4 ms·
The software industry has been extremely productive without any licensing program. It's hard to see what regulation could improve that. In fact, it's an argume
by WalterBright 4y ago
The software industry has been extremely productive without any licensing program. It's hard to see what regulation could improve that.
In fact, it's an argument that many other professions aren't actually benefiting from licensing, either.
It ain't broke.
- 9wzYQbTYsAIc 4y ago> it's an argument that many other professions aren't actually benefiting from licensing Citations? edit: I found this “The effectiveness of licensing” [1], which suggests replacing licensing with mandatory registration programs (or suggestions of improvements for licensing boards). I’d be curious to see any research that advocates for removal of licensing without the addition of something else. [1] https://link.springer.com/article/10.1007/BF01044517 https://link.springer.com/article/10.1007/BF01044517
- phpisthebest 4y agoMost recent is the news coming out of FL where they are arresting qualified licensed roofers who just happen to have the licenses in another state other than Florida for daring to try to help people repair their roofs after a hurricane [1]. They are not claiming he was scamming people, door poor work, or anything other than not paying the fee and kissing the ring of the local economic protection guild AKA licensing board Combination of this there are several licensed FL contractors currently being investigated for Fraud because they promised Homeowers work they never delivered. The Trades are FULL of licensed scam artists Building Contractors are by far the best example I could give. Personally I do not want to get to a situation where my Open Source Code is downloaded by someone in say California and the next thing I know I am being arrested for "Developing with out a license" Hell if there was a License requirement for development when I was a teen I likely would have never learned how to develop code in the first place, I though HN was all about equality seems odd you would promote blocking out people and only allowing in the elite. Here is your $25 pi to learn to code on, but if you want to release your code you must first pay thousands of dollars and get approval from the code guild.... No thanks [1]https://nbc-2.com/news/local/charlotte-county/2022/10/08/roofing-contractor-arrested-in-charlotte-county/ https://nbc-2.com/news/local/charlotte-county/2022/10/08/roo...
- 9wzYQbTYsAIc 4y ago> Hell if there was a License requirement for development when I was a teen I likely would have never learned how to develop code in the first place There are cadres of people that struggle through medical licensing just for the chance to make less money than an unlicensed software engineer. I can only imagine that licensing (or some functionally equivalent regulatory mechanism) would boost the quality of the labor pool. > I though HN was all about equality seems odd you would promote blocking out people and only allowing in the elite. You may not be aware of this, but there are a wide variety of ways for people to get into STEM or medicine without being an elite. Heck, there’s billboards showcasing the non-elites being given free access to Harvard training. > The Trades are FULL of licensed scam artists Take a look at that research article that I linked in my parent comment. It identifies some suggestions for how to fix the situation by switching to a mandatory registration program, instead of licenses, or how to fix the licensing boards so that they properly punish incompetence or other professional failings.
- WalterBright 4y ago> Citations? Frech's "Competition & Monopoly in Medical Care" says the original motivation for licensing doctors was driving the Jewish and Black doctors out of the profession. https://www.amazon.com/Competition-Monopoly-Medical-Care-Frech/ https://www.amazon.com/Competition-Monopoly-Medical-Care-Fre... Today, the motivation is to restrict the number of physicians in order to increase their pay.
- 9wzYQbTYsAIc 4y agoThat appears to be a broken link to some publication by a political punditry organization and not peer-reviewed research. I don’t find that to be interesting, sorry.
- WalterBright 4y agoSorry, copy/paste mistake with the url. Corrected: https://www.amazon.com/Competition-Monopoly-Medical-Care-Frech/dp/0844738840/ https://www.amazon.com/Competition-Monopoly-Medical-Care-Fre...
- verisimilitudes 4y agoAlmost all software is broken, in fact. A real field of engineering would prohibit most of the stupid shit idiots claiming to be programmers do before even reaching for the other issues.
- 9wzYQbTYsAIc 4y agoNotably, the ACM has been compiling a list of hundreds of instances of failures in software that have negatively impacted the public.
- cuteboy19 4y agoIn most engineering disciplines, some degree of fault is permissible and is accounted for
- 9wzYQbTYsAIc 4y agoThat is my understanding as well, but some degree of fault is also impermissible and people must be held to account for those kinds of fault.
- WalterBright 4y agoYou can't punish people into being better programmers. Things don't work that way. I know that it's popular to believe that companies improve productivity by beating their employees, but that doesn't work in practice.
- 9wzYQbTYsAIc 4y agoAs a thought exercise, replace “programmers/employees/companies” with “criminals/inmates/prisons” in your argument and you will see parallels with modern law enforcement. Law enforcement embraces rehabilitation, prevention, and other mechanisms as well as simple punishment. Seems rather disingenuous to make a claim that negative reinforcement doesn’t work, vis-a-vis not having to be in jail.
- liketochill 4y agoWould you consider that some type of license that holds the individuals writing software which is responsible for safeguarding human life or the environment accountable and liable for its performance might have prevented some disasters, such as for example the Boeing max acas crashes?
- WalterBright 4y agoThere's no evidence that punishing people who make mistakes is productive. What has contributed to safety was no-fault attempts to find out what went wrong. The reason is simple. Punishment means the person who made the mistake becomes highly motivated to cover it up, deny it, shift blame, and not fix it (because fixing it means admission of, and liability for a mistake). No fault means that people can fix it without fear. Destroying people for making mistakes is a really, really bad idea for improving safety.
- liketochill 4y agoFor whatever web app, it doesn’t matter, but are you saying blameless post mortems are the only solution to software caused deaths and evironmental catastrophes? The buck has to stop somewhere.
- WalterBright 4y agoNo-fault investigations have been very successful in making airline travel incredibly safe.
- 9wzYQbTYsAIc 4y agoCitations for this claim and the parent-parent comment claims that you made?
- WalterBright 4y agoI've watched every episode of "Aviation Disasters" on the Smithsonian channel. At the end there's a summary of corrective action taken. I can't recall one that resulted in a criminal prosecution for making a mistake.
- ghaff 4y agoAnd, in general in the US, there's not a lot of licensing for engineering professions in general. It's fairly common in civil engineering or other jobs where you're signing off on things like drawings that are being filed with regulators. But it's not common overall.
- teh64 4y agoYes, the industry has been extremely productive at data leaks and invading users privacy! With regulation, how could there be innovation in the "spying on every moment of your life" and "everything-as-a-subscription" industries.
- WalterBright 4y agoLicensing will fix exactly none of that.
- 9wzYQbTYsAIc 4y agoPray-tell, what will?
- CodeSgt 4y agoYou don't have to know the solution to a problem to know when a proposed one isn't it.
- 9wzYQbTYsAIc 4y agoYou do need to be able to articulate why a proposed solution isn’t the solution.
- teh64 4y agoIf developers would have to be licensed to work with sensitive user data, and the penalty for leaking it would be losing your license, penalties and being unhireable for IT work, there would be a lot more incentives for people to be more careful. Have any of the developers who implemented the systems that leaked data faced any consequences as a disincentive for it not to happen again? It would incentive developers to care more about security, not just say "eh, it happens, and it doesn't affect me, so who cares". For user "spying", the code of ethics could contain something like "respecting the user" (idk, just spitballing) and those who want to implement more "spying" into their software could be "disbarred" for unethical behavior. This is even mentioned in the IEEE text: "Approve software only if they have a well-founded belief that it [...] does not [...] diminish privacy". I guess you could codify this into law (i.e. GDPR style), but even that (at least to my understanding) does not hold developers individually responsible, but their company. E.g. when google gets fined, the developers who created the code are less likely to think "this is bad, therefore I will restrain for doing such things in the future" and more "how close to breaking the law can I push without the company getting in trouble".
- bigbillheck 4y agoI've seen what's been produced over the years and all in all I'd rather have the licensing.