[FM Discuss] residency, and floss for $

[adam hyde]

hey,

> >   
> The biggest issue I see here is how exactly to pay out the bounties to 
> multiple contributors (obviously if one guy writes the whole manual it's 
> easy), because all of the ways to divvy up money have problems. For 
> example, paying based on proportion of the final word count rewards 
> rambling, verbose text (not that I should talk); paying based on the 
> number of chapters doesn't take into account that some are much harder 
> to write than others; etc. The most flexible and least gameable would be 
> for some human to judge what proportion of the final product each person 
> deserves credit for, 
> 

Well, I thought the process of evaluation really comes down to two
issues - an evaluation of quantity, and an evaluation of quality.

The quantity can be done with scripts. Proffessor Luca De Alfaro has
worked trough a reputation system that enables an analysis of the final
content of a wiki page, and determines what quantity of content is
contributed by which users. There are many parameters that can be
tweaked according to Luca. So this would be the first interesting
possibility.

The second is quality. Much harder to determine and the only way through
it is to appoint an evaluation committee. They would review the content
and discuss exactly the issues you raised.

>but that adds a social problem if people feel like 
> their contribution was undervalued and get mad. I'll admit though that I 
> haven't thought about this at length or looked to see if anyone else has 
> come up with workable models; do you have any ideas on that?

I think there is a chance that people get mad. I don't know what to do
about it but do as much as possible to make the process fair, and then
accept that people can and do often see it as unfair. I think this
process will always have a component of fallibility but the point is to
be as clear and communicative as possible about what is trying to be
achieved and then be as fair as you think you can.

As for the clear communication - I think perhaps it would be interesting
to think about how to best communicate the process before it starts.
Would it be wise, for example, to publish a 'difficulty' rating for each
chapter and this rating would then help weight the final calculations? 


> >   
> 
> I think the big difficulty is that volunteers don't like to feel like 
> they're "working for free" when someone else is getting paid to do the 
> same thing.  I'm not sure of how exactly to handle it best, but it seems 
> like a way this might naturally work out in FLOSSmanuals is that the 
> paid people would have more responsibility.  For example, a paid person 
> might have deadlines, or they might be responsible for ensuring a manual 
> gets finished (writing the parts nobody else wants to, etc.), doing 
> editing/integration, meeting/negotiating with sponsors, etc.  The 
> volunteer person, of course, would just write what they feel like when 
> they feel like, though of course they would still coordinate with the 
> other writers as good project citizens.


souns like a good clear distinction to me. Perhaps Maintainers get paid
(where possible) but also have deadlines etc. This is what I was gettign
at with the post about added-value advertising. Each maintainer could
take some of the income from the ads for the manual they maintain.

> 
> Things should be fun and based on mutual agreement for both volunteer 
> and paid writers (making paid writers do more unnecessary work isn't a 
> solution), but it might smooth things over if the paid writers clearly 
> have more responsibilities... then a volunteer is less likely to think, 
> "wtf, that guy's getting $500 for the same thing I'm doing free; where's 
> my $500?"

yep, good idea. In the situation of maintainers being paid the role is
clear but should be written down. It would be something like:
* structure the manual
* ensure chapters reflect current software version
* look for new contributers
* keep the edit homepage up to date
* manage the print-on-demand 
* update the published manual as necessary

> 
> A separate suggestion might be to have some pool of discretionary funds 
> (if enough money can be raised) to reward longtime and productive 
> volunteers, either with direct payments, or with purchase of equipment 
> or sending them to conferences or something like that.  Basically some 
> funds that can be used for, "hey, this guy's written like 3 manuals for 
> free, we should do something for him".

also a good idea...


adam



> 
> -Mark
> 
> _______________________________________________
> Discuss mailing list
> Discuss at lists.flossmanuals.net
> http://lists.flossmanuals.net/listinfo.cgi/discuss-flossmanuals.net
-- 


adam hyde
floss manuals

free manuals for free software
http://www.flossmanuals.net

mobile : + 31 6 154 22770 (Netherlands mobile)
email : adam at flossmanuals.net