[consulting] Proper Collections Procedure

[Eric Goldhagen]

At 1:44 PM -0400 8/17/06, Gary Feldman wrote:
>Eric Goldhagen wrote:
>>Same with openflows. We have refused a number of times to sign 
>>contracts that define anything as "work for hire." It is our 
>>opinion that the legal reality of a work-for-hire agreement is in 
>>itself a violation of the GPL. That applies to our relationship 
>>with our clients as well as contractors.
>From an ethical perspective, this makes sense, but I don't think 
>that work-for-hire violates the GPL.  The GPL is about licensing, 
>not legal ownership.  There's nothing in the GPL that keeps any 
>copyright owner from selling or reassigning the ownership of the 
>copyright.  Indeed, FSF insists that their contributors do so (see 
>http://www.gnu.org/licenses/gpl-faq.html#AssignCopyright). 
>Work-for-hire is a little bit different (it changes the original 
>owner), but it's the same idea.


There is a difference between allowing a coder to re-assign copyright 
or ownership and entering into agreements that force such 
re-assignment. One is granting a freedom and power to decide; the 
other denies both.

(quoted from an online source I no longer have url for)
With a work for hire, the hiring party steps into the shoes of the 
creator and becomes the author of the work for copyright purposes. 
With a work for hire, all of the attributes of copyright ownership -- 
including credit and control -- vest in the hiring party, not the 
creator.
(end quote)

My concern is that work for hire means that the coder has no right to 
claim credit for their code. This does not allow for proper 
attribution of work. This also does not allow for someone to 
know/contact the person that wrote the code for advice later on. It 
also does not allow the coder any way of ensuring that the end 
product remains licensed under the GPL. Therefor, in my opinion, work 
for hire agreements are a potential violation of the GPL (maybe more 
the spirit of the GPL than the legal letter of the GPL).



>Please remember that public domain and GPL are incompatible in a 
>sense.  Once something is in the public domain, anyone is free to 
>make a derivative work that is no longer in the public domain nor 
>subject to the GPL.  Of course, anyone else is free to make a 
>different derivative work, also no longer in the public domain, and 
>choose to release it under the GPL.
>The usual "I'm not a lawyer" disclaimers apply.  Sorry if this is 
>stuff you already know, but enough people confuse "public domain" 
>with GPL that it's worth pointing out the distinction from time to 
>time.

I'm might not be using the term public domain in its official legal 
sense, as you make clear.  Thanks for pointing that out.

I'll be certian to make that distinction when I talk about this in 
the future, or at least to clarify my definitions before I do my next 
presentation on Free/Opensource Software and Licenses (which is 
scheduled for sept 16th at Simmons College, near Boston if anyone 
want to come and chat afterwards ;)

I agree with your point about the incompatibility. From my 
perspective, the GPL exists to ensure that things remain in the 
public domain (maybe publicly accessible is a better term?). If you 
simply release something into the public domain, anyone can later 
take that and make it theirs and deny you rights to it.

Indeed it is just this theft of code and knowledge from the public 
domain that motivated the creation of the GPL. (see the gnu manifesto)

The GPL boils down to saying "I own this, and now that we've gotten 
that legal crap out of the way, since I own this I am putting this in 
the public domain for anyone to use for any use and modification" it 
also make an attempt to ensure that those modifications are also made 
fully accessible to the public.

Thanks for forcing me to clarify.

--Eric

-- 
-------------------------------------------
Openflows Networks Ltd.
New York | Toronto | Montreal | Vienna
http://why.openflows.org
People are intelligent. Machines are tools.