FreeRouting is gone - Creator was bullied by Zuken to take it down

[Marco]

You probably have far more un-patented/un-copyrighted stuff. None of them belong to you then?

If someone can make a copy of it without first depriving me of physical ownership they are perfectly free to do so, by law and by any reasonable ethical standard.

[sleemanj]

Let's say that someone hands you out copies of Windows (or any other stolen goods) and tells you it is all legit. You are still liable to be prosecuted.

That somebody would be unlikely to be the sole and only programmer responsible for the creation of the entire codebase comprising "Windows".

In this case, the somebody is, unarguably, the sole programmer responsible for Freerouting, and he says he released it as GPL, and he released it as GPL before this Zuken stuff blew up, and before the Zuken stuff blew up nobody in the slightest bit questioned if the release under GPL was anything but completely 100% valid.

Not to mention to be prosecuted, you have to have violated some agreement, or law, in the jurisdiction in which you are bound.  So which agreement or law and in what jurisdiction, did  anybody who received or distributed the GPL released copy of the source commit said violation. 

[dannyf]

depriving me of physical ownership

How can you claim ownership if it doesn't belong to you?

In this case, the somebody is, unarguably, the sole programmer responsible for Freerouting

I thought it was established earlier, by the very author, that he is NOT the sole programmer responsible for Freerouting - he said, in a forum somewhere, that 1% of the code came from Zuken, reproduced below for your reading enjoyment:

in the forum he mentioned "All in all the code parts of pure 45-degree push and shove routing algorithms must be far less than 1% of the complete Freerouting code."

[Marco]

That is completely non sequitur, the quote doesn't even suggests he did not code that part independently.

[Marco]

even though algorithms are not patentable , the algorithm does belong to zuken...

This is a really poor way of putting it, if it's not patented all the algorithm can be is a trade secret ... you could say they owned (past tense) the trade secret, which might be true depending on his contract, but that's not quite the same as owning the algorithm.

In the mean time the algorithm can not be called a trade secret any more though. If he was wrong in sharing Zuken only has a claim on him, not on the algorithm. Contributory infringement can't be claimed either because the algorithm was opened up in a way third parties could not possibly know of trade secret claims ... the genie is out of the bottle.

PS. they might still own copyright depending on his contract and local laws, but that has little to do with the algorithm ... which unless patented is solidly public domain.

[sleemanj]

he said, in a forum somewhere, that 1% of the code came from Zuken

Where does that say that he did not write it? 

Zuken claims he wrote some of the code while under employ, not afaik that he took work somebody else wrote.

Nobody is saying he did not write it, only when and where did he write it, or for who it was written.

Sent from my HTC Dream using Tapatalk 2

[Bored@Work]

Even if he typed in all the code it does not mean he might have the right to license it under the GPL. He might have already licensed, assigned to, given away, handed over, or whatever the rights/code/ideas to, e.g. Zuken. For example as part of signing his employment contract.

And by his own description he already had that idea and wrote some code while being employed. That won't help his argument that he can license all the code.

Anyhow, now that the issue is discussed publicly you will have problems with plausible deniability. If he doesn't have the right to license it and you get the code you would get unlicensed code. I.e. now you would knowingly take part in a copyright violation.

[jucole]

Grab them while they last (If Zuken vaccums the "net")

The more it's mirrored, the harder that is.  Here's an S3 hosted copy.

yeah! free software for all rocks!! - now they can make a claim for extra damages and loss of revenue against him!   I think you need to brush up on "the unwritten rules"

[free_electron]

even though algorithms are not patentable , the algorithm does belong to zuken...

This is a really poor way of putting it, if it's not patented all the algorithm can be is a trade secret ... you could say they owned (past tense) the trade secret, which might be true depending on his contract, but that's not quite the same as owning the algorithm.

In the mean time the algorithm can not be called a trade secret any more though. If he was wrong in sharing Zuken only has a claim on him, not on the algorithm.

If the algorithm was made on 'zuken time' (meaning he invented it during working hours) it is now intellectual property of zuken (or whatever was the paying party at the time. The IP may have been sold from Theda to zuken or whatever. Point is : you invent something during working hours : belong to your employer unless you have written proof it doesn't. Words are volatile...

So, two things are problematic :
1) does he have a written release of ownership to him? No ... Which brings in 2:
2) the ip is property of zuken. Having it in his freerouter is amount to theft. He gives away IP , that isn't his to begin with. If they want to get real nasty : theft of ip, industrial espionage and they can claim financial loss as people are using the ip whenever a board is routed with that code.

This is a big can of worms....

The fact that the code is now up for everyone to see, And recopy , makes this an even bigger problem. Not for freerouter or the code but for the writer. At least as long as he only delivered binary blobs and kept his mouth shut it would be hard to prove...
Now the code is out and he admitted the algorithms are in there... Run for the hills...

[marshallh]

As far as the actula software, which I used to use a lot -

It is both an automatic and manual push/shove router.
You export designs from Eagle or kicad, or anther package to specctra DSN file.

You can then route the pcb manually by hand (which is very nice and good for digital boards) or you can use its autorouter (which is very good, better than Electra and altium's autorouter, and loooads better than Eagle's piss poor router)

The problem: It does not scale well. It is best for 2layer boards. It works with 4layer boards, but is vastly slower.
Board area and total traces affects performance as well. I did a dual fpga design with it, and it ran slower than 1 FPS on my quad core i7.

It is best for hobbyist boards, and makes EAGLE somewhat usable by bypassing the routing.

[zapta]

As far as the actual software, which I used to use a lot -  ...

Does it handle ground polygons (mixed with signal traces)?   If so, how does it work, do you export the polygon from eagle or add it later?

Also, when you are done routing, do you import it back to eagle (to generate gerbers, DRC, etc)?

[madworm]

Suppose they have known about it for years and tolerated it. Now they suddenly change their mind. That is quite weird.

It would be best if they made precise claims as to what part(s) of the code (line-numbers) supposedly violate(s) their copyright or whatever. Claiming ownership of the whole thing is a bit overzealous. It might be inspired by 'their' algorithm, but they surely can't claim ownership of stuff that he coded in the 7 years after he left them.

They should either make precise claims or shove it. There's nothing wrong with defending ones property, but these unspecific all-or-nothing claims... I hate it.

And in case he should get sued, a little crowd-funded support would certainly be possible ;-)

[Marco]

If the algorithm was made on 'zuken time' (meaning he invented it during working hours) it is now intellectual property of zuken

Fine, tell me then ... if it's not patented and it's public knowledge under what law could the algorithm possibly still be considered intellectual property?

Point is : you invent something during working hours : belong to your employer unless you have written proof it doesn't. Words are volatile...

As I said before it could very well automatically have been a trade secret, but however painful that is to Zuken or the Freerouting author that's NOW completely irrelevant to US. It was their responsibility to keep their trade secrets secret, when it's no longer a secret it's also no longer a trade secret. Unless it was patented this is now a contract issue between them and concerns us not one iota, without a patent that algorithm is public domain.

This is one of the reasons to patent something in the first place, trade secrets are very fragile ... you can make any trade secret you know a non trade secret, just post it here or on usenet and it's done. It might have consequences for you but not for anyone else wishing to use the former trade secret (assuming no other IP laws apply).

[marshallh]

As far as the actual software, which I used to use a lot -  ...

Does it handle ground polygons (mixed with signal traces)?   If so, how does it work, do you export the polygon from eagle or add it later?

Also, when you are done routing, do you import it back to eagle (to generate gerbers, DRC, etc)?

Yes you can have it via to an internal polygon... but it makes it much slower. the polygon is part of the DSN you exported with a ULP run inside eagle.

After you're done routing you generate an eagle SCR script and run it inside eagle, and it rips up and lays down all the tracks as you did inside freerouting.

[mikeselectricstuff]

Suppose they have known about it for years and tolerated it. Now they suddenly change their mind. That is quite weird.

Of course the net effect of Zuken doing this is that more people now know about FreeRouting, which is out there and can never go away.

[Rigby]

And by his own description he already had that idea and wrote some code while being employed. That won't help his argument that he can license all the code.

depends on the way his employment terms were worded.  some let you have the things you create on your own time, some claim ownership of everything you do during your employment (with none of the payment, responsibility, or liability that can often go with some of that stuff, of course), and some just want the stuff you write while employed that could be useful for your work efforts.

in all cases, always claim that you simply do not author software outside of work hours.  even if you write a million lines of code at home, off hours, and they get the cops to take your PC as evidence, deny deny deny.  if they can prove you wrote it during off hours, take a defense of "how can my employer own my free time or the work I do on time I am not paid for," or something like that. 

that door should work both ways, and does if you say it should.  if they want your personal time, tell them they need to pay for that time.  they need to supply you with a computer they own, a network connection at home that they pay for, a good, ergonomically sound chair and desk, coffee/soda provided free and all that. filtered air, etc. all the trappings they have at the office.  asking for all that is how I maintained ownership of all the code I wrote.  they didn't want to do any of that, so I asked why they thought they should be able to reap benefits from work they do not support.

there is probably a first amendment argument, too, if this chap is in the US.

I am not a lawyer.  don't listen to me too much.

[zapta]

Yes you can have it via to an internal polygon... but it makes it much slower. the polygon is part of the DSN you exported with a ULP run inside eagle.

After you're done routing you generate an eagle SCR script and run it inside eagle, and it rips up and lays down all the tracks as you did inside freerouting.

That's sound very good. For me, eagle weakest point is the manual router (I am very happy with the auto router).

[madworm]

Sooo... things might change after all!

Although it seems Alfons himself doesn't want to escalate the conflict, I at least think it has become a little bit safer to keep the sources of FreeRouting available online (e.g. on github).

https://groups.yahoo.com/neo/groups/kicad-users/conversations/topics/18240;_ylc=X3oDMTJzOXRiNXE4BF9TAzk3MzU5NzE1BGdycElkAzE2MDI3Njk4BGdycHNwSWQDMTcwNzI4MTk0MgRtc2dJZAMxODI0MARzZWMDZG1zZwRzbGsDdm1zZwRzdGltZQMxNDAzMjUxMDQz

I saw about your case a couple of days ago.

The Free Software Foundation will pay Till Jaeger,
a German Lawyer, to look at your case and give you advice.
If you wish.

I hope this isn't too late.

--
Dr Richard Stallman
President, Free Software Foundation

[gildasd]

If you add a couple of constants to an algorythm or "shape shift it" by numerical means, does Zuken (for example) still hold the proprietary rights?

[engidea]

Hello Everyone

Having some time available I had a look at the Free GITHUB repository of FreeRouting

I am writint to this gorup to point out that Alfons really did an incredible job of writing Freerouting but as far
as what I understand at the moment there is nothing special in it beside standard Vector, planar arithmetic and binary trees for searching.

Trying to follow the spirit of FreeSoftware, the current "exploration" is also on Github at
https://github.com/Engidea/FreeRoutingNew.git

I will continue to look into it until I have time

I try to commit working versions and I do respect the various discalaimers/licences that where posted on the original files.

We as a community should really be more vocal for FreeSoftware

I normally test what I change using Kicad and a board that I use as reference, however, if any of you wish to do some more testing, you are welcome. Please don't scream if something does not work.....

Damiano

[CJay]

It depends on whether Freerouting and the software from Zuken share the same sourcecode. If that is the case Zuken could claim partial copyright. Some companies like to claim copyright on any code written by their employees during their employment even if they wrote the software in their spare time. That is a bit of a grey area.

I've had similar clauses in contracts relating to hardware and software designs but it was only enforceable for designs that related directly to the employer's business. 

One particularly nasty clause that I hope was unenforceable, tried to give the employer rights over any and all future designs relating to their business, which I was told by a director would apply to any work I did no matter if I was working for them or not as they'd assume it was created using knowledge I had gained while in their employment.

While I was and am in no position to be able to fight that if they got legal about it I'd have loved to see them try it in court.

[janoc]

One particularly nasty clause that I hope was unenforceable, tried to give the employer rights over any and all future designs relating to their business, which I was told by a director would apply to any work I did no matter if I was working for them or not as they'd assume it was created using knowledge I had gained while in their employment.

While I was and am in no position to be able to fight that if they got legal about it I'd have loved to see them try it in court.

You shouldn't have signed the contract with such abusive clause in the first place. Strike it out and hand the paperwork back to them. They either accept or they won't.

Sometimes employers will try such things to see whether they will get away with it, but you are not obliged to accept that. Legal or not, it is abusive - nobody can claim copyright on your brain, otherwise you could never ever work for anyone else. They could always harass you by claiming that you couldn't have done this or that without using their "secret sauce". You would probably prevail in court, but why to even let it go that far - just don't accept such crap in the first place!

[CJay]

One particularly nasty clause that I hope was unenforceable, tried to give the employer rights over any and all future designs relating to their business, which I was told by a director would apply to any work I did no matter if I was working for them or not as they'd assume it was created using knowledge I had gained while in their employment.

While I was and am in no position to be able to fight that if they got legal about it I'd have loved to see them try it in court.

You shouldn't have signed the contract with such abusive clause in the first place. Strike it out and hand the paperwork back to them. They either accept or they won't.

Sometimes employers will try such things to see whether they will get away with it, but you are not obliged to accept that. Legal or not, it is abusive - nobody can claim copyright on your brain, otherwise you could never ever work for anyone else. They could always harass you by claiming that you couldn't have done this or that without using their "secret sauce". You would probably prevail in court, but why to even let it go that far - just don't accept such crap in the first place!

Should have made it more clear, I didn't sign that contract, I spotted the clause and refused to sign, they withdrew the job offer, refusing to rescind the clause.

The company is long gone now, they disappeared after a very expensive legal battle concerning copyright violations. Seems they wanted to protect their IP but had no respect for anyone else's.

[CatalinaWOW]

If you code for a company then all 100% of that source code belongs to them; in the forum he mentioned "All in all the code parts of pure 45-degree push and shove routing algorithms must be far less than 1% of the complete Freerouting code."     even using 1% of someone else's code is not good.

If it's only 1% of the code than he can just rewrite it or somebody else to do it in a clean-room mode. Contaminating free code with proprietary code is not a wise move in the first place.

While it doesn't matter from a legal standpoint, it is interesting that everyone on both sides of this argument is rounding "far less than 1%" up to 1%.

While clean-rooming can be done, the rub here will be identifying which lines of code need to be re-done.  I am sure the records are not clear, and that both sides have different opinions on the matter.

[hs3]

If you code for a company then all 100% of that source code belongs to them; in the forum he mentioned "All in all the code parts of pure 45-degree push and shove routing algorithms must be far less than 1% of the complete Freerouting code."     even using 1% of someone else's code is not good.

If it's only 1% of the code than he can just rewrite it or somebody else to do it in a clean-room mode. Contaminating free code with proprietary code is not a wise move in the first place.

While it doesn't matter from a legal standpoint, it is interesting that everyone on both sides of this argument is rounding "far less than 1%" up to 1%.

While clean-rooming can be done, the rub here will be identifying which lines of code need to be re-done.  I am sure the records are not clear, and that both sides have different opinions on the matter.

I'm not familiar with the issue but has it been said anywhere that someone is using someone else's code? What he seems to be saying in that quote is that far less than 1% of the total code is those specific algorithms. So it could already be a completely new implementation of those algorithms and may be also implemented in a different way to achieve the same result. Maybe also a different programming language using different kind data structures? Just an observation and I don't know the details.