Should programmers have to be part of a professional body to practice?

Question

Accountants, lawyers, doctors (and many other professions) have respected professional bodies that are responsible for awarding chartered status to the members. Through this mechanism, they gain trust and respect and can demonstrate that they adhere to a code of standards and ethics, and maintain their skills to keep them current.

Programming has a similarly large body of knowledge and a similar need for study and continual development. It is far younger a discipline than any of the examples quoted, but is of growing consequence to business, the economy, and most areas of life.

Will programming eventually become as respected a profession with a powerful professional body? (Lawyer and accountant jokes aside!) Should we be aiming for this?

Possibly similar to this question.

Answer 1

Some countries already have somehing like this.

In the Netherlands you can be registered. But you need to be invited which (I think) is an error in the system.

I'm not sure if this helps. Besides, if a customer loses faith in a software company, he is free to go elsewhere for another project.

Answer 2

Only if you wish to kill the profession. Credentialism only serves to prevent entry into the field. The only real benefit to members of such organizations is to prevent competition. Look at how many start ups have people with little or no formal training. Would the world be better off without Microsoft, Dell, or Google to name a few?

Answer 3

If you are interested in this topic, I can recommend the book "After The Gold Rush: Creating a True Profession Of Software Engineering". Not by Neil Young, but by Steve McConnell, the same guy that wrote "Code Complete".

Answer 4

Programming is in many ways a design activity. You cannot award a design status. Actually, you can but it won't mean anything.

Absolutely avoid this. Otherwise the industry will be set back through huge bureaucracy.

Answer 5

At first, I was tempted to dismiss the question but when thinking more about it, there is something. Doctors, for example, must follow the Hippocratic Oath. Lawyers will lose their job when they are found to be corrupt.

On the other hand, most developers I've met so far don't really care for networking or having a political body to watch over them. So what's in it for us? Surely, it would be cool if the script kiddies and the crackers would need to go through some kind of assessment process before allowing them to start crashing computers on the other side of the globe. Unfortunately, I don't think that this genie can be pressed back into the bottle.

Therefore, I still think that things like SourceForge and StackOverflow are better suited to our needs: To build a reputation among ourselves and to heck with the rest of the world. If they can't or won't install security patches as we advise, why should we feel mercy for their cries after losing their data?

Answer 6

I once read an amazing article on a nation wide magazine written by a famous economist (in my country) and it summary it said:

• I have a degree in economics.
• I have 3 phds.
• I have a second degree in management.
• I teach a post graduated course at the university XXX.
• I am not registered at the economist professional body (don't know if it's the write translation but you get the point.), so I cannot legally sign any projects. When I have to I just ask one of my STUDENTS to sign for me.

I guess this shows what I think of this matter.

Answer 7

I believe this will happen if and only if:

• Programmer compensation goes up. The average lawyer earns 3-4 times what the average programmer does. That's why so many people apply to law school. There is simply not enough money involved to induce people to study Comp Sci.
• Demand for programmers goes down. Though the current economy has shaken things up a bit, there's still far greater demand than supply for programmers.
• Offshoring is no longer common practice. There are very good reasons to hire a lawyer locally. Those reasons don't apply to programmers, which is one big reason so many programming jobs go overseas.

I honestly don't see any of these happening any time soon. Professional credentials create barriers to entry of the sort we simply don't need right now.

Answer 8

A professional body acts as both a barrier to entry, and a seal of approval. In law, and medicine, (I don't know about the others), the bodies are national.

Given that the IT sector invented the internet, and the transnational spirit that that encourages, I'm not sure if there is space for multiple national bodies to arise.

This would leave space for a single global standards body?

But what would be the point of its existence - a certification body to assert an equivalence of training? From a business point of view, either you are small enough to only be employing people from an education system you understand, or you're big enough to understand multiple systems.

Answer 9

I don't think in the general sense the idea will ever have merit. Consider:

• The discipline is so varied. What makes a good embedded developer has little to do with what makes a good web developer, or shrink wrap developer, or line of business developer, or data base developer, etc.
• what makes a good web developer, for example, is very different what made a good web developer even a few years ago, and accepted practices vary widely from place to place or platform to platform.

However, I feel there is a need/market/desire for a credentialing process for programmers that build systems for use in medical equipment or the space program, to name a couple examples. In short: systems where the software could endanger human life.

The problem here is that as soon as such a process is created, about 50 other places that don't really need this process will start enacting policies that require these programmers: military and other government agencies come to mind, as do things like ISO9001 certification.

Thus the market would drive up prices for these programmers. You'll start seeing a lot of people gaming the system, and if there's one thing programmers are good at, it's gaming these systems.

In the end, the credential wouldn't really matter. You'd have companies hire uncredentialled programmers for positions that require the credential, only to send them to a two week training course teaching them how to pass the credential requirements. Every CS university would have a course in how to do the credentialing. It would be meaningless, and serve only to limit access to the profession.

Answer 10

No. Anybody can download a basic compiler and start programming their very own "Hello, World" program, and nobody is hurt physically, legally, or otherwise by that. Medicine, law, and accounting, to use your examples, all have very real legal or physical consequences when an amateur starts practicing on their own without proper training and licensing.

There's a pretty obvious difference in consequence between an untrained, unlicensed programmer writing code and an unlicensed doctor performing surgery or an untrained lawyer representing somebody in court. The untrained programmer might not write the best code, but it's not very likely that their customer is going to wind up dead or in prison as a result.

Answer 11

No. We don't even have a single standard of practice--how could you possibly enforce one?

Answer 12

The accreditation process would simply not be able to keep up with the field...

Answer 13

So only graduates of Java Schools would be allowed to program?
Or you could have the industry run it, so a few $$$ to MSFT for your VBA professional engineer certificate and you're all set.
It would get rid of Linux, Apache, C, the Web etc.

Unless of course it was possible to write code in other countries. You could call it "out-shoring", or "off-shoring" or something!

Answer 14

Accountants, lawyers, doctors (and many other professions) have respected professional bodies that are responsible for awarding chartered status to the members. Through this mechanism, they gain trust and respect and can demonstrate that they adhere to a code of standards and ethics, and maintain their skills to keep them current.

The truth is that through this mechanism, entrenched interests create barriers to entry in a market that they want to continue to control.

The way to establish trust and respect is through the competition mechanism of the free market. Programmers who meet needs will be rewarded; those that suck or harm or defraud their clients will go out of business

As long as somebody's willing to continue to pay for a particular programming service, it obviously meets their needs to their satisfaction. Businesses don't need a programmers' guild telling them whether or not their programmers meet their needs.

Answer 15

I think there is a vast difference in the manner professions like Accountants, Doctors work and the way the software profession works (I better not comment on the legal profession). Doctors and Accountants are a methodical lot. The community has grown on the foundation of methodical practice.

Its not that Software profession has not tried to build methodologies but then it has evolved more on the creative side where programmers have found it best to break the rules and get things done. Also the element of risk as it is apparent in the accountancy or medical profession is not easily visible in the software profession. Programmers hardly think of risk before they start building things. The assumption mostly is that it can be changed later if it doesn't get the job done.

That said I think software must as a first step try to emulate its better organized cousin - the hardware industry. That is where we better be aiming at.

Answer 16

I'm going to be contrary and say that a professional accrediting body would do a lot for the field of software engineering and move it from software soloing, with the bugs we know and love, towards creating an actual professionalism and engineering discipline. Yes, that serves to keep out the hyper-independent.

Frankly, most of the complaints I read on SO stem from a lack of professional skill and ignorance about other people. An accrediting body, properly set up, would serve as a filter to keep the semi-competent and hacks out, and ensure solid engineering practices taught and used.

Answer 17

Professional bodies are always started with the best intent of ensuring that certain level of knowledge, quality and ethics will be professed by the members of certain occupation.

The problem is that there's essentially no external control over such organizations. After all, there's no else suited to govern the profession than the people that are most knowledgeable in it. Yet these same people are the most likely to be biased and put the interest of their guild above the interest of the society.

If left unchecked, professional bodies quickly erode in a protectionist organisations that are only aiming at ensuring the demand for professionals is bigger than the supply by creating artificial barriers to entry and raising the cost. In essence, they become an elitist type of unionized organization, aiming to protect the status quo, maximise the profit of their members, and prevent the profession from commoditization.

Answer 18

I despise professional licenses (required by law). Their sole purpose is to deny competition and dramatically increase costs. If I have a cold or flu, I don’t need to see a doctor (around $200), just give me somebody that will prescribe this year’s batch of medication. After all, the doctor is just prescribing what the salesperson told them to prescribe this year. If I have a friend that is willing to do all the grunt work of going to small claims court for me, then I should be able to pay them to do it rather than a lawyer (around $1000). They shouldn’t be deemed illegally practicing law without a license.

How much more reasonable would these professionals’ prices be if licensing were not required? I could then choose whether to use someone with a license or not and that licensed practitioner's price would be far closer to the actual worth of their services.

On the other hand, if we had a professional software developer’s license then I assume that malpractice would go along with that license. Thus, the quality level of software done by licensed professionals should greatly improve along with dramatically higher prices. However, this would also have the side-effect of dramatically lowering the number of people who could qualify to develop software. Innovation would most likely suffer. Many developers would be driven out of the business because of malpractice. We would have to schedule several weeks a year for court time defending ourselves for any bugs we create.

In short, I don’t think we should be pushing to license software developers. We should actually be pushing to remove much of the bureaucracy of licensing that is already in place.

Answer 19

This is an interesting question, but for other reasons. Most users only notice what we do when we do it wrong, the rest of the time most users aren't even aware that effort and experience are requried to make computers do what they do. What is really lacking is a group of professional bodies letting people know just what it is we do. Other engineering disciplines have this - electrical/electronic, civil, mechanical, etc. Software is often absorbed into the electronics field. Until people understand how difficult software is, we're going to be constantly undervalued:

3D IM Client - year and a half for a few thousand USD.

Company Portfolio $500 for about 90 days work.

Captcha Solver $250 to automatically solve captchas.

(Oh, and thanks Jeff et al for putting SO together so quick, it makes the rest of us look bad ;-) )

But to answer the question, no, programmers should not need to be part of professional bodies to practice. Work experience is always the most important factor. But, professional bodies could make our life as programmers better.

Skizz

Answer 20

I think, perhaps, the biggest difference between a programmer and a doctor or lawyer is that programmers have the luxury of QA and UAT (I know it doesn't always seem like a luxury, but it is). Computer programs can be dangerous. They can delete databases and backups, they can inadvertently expose critical information to the public, and all manner of manifestations of malfeasance or ineptitude. However, a prudent company realizes this and thus will not send an application to production until it is fully tested, security audited, etc. If the program is defective, then it's back to the drawing board to correct the error.

Doctors, lawyers, and accountants do not have testing. If the doctor is a quack who accidentally kills you, it happens "in production." Thus, before you go to a doctor you want to be certain that this doctor has the necessary skills to perform the task he is given, and the best way to scale this process to to have official accreditation. It's a necessary evil for doctors, lawyers, architects, and accountants, but completely unnecessary for software engineers.

Answer 21

Don't think it's necessary for someone to be part of a professional body to code, but it would definitely be nice to have more professional bodies that test code and certify their use in certain fields or in high-risk situations.

If a website is handling my credit card information and there's a certification from a well-recognized professional body assuring its safety, I'd feel a lot better about it.

Answer 22

I'm in favor of it. But there would have to be some kind of "Bar" exam and I just don't know how that could be implemented for programmers, software engineers, etc...