Draft of LUG letter

[Arthur Ogawa]

Sebastian Rahtz wrote:

> Arthur Ogawa writes:

> i am not sure you realize what you are taking on. there are
> 533+ packages to look at... but I do have some information coming in
> now about copyright after a massive mailing. dont get stuck into this
> without talking to me further....

OK, let us talk further. I cannot do much without the information anyway.

I had intended /COPYING to explain things in more detail, to refer to the
licensing requirements of the packages for which those requirements are
clearly spelled out (e.g., LaTeX and GS among others), and to try to organize
as many as possible into common classes, e.g., freeware, GNUware, LPPLware, etc.

I intended ultimately to fall back on "The recipient of the CD itself can only
copy and redistribute the software thereon by complying with the conditions
placed on the relevant packages involved." That some packages are mute on the
point is a pesky detail.

> 0.7 per CD is probably excluding the plastic sleeve, at least. you
> need to decide how to pack it.

Yes. Patricia got a bid for the CDs on spools. But we need them to be in
either paper or plastic sleeves in order to put them in TUGboat. She'll take
care of this item somehow.

>  > I'd like to do it your way, but Peter Deutsch's terms are GNU-like, 
>  > and require no prohibition on further copying...
> is the tail wagging the dog here?...

Well you may ask. Let's try it this way: does the presence of multicol.dtx
mean that Peter Deutsch's requirements cannot be satisfied? Is there any other
content on the CD that *must not* be copied under any circumstance?

To first order, I don't see that Peter's conditions throw a spaniard into the works.

>  > The recipient of the CD itself can only copy and redistribute the
>  > software thereon by complying with the conditions placed on the
>  > relevant packages involved...
> ok, so long as you accept that a good many packages will be "unknown,
> cannot contact author"

The individual who wants to distribute the S/W has to accept the potential
liability for doing so. That's what it comes down to.

>  > ...each package itself contains
> or maybe not....
All right, noted.

>  > * Not be distributed for profit.

> this is not [a provision of the LPPL]

Yes, I know. You see, I am presenting a list of the possible restrictions, so
that the end-user can be aware of what sort of actions on his or her part
*might* constitute a violation of the license. Perhaps I ought to rewrite this
portion from that standpoint.

In short, if you do anything more than copy the S/W to your computer or your
server, you must take the trouble to ensure that you are in compliance with
the licensing conditions of the S/W.

>  > * Not be used for commercial purposes without license directly from
>  > * the author.
> I hate Bernard

Actually, I had multicol.dtx in mind here. Has FMi any plans on relaxing that
restriction? I swear, demanding a license for S/W that constituted the basis
for his Knuth Scholarship seems a bit strange. Not to speak of its not being
revised in 10 years.

>  > ...the licensors of the software simply wish to ensure that their
>  > work continue to benefit people like you...
> some of them want to make money

Which packages fall in this category?

BTW, I have a retronym in mind for S/W that is free to use by not-for-profit
enterprises, but which entails a fee when used by a commercial entity:
grubware ;-)

-- 
Art