe at the idea of having to pay a technically adept worker what they are worth. Everyday on my ride in to work I have to listen to these stupid radio commercials trying to sell people on IT by simply taking a 6 month online course. Give me a break... If it were that easy, then yes, everyone would probably be a coder or IT professional, but it's not and the industry is changing at such a fast pace that even IT professionals have to keep up to date with industry standards, new technologies and best practices regularly in order to not be left in the dust. This propaganda is nothing more than an ill conceived attempt at driving down labor costs. Companies need to suck it up and pay people what they are worth whether they are coders or not! People claiming to be programmers are a dime a dozen but good programmers are very hard to find.

Re:I know how to change my oil.

[Hognoxious]

You'd be surprised how many can't even file a decent bug report, even if you give them a template with six simple questions on it.

"Oh, I thought I only had to answer one".

Re:Will never happen

[tepples]

why would you buy an iPad if you want to do programming?

I was thinking of the case of someone who already owns an iPad and wants to start doing programming without having to spend hundreds on a new computer.

I am a coder and... no

[GuB-42]

I code, a lot. But I don't need an API for everything. Beside my work, I'm at 99% a regular computer user. If I want to read twitter feeds, I go to twitter.com, why should I mess with the API? My needs aren't so specific that I should write my own code.
And even when I do the remaining 1%, I rarely use website APIs, even when they are available.

Title Should Read

[Afty0r]

"Saddles, Not Sore Feet": What the Future Will Be Like When Everyone Has Leatherworking Skills To Make Their Own Saddles

- Daily Tribune, 15th March 1789

Re:Lock-in and dependence

[Hognoxious]

Each website will inevitably have its own API that is incompatible with every other similar service.

Nothing you can't solve with an adapter API controlled by 47 tons of incomprehensible XML.

Of course, now you have two problems.

https://xkcd.com/927/