Re^4: Responsibilities of a module author

[BrowserUk]

Adding the requirement that outsiders may take over authorship and copyright of a module isn't a good thing in my opinion.

I'm at a loss to see where anyone suggested that this could/would or should happen?

I'll restate that I was not suggesting anyone should be able to "take over authorship and copyright". All existing copyrights and licences would remain in place and sacrosanct, as with any other OS code.

The idea was to provide a somewhat formalised mechanism for the extension and improvement of modules through the collaborative efforts of it's users, that maintained a continuity between the developments that would aid both the original authors; and existing and future users.

If the original author is unable to accommodate changes that users require for their use of the module, whether through

1. being uncontactable;
2. unable to continue maintenance for whatever reason;
3. sees the requested changes having undesired affects upon other users of the module;
4. or simply that they take the module in directions he never intended it to go;

A somewhat formalised mechanism for forking the development would ensure not only that the collaborating users could find each other and achieve their needs; it would also ensure continuity of both authorship and licencing of the original elements of the module.

Not only does this help ensure that you don't get half a dozen unilateral forks by disparate individuals; it also allows future users to find and follow the developments rather than their being spread all over the Internet, (or worse, unavailable publicly!). It could also allow the original author to "keep a watching brief" on where the fork is going, and even offer consultative advice to it's progress.

There was no intent to impose extra restrictions upon authors, nor in any way diminish their rights. Just to formalise those rights and procedures already in place--to everyones benefit.

[Perl Mouse]

Adding the requirement that outsiders may take over authorship and copyright of a module isn't a good thing in my opinion.

I'm at a loss to see where anyone suggested that this could/would or should happen?

mifune wrote: I feel that thought could be given to a natural and agile way (built into the structure of CPAN) of branching or evolving a given module, a way that automatically took care of authorship, copyright and and other important formalities.

If authorship and copyright are to be stayed at the original author, what are the "formalities" regarding authorship and copyright then?

A somewhat formalised mechanism for forking the development would ensure not only that the collaborating users could find each other and achieve their needs

There's no reason why this mechanism should start with changing CPAN. I'd say, hammer out the details of your mechanism. Start acting on it. Find the quircks. Work them out. Then, after you have a working mechanism, first contact the modules mailing list with your suggestion, then try to get PAUSE changed to handle your (by now proven) mechanism, and then you can see whether you can build this into CPAN.

it would also ensure continuity of both authorship and licencing of the original elements of the module.

No change to CPAN is required to do. Copyright and licensing laws already force this to be the default.

Not only does this help ensure that you don't half a dozen unilateral forks by disparate individuals;

Actually, it won't help your ensure that you don't get half a dozen forks. At best, it will prevent some forks, but there will still be people ignoring whatever framework you create, and creating their own forks. This is an open source world we live it.

[BrowserUk]

If authorship and copyright are to be stayed at the original author, what are the "formalities" regarding authorship and copyright then?

If nothing else, ensuring that when the code is forked, nothing gets accidently removed that shouldn't. And, assuming that the fork was done using version control software, then any all all subsequent modifications, including any that accidently affected these issues, were recorded and reversible.

There's no reason why this mechanism should start with changing CPAN.

That's true. It could be done on sourceforge, or on another accessible server somewhere, or my harddisk, but that just emulates what already happens.

Equally, I or others could attempt to put together some mechanisms independant of cpan in the hope that they might be incorporated there at a later date, but that leaves us without detailed knowledge of the cpan infrastructure and we would probably arrive at something that was incompatible with it.

Even if we lucked out and came up with something that could be added, the chances are that it wouldn't be, because the incumbant denizens of CPAN would disagree with our choices of versions control software, or authentication mechanisms or whatever. You'd also get multiple, incompatible solutions developed in parallel, wasting everyone time and efforts.

... but there will still be people ignoring whatever framework you create ...

That's also true, but if there is no system in place, then everyone has no choice but to ignore it and go their own way. If there was something to ignore, then they would at least have the choice of not doing so.

I don't get it. I mention an "idea", and clearly label it as such, and you have already condemned it as being tantamount to piracy whilst simultaneously stating that the legal protections are already in place. You then suggest that the way to achieve collaboration is to do it yourself.

If CPAN is already so perfect, or so fragile, that even thinking about ways to change it are seen and condemned as heresy or sedition or illegal or immoral, then I guess it's best left exactly as it is.