On Thu, 2012-07-05 at 14:01 -0400, Rob Weir wrote:
> On Thu, Jul 5, 2012 at 1:25 PM, Dennis E. Hamilton
> <dennis.hamil...@acm.org> wrote:
> > Let's step back a little.
> >
> > The markup that I created was designed to morph the Oracle terms of service 
> > to one that was similar but modified to use of ASF as the HOST and 
> > eliminate those considerations that do not apply (e.g., the use of distinct 
> > projects on the web site) under ASF custodianship.
> >
> > This approach was so folks could compare with what was already in place and 
> > produce a harmonious replacement.
> >
> > I'm not a lawyer and not prepared to say what is too much or too little.  I 
> > agree that plain language statements are preferable.  My favorite document 
> > of this kind is, after all, the Creative Commons Attribution Deed 2.0.
> >
> > However, there are some essential considerations, it seems to me:
> >
> >  1. Privacy should probably be separated as is commonplace on sites of this 
> > kind.  Privacy should extend beyond what is done to support performance of 
> > the site and what the monitoring is, to embrace more clearly what is done 
> > with anything that is considered personally-identifiable information.  In 
> > particular, the reliance on email addresses being made visible as a policy 
> > is something that users need to be aware of and why, and the need to be 
> > able to contact people by the e-mail address provided is also a matter of 
> > concern.
> 
> What I have here is the minimum we need, mainly to satisfy Google, via
> their Terms of Service on the use of Google Analytics.  If you want to
> add more, feel free.

Howdy Dennis, Rob, etc


I agree that it is common to see it as a separate page, but can't say
that I know of a why, beyond convention. 

Using a a single TOU for _everything_ from the main site to the blog it
would seem easier to move it into the common text, but the one thing I
don't want to see is a privacy policy listed in two places ;)

I don't have a big preference as to pp in the tou or a link to it.

The text in the privacy section in Rob's proposal looks like a straight
copy from http://www.openoffice.org/privacy.html with an addition of the
paragraph to handle the wiki(s)/forums registration and would need then
to update http://www.openoffice.org/license.html also. (I did not check
if the stand alone privacy page is linked from anywhere else)



> 
> >     If there is any kind of promise to be careful with personal data, that 
> > needs to be made in a way that the ASF can demonstrate diligence about and 
> > the ASF needs to be aware of the obligation the promise represents.
> 
> My draft did not make any promise, so I don't see an issue.
> 
> >     Privacy statements should be dated and back versions should be 
> > accessible.
> >
> 
> Yes, the ToU in general should be dated, etc.

OK

> 
> >  2. The Terms of Use should be clear about what the HOST (the ASF) is 
> > granting generally to users of the site and what contributors to the site 
> > are granting to the HOST and all Users, absent any clearly-established 
> > special cases (licenses, contribution agreements, etc.).  The outgoing 
> > license is a clear condition of a TOU and not having one for readers of 
> > general web pages seems erroneious.  The actions that the HOST will take 
> > without notice and at its own discretion need to be clear also.
> 
> Obviously I disagree.  IMHO, An explicit outgoing license is not at
> all necessary in the ToU.  Any public website has an implicit license
> that users can read it.  That's all we require as a general statement.
>  For most websites in the world, that is all they have.
> 
> Beyond that, in a website like ours, with eclectic content under
> various licenses, in some cases undetermined licenses, we cannot make
> any simple, general and true statement about the outgoing license.
> 
> It might be possible to make some extraordinarily complex statement
> about the outgoing license the website, but I don't see the value of
> that.  In the end we're publishing software, not a website.
> 
> It should not necessarily for someone to copy the website and create
> derivative works any more than it is necessary to copy posts from the
> mailing list to do the same.
> 
> >     In this case, I think that the terms of use definitely need to be dated 
> > and back issues available.
> >

IMO handle this by placing the license for a page on the page, and would
not expand the text in the TOU proposal. 

So I suppose what that means is I would favor a standard footer for the
different websites covered, using the current main sites footer as the
template, over time.

> 
> Sure.
> 
> > <orcmid comments="more in-line below"/>
> >
> >  - Dennis
> >
> > -----Original Message-----
> > From: Rob Weir [mailto:robw...@apache.org]
> > Sent: Wednesday, July 04, 2012 06:50
> > To: ooo-dev@incubator.apache.org
> > Subject: Re: Terms of Service on Forums
> >
> > On Tue, Jul 3, 2012 at 11:11 PM, Dennis E. Hamilton
> > <dennis.hamil...@acm.org> wrote:
> > [ ... ]
> >> I created an issue that proposed a new terms of use that was consistent 
> >> with the Oracle ones for ASF and would have not made this problem worse, 
> >> as far as I can tell.  That was long ago and it went nowhere.  The JIRA 
> >> issue is here: <https://issues.apache.org/jira/browse/LEGAL-104>.  Here's 
> >> the connected issue on our Bugzilla: 
> >> <https://issues.apache.org/ooo/show_bug.cgi?id=118518>.  I came to our 
> >> Bugzilla because LEGAL-104 can't have attachments.  The attachment on the 
> >> Bugzilla provides a red-lined transformation of the Oracle ToU into one 
> >> that could work for the forums, wikis, and web pages now under ASF 
> >> custodianship.
> >>
> >
> > OK.  I read that over.  IMHO it is still too much.   It is dressed up
> > in chain mail armour, as befits a big corporation with the risk
> > profile of a big corporation.  Large corporations are a magnet for
> > petty lawsuits.   But the circumstances are different with the ASF.
> > Our main concern is (or should be) clarity and helpfulness to the
> > reader, not to protect our multi-billion dollar corporation.  I''d
> > also entirely separate the outgoing license question.  This is not a
> > ToU question.  That can be covered separate by a "license" link on the
> > various websites, which might be simple in some case (incubator site
> > is 100% ALv2) and less so on some legacy sites.
> >
> > <orcmid>
> >   What do you mean by the incubator site?  I disagree about ALv2
> >  for http://www.openoffice.org since the honoring of ALv2 is more
> >  restrictive/intrusive than the incoming grant of legacy material
> >  requires.  Furthermore, patent non-assertions were never collected.
> > </orcmid>
> >
> 
> The incubator site is the one at
> http://incubator.apache.org/openofficeorg.  Since that site is 100%
> from Apache committers, submitted under iCLA, we can state
> unequivocally that that site is ALv2.  Agreed?
> 
> > So a minimal approach might look something like this, which could be
> > applied across all of our websites:
> >
> > ----------------------------------------------------------------------
> >
> >
> > Consolidated Terms of Use for Apache OpenOffice Websites
> >
> > The Apache Software Foundation hosts several websites on behalf of the
> > Apache OpenOffice project:
> >
> > -- under openoffice.org
> >
> > - under incubator.apache.org/openofficeorg
> >
> > - under cwiki.apache.org/confluence/display/OOOUSERS
> >
> > - under issues.apache.org/ooo/
> >
> > -- under blogs.apache.org/ooo/


> >
> > By your use of these websites you acknowledge that you have read these
> > Terms and agree to them.
> >
> > <orcmid>
> >   I think it should apply to the site being accessed and from which the
> >   ToU is linked. Also, the "By your use ..." without even a click-through
> >   the claim of reading and agreement is about as corporate as it gets
> >   and indefensible.
> > </orcmid>
> >
> 
> Defensible against what exactly?  What risk are you concerned about?
> 
> >
> >
> > 1. Privacy
> >
> > [ ... ]
> >
> > Additionally, some of websites  have a user registration systems that
> > queries you for information such as name, password and email address.
> >  This information is collected, stored and used in order to provide
> > you a unique login and to associate that login with your activity on
> > the website.  We endeavour keep this information private, but it may
> > be revealed to a competent authority under a lawful order.
> >
> > By using this website, you consent to the collection of this data in
> > the manner and for the purpose described above.
> >
> > <orcmid>
> >   This might need to be separated for what the agreement is when people
> >   register/subscribe and provide information solicited to accomplish
> >   that.
> >     This seems like too broad an umbrella for what happens when folks
> >   register versus what happens when accessing sites versus what happens
> >   when sending an e-mail somewhere.
> > </orcmid>
> >
> 
> It would be good to link to the ToU from any registration.  But note
> that we don't always have that access where it is a shared Apache
> service, for example CWiki.
> 
> Nothing in the ToU speaks about emails, so that is red herring.

A red herring? I don't think so - why should it only be valid if already
there. The site references our mailing lists and certainly did, likely
still does, IMO a comment on the public nature of mailing lists is
really appropriate here.

> 
> >
> > 2. User submitted content
> >
> > By submitting content to the website, you agree that this content may
> > be hosted on the website, for the benefit of other website users.
> > Additionally, you agree that your submissions may be edited, modified,
> > translated, copied, within the website.  You warrant that you have
> > sufficient rights to any content that you submit to the website
> > necessary to grant the above license.

I would add one thing (word) here - we have the right to _remove_ any
content,  picky I know but I would say it directly. 

> >
> > <orcmid>
> >   The "within the website" basically provides no outgoing license to
> >   the contribution.  The Oracle ToU is not clear about that, but a
> 
> Correct.  Nothing more than the implicit one.
> 
> >   counterpart to the incoming license would be very good there.
> >   I would not allow stipulation of broader licenses.  That makes this
> >   all impossible and is sort of anti-community, it seems to me.
> > </orcmid>
> 
> 
> Why would a broader license be anti-community?  And remember, we can
> never prevent a user from putting a broader license on a contribution.
> 
> >
> > If you wish to offer a broader license, to allow 3rd parties to reuse
> > your content outside of this website, then you may do so, provided the
> > license is compatible with the above requirements.  Apache License 2.0
> > is especially recommended.   Please mark the license prominently in
> > your contribution.
> >
> > <orcmid>
> >   To emphasize, this is too burdensome and it creates a problem around
> >   permissible use and who determines what that is.  Having bits and
> 
> Burdensome for whom?
> 
> >   pieces under individual license notices makes no sense.
> 
> We already have that, with openoffice.org website, with our incubator
> website, and even with other parts of Apache.  Take for example JIRA,
> where an attachment can be marked as being a contribution or not.
> 
> If you read the iCLA you see that as a committer you have that ability
> as well, to indicate in an email, or in subversion, or on the website,
> whether or not something is a contribution.
> 
> So we're not starting from some pure world where we can assume a
> single incoming license.
> 
> Another proof point for how this works is with the extensions and
> templates websites.  They manage to have eclectic licenses.  So long
> as they are each declared, there is no need for a default license or
> to exclude per-item licenses.
> 
> 
> >     It might be useful to have the conditions for submission to the
> >   site be at a place where submissions can happen, and deal more with
> >   what the outgoing license is in the absence of any notice to the
> >   contrary.
> > </orcmid>
> >
> 
> If we were interested in enforcing ToU against a user, then yes, we
> would make this bulletproof and put it in their face at registration
> time and at the time of their contribution.  But I don't see us having
> that need.  For us the ToU is more a set of notices that we want the
> user to be aware, for their benefit and to avoid confusion.  We're
> trying to be helpful.
> 
> > 3. Exclusions
> >
> > The websites at extensions.openoffice.org and templates.openoffice.org
> > are not operated by the Apache Software Foundation and are not covered
> > by this Terms.
> >
> > 4. Changes to these Terms
> >
> > We may change these Terms from time to time.  When we make substantive
> > changes we will also make an announcement on the ooo-announce mailing
> > list.
> >
> > <orcmid>
> >   These definitely need to be versioned and the older versions archived.
> >   I favor having references back to the previous one in a chain
> >   that anyone can chase.
> > </orcmid>
> >
> 
> That would be fine.

OK

> 
> -Rob

Ok - well, I think that the current TOU page should be updated sooner
rather then later now - it's been deferred long enough huh? *smile*.

I think it is easier to start with simple and embellish as we see fit.

so how about let's use this page:
https://cwiki.apache.org/confluence/display/OOOUSERS/website-terms-of-use-draft
fir a quick white board. 

Will start from a copy of Rob's text - add my two edits - and lets just
work changes if folks have them, if none move it to the website page
proper on Monday - sound good?

//drew

Reply via email to