Live data from Hacker News

If Software Is Funded from a Public Source, Its Code Should Be Open Source

linuxjournal.com

201–210 of 277 posts

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#201

Earlier quoted context omitted.

What usually happens is that it doesn't get released at all. It isn't worth the effort. I've got tons of code I've written over the years that, if I could just throw it onto github, I would. But since it would take me six weeks of pain, I don't. Some of that code I'm kind of proud of, or was at the time I wrote it, anyway. I wrote an rviz app a decade before ROS took off.

Would there be some way of just automatically releasing everything in a git repo after X days (30? 60? 90?) unless some action is taken? Commit/push like normal, and some other process pushes to a public repo after X days, but anyone can review and block in the meantime for security reasons when needed?

No. “Automatic release” is a contradiction in terms.

It could happen, but it would take something like the Secretary of Defense ordering it. Or a White House directive. The folks at my level, or at any level I can actually talk to, don't have the authority. The thing is, they aren't making these policies and procedures up just to be dense. There are real, hard and fast laws and department-wide policies that require the system to look the way it does.

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#202

Consider the Feds tried doing this with IP post WW2 and the results were disasterous. Turns out govt. employees aren’t allowed to be paid enough to make 100 hour work weeks reasonble - the level of effort required to get some ideas off the ground. Once the Bayh-dole act was passed to allow citizens to take ownership of IP related to publicly funded endeavors, only then did the public start benefiting.

I came here and searched for the Bayh-Dole Act and am surprised to find only this single mention of it. It really does seem to me to be at the root of the current policies around federally funded research artifacts.

I disagree with the causation you infer regarding public benefits as a result of this change in IP rules. After all, we saw many post WW2 benefits of federal spending long before 1980. The most visible were in aerospace developments that gave us the jet age, but of course even Silicon Valley was well on its way in the 1970s with lots of computer industry groundwork already in place.

I think a lot of the computer industry developments of the 1980-1990s were almost inevitable once that stage was set. It was mostly a coincidence that Bayh-Dole was passed and universities ramped up their strip-mining of the federal budget around the same period. An awful lot of the current Internet age was built by people like me, working on open source projects and federal funding in spite of Bayh-Dole, not because of it.

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#203
post #191

Earlier quoted context omitted.

Psst, it's not our state of the art software that's holding this ball of mud together.

It's a competition nonetheless.

Is it? What would it look like to win this competition? To lose?

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#204

Government scientist here. IANAL, but my understanding is that all software written by US government civil servants is in the public domain. It cannot be copyrighted or otherwise licensed (although the government can patent things). This does not necessarily mean that it is available to the public, as noted in the other comments; it may be classified, considered For Official Use Only, or restricted from release by IT…

Lawyers for at least some multinationals get hives when you talk about public domain software. I've had to switch libraries or participate in cleanrooming a single PD function a couple times to get them to calm down. Apparently in Europe and some other places it's not so easy to disavow all rights to a bit of software. BSD- or MIT-style licenses are much more comforting.

This is why the CC0 is way more useful than just putting "This code is in the public domain" at the top of every header file.

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#205

Government scientist here. IANAL, but my understanding is that all software written by US government civil servants is in the public domain. It cannot be copyrighted or otherwise licensed (although the government can patent things). This does not necessarily mean that it is available to the public, as noted in the other comments; it may be classified, considered For Official Use Only, or restricted from release by IT…

On the other hand, some branches like the Department of Energy are making a big push for open source. You can see how many projects Los Alamos has, for example:

https://github.com/lanl

And:

https://github.com/laristra

Keep in mind, Los Alamos is one of the big labs doing nuclear work for the DOE, so obviously a lot of what they do is classified, but of the stuff that isn't they're actually doing a pretty impressive job of open sourcing it.

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#207
post #102

Government scientist here. IANAL, but my understanding is that all software written by US government civil servants is in the public domain. It cannot be copyrighted or otherwise licensed (although the government can patent things). This does not necessarily mean that it is available to the public, as noted in the other comments; it may be classified, considered For Official Use Only, or restricted from release by IT…

I used to work in a state government job as a student intern, and they said they wanted an internal directory for some various tasks. You can think of it like an inventory manager sort of website, purely on our intranet. The mad lad my lead was, said we would need a database for retaining some of the basic data we needed to keep track of. Nothing all that special, just a few tables, and by nature nothing sensitive wo…

This site, Hacker News, uses flat files for its database.

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#208

Earlier quoted context omitted.

Mostly sortof true, but there are lots of complications. I've delved into this. For lots of detail on the current situation, see my paper, "Publicly Releasing Open Source Software Developed for the U.S. Government" by David A. Wheeler, https://www.csiac.org/journal-article/publicly-releasing-ope... In almost all cases, if a US Federal employee writes software as part of his/her official duties, then there is no copyr…

This DoD CyberSecurity Policy Chart in the sidebar is epic: https://dodiac.dtic.mil/dod-cybersecurity-policy-chart/ Clearly, they don't want new ideas.

I think they just try to idiot proof too much, which ends up a bureaucratic nightmare.

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#209

Government scientist here. IANAL, but my understanding is that all software written by US government civil servants is in the public domain. It cannot be copyrighted or otherwise licensed (although the government can patent things). This does not necessarily mean that it is available to the public, as noted in the other comments; it may be classified, considered For Official Use Only, or restricted from release by IT…

I'm joining the Air Force as a Civilian with the explicit purpose of solving this. In fact we've made a lot of progress with Continuous ATO and some other efforts to get faster iteration, deployment, sharing CI/CD etc... processes in place. It's going to take time but we're on the right track. Send me a message if you want to be involved in this culture change.

Very cool, I recall seeing a lightning talk at KubeCon on this very thing: https://schd.ws/hosted_files/kccna18/45/NA%20KubeCon%202018%...

Re: If Software Is Funded from a Public Source, Its Code Should Be Open Source

#210

Earlier quoted context omitted.

> The bigger problem, though, is that much of the US government just isn't technically set up to release software (as opposed to documents). Setting up a git repository, in the DoD, is an utter nightmare. Every single patch has to be pushed through a public release process that can take weeks or months and involves review by as many as ten to fifteen different offices, few if any of which even know anything about sof…

I was told by one government contractor that it takes forever to review the code and pull out any hard coded passwords.

That can be corrected because hard coding passwords fail the government secure coding policies. So if the contractor hard coded then they failed the quality requirement. So it gets them into a fun for the public catch22.

The most effective method is to have continuous development through an open source structure so any failures where someone hard codes sensitive info in source get found quickly.

Post reply on HN