[Open-access] Letter to publishers - URGENT

Mon Mar 5 11:41:38 UTC 2012

On Mon, Mar 5, 2012 at 11:45 AM, Mike Taylor <mike at indexdata.com> wrote:

> On 5 March 2012 10:38, Douglas Carnall <dougie.carnall at gmail.com> wrote:
> > On 5 March 2012 09:51, Peter Murray-Rust <pm286 at cam.ac.uk> wrote:
> >
> >> We assert that subscribers [1] have an absolute right to use machines to
> >> extract facts [2] from the scholarly literature in all forms [3]  in
> >> arbitrarily large amounts [4] whenever they wish [5] and publish the
> output
> >> as CC0 [6].
> >
> > Your footnotes make your actual intention clear, but when I first read
> > this sentence I thought "this sounds like a recipe for crashing
> > servers;" or at least that's one possible interpretation for the
> > "absolute right" to extract as "large amounts" of information
> > "whenever."
> >
> > I think you could address this potential criticism by deleting the
> > word "absolute" [very few rights are indeed so, sigh]
>
> Agreed -- this small change would make a big difference to the
> perceived tone without actually affecting the meaning.
>

For me the strong tone of the message is absolutely OK. Of course we do not
want to be to harsh, but I don't think the letter is. Replacing the
absolute with 'fundamental' would be expressing my conviction on this topic
quite nicely, where 'absolute' has an undertone of frustration, this just
would state matters as a fact.

>
> > and inserting an
> > additional footnote after the word "right" that makes clear that you
> > of course intend that you (and everyone else exercising the right) be
> > a good internet citizen. Cameron's point about respect[ing] "API
> > service limits where posted and develop[ing] polite tooling with
> > exponential back-off where appropriate" could serve as a basis for the
> > footnote.
>
> I agree that a note should be inserted to this effect.  But I think
> you should avoid trying to integrate (any version of) Cameron's text
> detailing what that means.  This is an email about principles, not
> implementation, and in this context it suffices to say we intend to
> play nice.  (Otherwise we're just inviting the publishers to get into
> a lot of irrelevant detail about what they consider a suitable
> exponential back-off factor to be.)  We want this to be short, sweet,
> instantly comprehensible, and incapable of misinterpretation.
>

I agree

>
> > The Un*x command that allows users to prioritise kernel time between
> > different users is called, fittingly, "nice," and I don't think it's
> > quite good enough to say that the technical details don't concern you
> > when proposing automated access to servers. After all, you couldn't
> > abstract a chemistry article that day if it was bound in a volume that
> > had been borrowed by another user.
>
> I agree that we can't say technical details don't concern us.  But we
> can say that they are not the subject of THIS discussion.  We can
> recognise the importance of that separate discussion without getting
> sucked into having it now.
>
> > The current tone of your draft betrays the frustration you have
> > experienced over the past few years, and while that is understandable
> > to me as a list member here, a more even and temperate tone that
> > acknowledges some of the potential practical difficulties would
> > perhaps be more effective. Murmur gently in a voice of sweet reason...
>
> Agreed.
>
> There's a paradox there: only years of abuse can give rise to the
> level of fury that makes letters like Peter's happen at all; but only
> a more sober tone than such fury expresses can achieve all that we
> hope.  So it's good that we have among us some people who have not yet
> been pissed of quite so royally as Peter.
>
> -- Mike.
>
> _______________________________________________
> open-access mailing list
> open-access at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-access
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-access/attachments/20120305/fac2f480/attachment.html>