The I think you will agree there are many problems with both the open access and closed access models of academic publishing. The closed access model forces the reader to pay (often within the region of US$25) per article, while the open model forces the author to pay (often within the region of US$1500) per article. The ACM seems to have found an excellent middle ground with their new Author-Izer service.
This new service offered by the ACM enables authors to place a link on a personal website which resolves to a specific PDF document lodged within the ACM Digital Library. Readers who select this link are then not charged to view the PDF and indeed by selecting this link you get free and open access to the publication for as many views as you like. All the statistics for downloads and citations are still recorded and so the service has an advantage because authors are able to show public/professional interest in their work; via the download statistics.
This works to the ACMs advantage too because now there will be huge amount of Web referrers to the ACM Digital Library – as the single authority or hub on the web for computer science academic journals. This can only make the popularity of the ACM increase and will, I suspect, not impact their revenue streams too hard as most institutional libraries will still wish to pay for the full-service, including some committed individuals still paying the small US$99 fee for a years unlimited access to everything within the ACM Digital library.
The people who will really benefit will be academics at smaller institutions without institutional access and those with a passing interest in the work but without the means to purchase the closed version. Indeed with this kind of open access we may also see interested members of the public or the media using the service to pursue their interest, if not their profession, in computer science.
Full credit to the ACM for this excellent step forward in open publishing.
Bugs Become Features... 
I never understood why open access publishers get away with 1500 dollars per article. JMLR, the leading journal in machine learning is free all round. MIT donate the server space. The problem seems to be with publishers, not with the open access model.
I also am very sceptical of anything coming from the “Learned Societies” they are large organizations with large costs to cover. I think they bring very little benefit alongside an enormous overhead.
In the long term I see no reason why all articles (in computer science at any rate) aren’t available for free. If you use latex there is no type setting cost, the main “cost” is reviewing and we donate that for free as well. It sounds like this project is to stop the inevitable happening and keep the ACM in the loop. If you want a digital archive system, why not use arxiv (also available for free).
Hi Neil,
I see your point, but as you imply, someone needs to pay for – at the minimum – server space and infrastructure such as citation counts, downloads, metrics, and guarantee the archive will remain available over time – oh, and also the DOI purchases. The ACM is a Learned Society but one that I think works for it’s members – I’ve also no problem with some administration overheads as long as they are not over-blown. While you may use LaTeX (as do I) this is not the case for all – and I’ve seem some terrible ‘typeset’ and typographically error prone drafts which have to be later proofed (a job I would not want) to be included in the literature.
It seems to me your assumption is that everyone is a conciousness good citizen as you are – this is not always the case. While voluntary academic review takes care of the science – it doesn’t in most cases take care of the language, incorrect typesetting etc.
This said, I agree that the large publishing organisations (not the Learned Socs.) need to come to terms with a reduction in profits and a new business model, but as downloads become evermore important I also still want infrastructure and a central download count, beyond just citation.
Finally, arxiv isn’t bad although all of our work including code, data, analysis and the technical reports – from which we create our papers – are all open and available under creative commons on our own ePrint server – however as with arxiv, there is no review process.
I’ve more to say – let me continue via your next post…
As a CS researcher, I feel proud that the ACM has developed this initiative; as an OA advocate I am astonished that they have come up with a new take on the issue!
However, I am not at all convinced that it will work in practice. As a “Web Ergonomics” expert, could you comment on the user engagement that the Author-izer process demands?
Hi Les,
I think the process both ways (author, and reader) are pretty straight forward. ACM authors login to the Digital Library (Free and Open), they select the paper they want to provided access too, copy the links provided, and paste them into their target page – no more difficult than my reply to this post. The reader just selects the link provided by the author on their site; no more difficult than selecting this post to read.
For the author I see big benefits because, even though the ACM allows a copy of any published paper to be held on their own site, a download of this paper will not register as a download in the ACM metrics – and I think downloads are a new metric which are a way more telling measure of ‘reach’ and influence than citation count alone.
I also see benefit for the ACM in that they will still be able to sell their organisational subscriptions (their main DL revenue stream), along with a years full access to everything for $99 (US) to researchers without organisational access but who require a premium service.
Those of us who are occasional readers, or non-academics wishing to pursue a line of enquiry can use Google Scholar to find the work, and the free to download version to read. Indeed, as the abstracts and citations are all free via the DL – authors who have not provided Author-izer links can be emailed by the reader to request free access.
There is a long post about this step (which expresses my concerns much more eloquently than I did) here: http://blog.dshr.org/2011/10/acm-copyrights.html
Hi Neil, and thanks for the link, it’s an interesting post but before I comment, and in the interests of full disclosure, I’d like to declare that I’m not much impressed by the author beating-on the ACM publishing policy especially when there are more deserving, and harder, targets in the publishing world – and in this case when actions to correct errors in copyright procedure seemed to have been taken in a timely manner with ACMs mistakes being freely admitted. Now on to the meat…
Seems to me we as Computer Scientists have a number of choices, we can publish reports on arXiv, or indeed everything we produce reviewed or not; we could set up our own open-access journals; or we could take a trip down the yellow brick road with the Wizard-of-Oz as per last months CACM Viewpoint. However, in all cases we don’t have time to do everything (do we) so we’ll need to employ someone, and therefore incur a cost, which someone will have to pay.
And this is the point, as a researcher my capital is data and the research papers based upon it, anything which diverts my attention from generating and disseminating data is a cost, and so we employ institutional administrators to deal with the day-to-day clutter which detracts me from generating my capital.
In the case of publications, we don’t want to do the administration, so we’ll have to employ someone to look after the journals (day-to-day) we’ll be creating – how will they be paid?
Now don’t get me wrong – I’m not arguing for closed access – I support open access when practical – and pragmatic access in other cases. I think the ACM are treading a middle ground and I’m all for this approach in the absence of something better which benefits all employees (not just us CS academics / authors).
I also have 3 pragmatic steps to open access…
However, if we expect publishers to move to open access (thereby threatening their currency) when we better make sure we make our currency (data and its analysis) open and available too; publication or not.
Kudos to ACM for implementing the Author-izer!
The American Educational Research Association has actually been providing such tollfree hyperlinks to its authors for the last 4 years since we began publishing through Sage Publications. The HighWire Press platform that Sage uses supports these links, and on Sage’s recommendation we began offering them to our authors. In addition to numerous individual authors, at least one institutional repository has accepted a tollfree hyperlink to an AERA journal article in lieu of a PDF posting. In our case, the links lead to the fulltext PDF and HTML.
i believe there is a handful of other publishers on HighWire Press who also offer these links, and I am happy to see additional platforms beyond HighWire beginning to support these.
Now if only we could use ACM’s great name for the service!
Todd Reitzel
AERA Director of Publications
Hi Todd,
Thanks for your comment – I do like the HighWire concept, and I’ve looked at your solutions in the past for possible journals we have been looking to start in Web Ergonomics – however, as far as I can tell there is a cost attached here too? I wonder how your funding model compares to the general publishing model?
Cheers
Si
Pingback: ACM ASSETS 2012: The Best Paper According to Me! | Thinking Out Loud…
Pingback: ACM Open Publishing | Thinking Out Loud…