Re: Open Access publishing of Semantic Web research

Hi Aidan.
My thanks to you for writing such a comprehensive and well-elaborated 
article on the state of open access, and the various alternatives 
possible for organisers. This is really helpful to have. I've been 
mulling over these questions for a lot of time now.

In addition to your points, I am deviating the discussion a bit to also 
how this affects research and researchers in general. I appreciate you 
providing a hyper-focused outlook for this particular community, but I 
would also like to draw attention to some things we can do or think 
about for all communities and venues.

1) Preprints

Preprints serve two purposes AFAIK. One is to provide earlier access to 
work when formal publications can take time. Second is to provide access 
when the work is locked behind a paywall or is not accessible. We're 
increasingly seeing this become common as good practice, but it is still 
not uniform or widely adopted. In cases where we don't yet have a gold 
or diamond OA means accessibility to preprints is reliant on the authors 
good-will.

I suggest we raise the bar to have preprints be mandatory (e.g. in a 
conference state this as a required condition similar to author required 
to present the work). Authors upload their accepted work in a publicly 
accessible open repository (e.g. institution, arXiv, OSF) for accepted 
articles. In both short and long term, this provides a good way to 
ensure continuously generated knowledge is made accessible while the 
slow-movement towards actual OA takes its time. It also provides an 
always-open way to read the article on the conference website.

2) Why be a Reviewer

Most reviews are done without any benefit in return. We do this because 
we want to use peer-review to strengthen our knowledge by a rudimentary 
filter mechanism. However, as you rightly point out, if the end result 
is locked away, used only to increase profit margins, and can be easily 
removed any year - we are being taken advantage of.

I do not believe (only) compensating reviewers is an adequate or 
proportional option. What would 'fair' compensation even be in this 
case? A fixed amount - and how would that compare with our hourly rates 
based on salary? Such measures (compensation) have to be in-line with 
the demands of knowledge being open. Which basically means stop 
reviewing for non-OA venues. This is another way to 'encourage' adoption 
of OA.

And I have to emphasise here - having links to 'free/open' articles on 
conference websites should NOT be considered OA. Its either open for 
everybody everywhere or not (think internet pages being 'open'). We're 
the semantic community - we know what "identifier" means. So either the 
permanent identifier leads to OA from anywhere or its not OA.

This is another interesting area where Google Scholar (its amazing, but 
its also prone to peculiar shortcomings) will mark an article as OA when 
access was through a conference website, or in some cases - because the 
page got crawled by the Internet Archive (so you can get the PDF with 
wayback machine + conference website). So we should remember: "open 
access" does not mean "for the conference audience", but for all of society.

3) Should 'knowledge' be hostage to a third-party controlled URL?

In an ideal world, the article is open access which means anyone can put 
the DOI or PURL and get an accessible copy. In a less ideal world, this 
can break for several reasons: article was not published as open access 
in the past, there's a paywall for what was supposed to be open access, 
or a previous agreement granting access has been revoked.

In all these cases, we consider access to preprints as being a 
mitigating factor - since the same article can be accessed from another 
source. This is good. BUT. How does one find an authoritative preprint?

We go to Google Scholar, it pulls in any PDF that has the same data, and 
considers it an 'authoritative copy'. So I can post a copy of your 
article (with or without edits) on my website, and you can post your 
article in your institution's repository - and Google Scholar will 
combine them as being the same article. This is not to fault Google - 
indexing article PDFs has been an immense boon, but rather to state that 
this is a research problem that the semantic web community has answers to.

Quite simply: an article is an IRI, a Preprint is an IRI. How to enable 
the mapping between IRIs and to calculate 'trust' (e.g. institution 
repos or conference websites are more trustworthy than other websites). 
This is basically what SciHub does - it allows you to put in a DOI and 
gives back open-access versions of the same article. So why not use our 
community and expertise to index such 'legitimate' preprints into a 
KG-based community resource like WikiData or DBPedia? This will solve 
the discovery problem, and will also enable us to create short-term 
solutions while the long-term OA movement happens slowly.

[1] 
https://www.scienceguide.nl/2019/01/editorial-board-of-informetrics-resigns-at-elsevier-and-starts-new-journal/

[2] https://petsymposium.org/publishing_announcement.php

Thanks again for the article and raising this discussion.
Regards,
Harsh

On 03/11/2022 00:32, Aidan Hogan wrote:
> Thanks Axel!
> 
> A couple of people were questioning what the status is for OA on ISWC, 
> and indeed Daniel Schwabe (in copy) mentioned that he negotiated OA for 
> ISWC indefinitely in 2006. It seems that many assumed that ISWC was 
> still OA. However, recent years of ISWC have been pay-walled.
> 
> I checked through previous years of ISWC, and found the following:
> 
> https://docs.google.com/spreadsheets/d/1RU4rP5tg_IECE3KBeGdjbwCdcC0N0R5PYX2w9-Kv9gg/edit#gid=0
> 
> In summary, up to and including ISWC 2017 (with the exception of ISWC 
> 2004), SpringerLink offers free PDFs for all papers with subscription, 
> making all but one ISWC prior to 2018 Diamond OA (I don't recall any 
> author fees/APCs in this time).
> 
>  From ISWC 2018 on, most papers are paywalled on SpringerLink, and no 
> preprints were published on the homepage (in some years prior, including 
> ISWC 2017, they were). Only when authors have paid APCs, it seems, are 
> ISWC 2018+ papers available as OA. Otherwise all other papers appear to 
> be paywalled.
> 
> So if the analysis is correct, ISWC regressed from Diamond OA to Hybrid 
> OA (OA only when authors pay) in 2018.
> 
> Best,
> Aidan
> 
> On 2022-11-02 12:04, Axel Polleres wrote:
>> Dear Aidan,
>>
>> Congrats, great piece which I really enjoyed reading!
>>
>> Two small remarks:
>>
>> 1)
>> FWIW, I would add one more point in "Journal pain-points"
>> the strict deadlines in conference revieweing and - at least perceived 
>> - less strict deadlines in journal reviews, community journals slower 
>> overall in CS, compared to other communities. I have to admit that I 
>> only have annectotal proof for this, it was mentioned I believe in a 
>> panel on the issue with e.g. Moshe Vardi and Sweitze Roffel onboard at 
>> FLOC in Vienna's summer of logic a couple of years ago.
>>
>> I think this is an important factor across conferences and journal in 
>> a confrerence-culture heavy community like ours.... could be added 
>> under "Poor reviewer incentives:".
>>
>> 2) While sthey're still exclusively publishing with Springer, I found 
>> the 2-tier publishing model of BPM conference very interesting, where 
>> papers can be accepted as full conference papers (in Springer's LNCS) 
>> or in the "BPM Forum", see https://dblp.org/db/conf/bpm/index.html 
>> <https://dblp.org/db/conf/bpm/index.html> for less mature papers 
>> (published in Springers less prestigeous(?) "Lecture Notes in Business 
>> Information Processing" series.
>> This allows them to accept more papers, and would be combinable IMHO 
>> with a journal model for the top tier papers.
>> I find this model interesting to consider when re-thinking a mixed 
>> journal/conference model and it would be interesting to explore routes 
>> to combine with the VLDB model, or in other words would lend itself 
>> to a *combination of Option1+Option2* as one.
>>
>>
>> just my two cents,
>>
>> Axel
>>
>>
>> p.s.: also left this as comment on your blog, thanks again, also for 
>> the great gathertown discussion with Claudia and yourself at ISWC!
>>
>> -- 
>> Prof. Dr. Axel Polleres
>> Institute for Data, Process and Knowledge Management, WU Vienna
>> url: http://www.polleres.net/ <http://www.polleres.net/>  twitter: 
>> @AxelPolleres
>>
>>> On 02.11.2022, at 05:16, Aidan Hogan <aidhog@gmail.com 
>>> <mailto:aidhog@gmail.com>> wrote:
>>>
>>> Hi everyone,
>>>
>>> Inspired by discussions at the ISWC Townhall, and previously at the 
>>> JWS plenary call, I published a blog post here discussing the issues 
>>> of Open Access publishing in the context of Semantic Web research:
>>>
>>> https://aidanhogan.com/blog/index.php/2022/10/27/publishing-research-semantic-web/ <https://aidanhogan.com/blog/index.php/2022/10/27/publishing-research-semantic-web/>
>>>
>>> Also adding in CC members of the SWSA task force set to look into 
>>> this issue; as well as Uli and Claudia with whom we discussed this 
>>> issue for ISWC 2022; Pascal and Krzysztof as EiC's of SWJ; as well as 
>>> Jim, Zhixiong and Ying as EiCs of Data Intelligence. (Apologies; I 
>>> know some of you are already on the list.)
>>>
>>> The post covers the context of the OA issue, why there is a push to 
>>> move away from publishers like Elsevier (in particular) and Springer, 
>>> different types of OA, advantages and disadvantages of different 
>>> types of OA, precedents for conferences and journals in CS that have 
>>> moved to (zero-fee) Diamond OA and (inexpensive) Gold OA, 
>>> general issues relating to the conference vs. journal paradigm in CS, 
>>> precedents of CS conferences that have moved partially towards a 
>>> journal model, and a concrete proposal for how publishing research on 
>>> the Semantic Web could potentially look in future (as some food for 
>>> thought).
>>>
>>> I hope you might find the post interesting! And hopefully it's not 
>>> too long (I wanted to cover the issue in a self-contained and fairly 
>>> complete way for those also maybe new to the topic).
>>>
>>> Comments welcome!
>>>
>>> Best,
>>> Aidan
>>
> 

-- 
---
Harshvardhan J. Pandit, Ph.D
Assistant Professor
ADAPT Centre, Dublin City University
https://harshp.com/

Received on Thursday, 3 November 2022 16:01:00 UTC