Arthur is wrong on his final point.  When an author 
assigns copyright to a publisher, the author gives away 
all rights.  It is equivalent to selling your house, your 
car or anything else.  Once you've sold it, you've no 
right to enjoy it's use any more, even  though you were 
the previous owner.

So when an author assigns copyright to a publisher, he or 
she has no rights to keep a back up copy, store it in a 
repository, etc., UNLESS the publisher graciously gives 
permission for the author to do so.  But what the 
publisher cannot do is demand deletion, etc., of earlier 
drafts of the manuscript, because the author has only 
assigned the final accepted version to the publisher.

Charles

On Sat, 14 Feb 2009 15:01:59 +1100
  Arthur Sale <[log in to unmask]> wrote:
> Talat
> 
> 
> 
> Let me assure you that you should credit that a court 
>would accept a case
> that repositories fulfil other functions. Indeed in 
>Australia we could argue
> that they are required by the Federal Government for the 
>purpose of
> institutional publication reporting and research 
>evaluation. Tasmanian law
> requires the university to keep records for long-term 
>preservation under the
> Archives Act and so do most States.
> 
> 
> 
> The other point you miss is that publishers have no 
>rights to prohibit a
> restricted copy being mounted in a repository. If an 
>author chooses to keep
> a copy of his or her article in one computer system or 
>another (or is
> required to place a copy in a particular one) is of no 
>concern whatsoever to
> a publisher. They might as well demand that the author 
>delete the manuscript
> from their personal PC once it has been published! 
> Indeed my departmental
> backup system makes regular copies from my PC somewhere 
>and I don't bother
> enquiring where, nor does any publisher of my work. 
>Neither do they demand
> that a particular filing cabinet be used for any paper 
>drafts. None of this
> is of any concern to a publisher.
> 
> 
> 
> You and I have had this argument before and you persist 
>in this view, but it
> cannot go unchallenged if you keep making it. It does 
>not stand up to
> examination.
> 
> 
> 
> Arthur Sale
> 
> University of Tasmania
> 
> 
> 
> -----Original Message-----
>From: Repositories discussion list 
>[mailto:[log in to unmask]]
> On Behalf Of Talat Chaudhri
> Sent: Saturday, 14 February 2009 2:57 AM
> To: [log in to unmask]
> Subject: Re: Fair-Use/Schmair-Use...
> 
> 
> 
> Yes, as Charles points out, the only way of doing it 
>legally is to get 
> 
> the author to give permission to a private requester. 
>The technology 
> 
> used has to allow for that, with an appropriate security 
>mechanism to 
> 
> make sure nobody else gets it.
> 
> 
> 
> I've never been clear whether a repository has a right 
>to store such a 
> 
> copy on a restricted basis in the first place if the 
>publisher says it 
> 
> may not, but it's often contended that this is allowed 
>for 
> 
> administrative purposes. Since repositories are 
>demonstrably not 
> 
> intended for that, and since the button is in place in 
>such a scenario, 
> 
> the intent seems to me to be to solicit requests and 
>therefore to 
> 
> redistribute. To my uninformed mind that would look like 
>storing a copy 
> 
> in breach of copyright. The response I've had in the 
>past from Stevan 
> 
> Harnad and others is, uncharacteristically I might add, 
>that 
> 
> repositories may serve purposes other than distributing 
>content on the 
> 
> web. I find it hard to credit that any court would be as 
>simple-minded 
> 
> as to swallow that argument, should the issue ever 
>arise. It hasn't come 
> 
> to court so far, so let's hope it never does. Perhaps 
>any such case 
> 
> would be unprovable anyway, but it would be nice to know 
>the legal 
> 
> position on storing a copy in a retrieval system, as one 
>often sees the 
> 
> phrase in copyright statements in published books.
> 
> 
> 
> Perhaps Charles has the answer to this? Clearly if one 
>can't store a 
> 
> copy, the issue of the button would never arise unless 
>the publisher 
> 
> specifically allowed a restricted-access copy to be 
>held, rather 
> 
> contrary to the purpose of their opposition to OA and 
>presumably thus a 
> 
> rare occurrence. If there are reasonable legal grounds 
>to hold a 
> 
> restricted-access copy in a repository despite the 
>publisher's lack of 
> 
> permission, then the issue evaporates.
> 
> 
> 
> 
> 
> Talat
> 
> 
> 
> Richard Rankin wrote:
> 
>> I may have misunderstood the thread
> 
>> 
> 
>> I had read it that the eprint request would 
>>automatically email the paper
> without intervention from the author.
> 
>> 
> 
>> Is it that the eprint request sends a message to the 
>>author to email the
> paper?
> 
>> 
> 
>> Ricky
> 
>> 
> 
>> ________________________________________
> 
>> From: Repositories discussion list 
>>[[log in to unmask]] On
> Behalf Of Talat Chaudhri [[log in to unmask]]
> 
>> Sent: 13 February 2009 15:32
> 
>> To: [log in to unmask]
> 
>> Subject: Re: Fair-Use/Schmair-Use...
> 
>> 
> 
>> Any transport protocol would seem fine provided that the 
>>author's active
> 
>> consent is incorporated into the process for each 
>>personal request.
> 
>> Unfortunately, HTTP doesn't work that way. I suppose the 
>>system could
> 
>> instead send a unique authentication code by email to 
>>the person who
> 
>> wanted to see it by HTTP whereas normally it would be 
>>restricted, but it
> 
>> would have to work only once in case they shared it 
>>against the author's
> 
>> wishes - so what would be the advantage over sending it 
>>by email? Unless
> 
>> of course one has an account on the repository or 
>>another linked
> 
>> service, again requiring the author's permission - but 
>>are people really
> 
>> going to use one FB-type service for academia and 
>>request/give
> 
>> permissions through it? Isn't it simpler to use email?
> 
>> 
> 
>> Ian Stuart wrote:
> 
>>   
> 
>>> Ricky Rankin wrote:
> 
>>>     
> 
>>>> Sorry to return to this.
> 
>>>> 
> 
>>>> As I understand the arguement goes that by including a 
>>>>eprint request
> 
>>>> link/button it is OK to email the publishers pdf as this 
>>>>is the same
> 
>>>> as sending a copy through the post.
> 
>>>> 
> 
>>>> A computing colleague has asked why is ths different 
>>>>than displaying
> 
>>>> the pdf on the screen of the requester's screen as http 
>>>>is just
> 
>>>> another transport protocol?
> 
>>>>       
> 
>>> The difference is that the paper is not Openly 
>>>available: The author
> 
>>> of the paper makes a concious decision to produce a copy 
>>>for a fellow
> 
>>> academic
> 
>>> 
> 
>>> An Open Access (true Open Access) article would have the 
>>>paper
> 
>>> available without restriction.
> 
>>> 
> 
>>>     
> 
>> 
> 
>> --
> 
>> Dr Talat Chaudhri
> 
>> ------------------------------------------------------------
> 
>> Research Officer
> 
>> UKOLN, University of Bath, Bath BA2 7AY, Great Britain
> 
>> Telephone: +44 (0)1225 385105    Fax: +44 (0)1225 386838
> 
>> E-mail: [log in to unmask]   Skype: talat.chaudhri
> 
>> Web: http://www.ukoln.ac.uk/ukoln/staff/t.chaudhri/
> 
>> ------------------------------------------------------------
> 
> 
> 
> -- 
> 
> Dr Talat Chaudhri
> 
> ------------------------------------------------------------
> 
> Research Officer
> 
> UKOLN, University of Bath, Bath BA2 7AY, Great Britain
> 
> Telephone: +44 (0)1225 385105    Fax: +44 (0)1225 386838
> 
> E-mail: [log in to unmask]   Skype: talat.chaudhri
> 
> Web: http://www.ukoln.ac.uk/ukoln/staff/t.chaudhri/
> 
> ------------------------------------------------------------
>