On Nov 27, 2015, at 6:25 PM, Andrew A. Adams <[log in to unmask] <mailto:[log in to unmask]>> wrote:
> Richard Poynder wrote:
>> As a matter of interest, what is the average time span between
>> acceptance and publication? 
> I'm not sure an average number is really useful here. More useful but still 
> aggregate data would perhaps be something like each decile of delay, to pull 
> some random numbers out to llustrate only:
> 10% of papers are published with no significant delay;
> 20% of papers are published with up to three months delay;
> 30% of papers are published with up to six months delay;
> 40 % of papers are published with up to eight months delay;
> 50 % of papers are published with up to a one year delay
> etc.
> I would also find it useful to know about the difference between advanced 
> online release and final publication with volume and age numbers. I find it 
> very frustrating that a paper of mine which was accepted for publication in 
> January 2013 wasn't given a volume and pages until July 2015, a delay of two 
> and a half years! I was also not informed when it actually appeared.
> -- 
> Professor Andrew A. Adams     アダムス アンドリュー 
> Professor at Graduate School of Business Administration, and
> Deputy Director of the Centre for Business Information Ethics
> [log in to unmask] <mailto:[log in to unmask]>

Andrew Adams is (as usual) spot-on:

Date of publication, which is relative to date of acceptance, is extremely variable, can only be estimated on average, and is highly arbitrary.

This is another reason (if yet another reason was needed) to peg deposit date requirements to acceptance date and not to publication date:

The earlier the deposit date — as well as the OA date — the better (for research,  researchers, their institutions, their funders, the tax-payers that fund the funders,  and for the “public good”). The purpose of OA is to eliminate research access and impact loss in the online era because of needless access denial.

Another reason is that the Copy Request Button <> can start to provide “Almost-OA” as of the date of deposit, not before.

Yet another reason is that — whether they like it or not, and whether they admit it of not — virtually all journal publishers have by now conceded on “OA” after a one-year embargo (from publication date!): 

Hence the needless access/impact-loss figure to beat is P + 12.

It stands to reason that if P =  A + 3, 6, 12 or more then the needless  access/impact-loss becomes not just 12 months, but (unpredictably) 12, 15, 21, 33 months or more.

OA means immediate OA (or at least immediate OA + Almost-OA) to refereed research -- not embargoed access.

“OA” after 12, 15, 21, 33 months or more, on the other hand, is a not OA but a grotesque and cynical caricature of OA.

Hence HEFCE have adopted exactly the right mandate: 

1. Deposit immediately upon acceptance for publication  (with 3 month grace period) plus 
2. OA as soon as possible thereafter.  (Meanwhile, at least almost-OA immediately.)

Stevan Harnad

PS Another gratuitous source of delay is refereeing lag: Publishers’ subscription revenues depend on how many accepted articles they have (to sell). Like mammals and fish, publishers have a choice between the “K” (fish) strategy (invest in engendering as many offspring as possible,  without investing anything in to ensuring that the individual offspring survive  to reproductive  age — count instead on large numbers of offspring and some proportion making it through) or the “R” (mammal) strategy (engender few offspring and invest instead in nurturing them to make sure they survive to reproductive age. Many publishers processing submissions in an industrial way, not worrying that long refereeing lags mean lost research access and impact too. It is  undeniable that referees are a finite resource (and work voluntarily, and gratis) and that there are potentially huge numbers of submitted papers) but the revenue motive gives little incentive for adopting the R strategy rather than the K strategy.

The solution, of course (as I’ve too long preached to deaf ears), is for research institutions  and funders to mandate universal Green OA, which will make subscriptions unsustainable,  forcing publishers to minimize costs and down-size to becoming no-fault peer-review service providers and certifiers <>. irrespective of whether  the outcome is rejection, acceptance  or revision & re-refereeing. This service would be paid-for as a Fair-Gold OA fee out of a tiny fraction of institutional subscription cancelation fees — but Fair-Gold is only possible after universal Green has made subscriptions unsustainable. Till then, pre-Green (paid) Gold fees are Fool’s Gold fees: unnecessary, arbitrarily bloated and unaffordable  double-payment (while subscriptions still need to be paid).