American Chemical Society AMA: I am Stefano Tonzani, Executive Editor of ACS Omega. Ask me anything about the publishing process!

Abstract

Hi Reddit,

I am Stefano Tonzani, Executive Editor for ACS Omega (http://pubs.acs.org/journal/acsodf ), the new multidisciplinary open access journal from the American Chemical Society. My background is in science, as a chemist/physicist (simulations of electron-molecule interaction phenomena, self-assembly, and nucleic acid dynamics, to be precise).

Through my career in publishing I have worn many hats: editor, business development manager, technical sales, you name it. I am clearly having fun, as I have been doing in publishing for 8 years now. My first role was as materials science Editor for Nature magazine; there and in successive gigs I quickly discovered a passion for developing new products, which has traveled with me ever since.

Ask me anything about science publishing, what Editors look for in manuscripts, career advice, open access, starting new journals, and of course my latest baby: ACS Omega and what we hope to accomplish with it.

Time is over for this session: I hope to talk to you again soon!

What is the editorial process at ACS for table-of-contents graphics? Is there any review of them whatsoever? You get some pretty wild stuff on ACS journals as TOC graphics. Have you ever had an editor reject one for being offensive or too silly?

iorgfeflkd

The ToC graphic is built as part of the paper, so Editors and referees see it when shuttling it through peer review, and can block images they find offensive or scientifically misleading. However, we do not claim that ToC graphics are peer reviewed, as their purpose is to "give the reader a quick visual impression of the essence of the manuscript without providing specific results". Our production team, after acceptance, also strives to prevent offensive and copyrighted material from being included in the eToC.


I've read a lot about the problem of data manipulation such as p-hacking in order for scientists to improve their odds of publication. How seriously do you think this is currently affecting the literature? Do you think the industry is capable of overcoming this?

that-is

Selective reporting of data is a problem. There are a number of initiatives ongoing to deal with it. For example, a couple of companies (researchsquare.com and peneloperesearch.com) are using different approaches to improve reproducibility; publishers (Wiley and Nature come to mind) are also working on this same problem, which is seen as important especially by funders.

I think the industry can overcome this issue, but the main problem is of incentives and author behavior: publishers, funders and institutions need to foster a culture of openness in reporting, giving authors strong incentives to do so, and making it easy to comply. If it's not easy to report all data without bias (easier than continuing to report only the data that fits the hypothesis), authors are not going to do it.


I wonder if you could talk about access, particularly how the rise of open access has affected your journals approach to pay walls. Additionally, I'm curious if you can address the peer review process, and how it could be improved or streamlined.

Thanks!

Izawwlgood

Open access is the present AND the future of scientific publishing. ACS Omega is a born-open access multidisciplinary journal: in a field such as chemistry, which is not at the forefront of the OA movement, this would not have been feasible say 5 years ago.

I have worked on open access titles for quite some time, and I think this is a useful business model, which offers authors many advantages, such as higher exposure and, seemingly, higher citation counts. That said, coming to grips with open access has been, for some publishers, a path through the five stages of grief, whereas others have been much more eager to embrace it. Speaking from personal standpoint, I think one of the most interesting effects stemming from OA on paywalls is the fact that several publishers are now offering ALL their content to their subscribers, as opposed to individual titles, or picking and choosing. This will do marvels for access counts to all paywalled research.

Peer review: it is not perfect, but are non-peer reviewed articles more reliable? On average, I would think this should be a resounding "no". How can we improve it? Avoid reviewing ex novo the same paper at a different journal after it has been rejected by a first choice journal: reviews should travel with the paper through its route to publication. Across publishers if needed. I am a believer in open peer review, in the sense that I think reviews (not necessarily the names of peer reviewers) should be disclosed when a paper is published: this has the side effect of showing very clearly which journals are serious about peer review, and which aren't. I also think that peer review could be complemented by other (potentially commercial) services which check things such as reproducibility more deeply.


having published in open access journals myself (pleasant experience) I was wondering, how does the business side of open access journals look like? are the profits better, equal or worse than the traditional journals behind paywalls?

I fully understand if you are not allowed to answer this question

nixielover

I cannot speak to the detail, but I will note that there are a number of purely open access publishers, and they are doing just fine (the reputable ones, not talking about predatory publishers), so open access can be a perfectly profitable business model, just as subscriptions are.


How seriously does your journal conduct peer review and what steps do you take to prevent fraudulent articles?

razerxs

Very seriously. The tagline for our publications is "most trusted, most cited, most read", and maintaining our readers' trust in what we publish is fundamental for us. Our level of retractions is consistently low, which I think speaks to a fair but thorough peer review process.

There are a number of things we do to prevent fraud. Of course having a robust peer review process is number one, but we also aim at rooting out plagiarism as much as possible by passing all manuscripts submitted to ACS journals through anti-plagiarism software to detect text copy-and-paste. When ethical issues are discovered, we aim first to educate authors about what is scientifically ethical behavior. In cases in which authors are found to be willingly engaging in unethical practices, or they are recurring offenders, more drastic measures can be brought in, but I would like to stress that our main aim is to educate authors.


What steps do you take to avoid problems like those described in How to Publish a Scientific Comment in 1 2 3 Easy Steps?

filletrall

The linked piece is hilarious. Most journals now are online-only, so they (fingers crossed) do not impose restrictions on the length of comments/errata. This is anyway the case for ACS journals, as far as I know. Many disciplines now have preprint servers, which are very useful for this type of activities (chemistry will soon have one as well). Some journals are hosted on sites which have commenting features: this can also be a quick way to update the record. My advice is the following: if within two months from submission of your comment this has not been peer reviewed and given the thumbs up/down based on MERIT (not formatting), try and publish it elsewhere.


What are your thoughts about the increase of "clickbaits" in journals. One example that comes to mind is the last issue of Energy and Environmental Science (Impact factor: 25). Their titles are somewhat ridiculous sometimes.

Using words as novel, promising, critical, excellent, unique... in the title. I know some journals that actually forbid these words. Do you think this is hurting the publishing environment?

desconectado

I am not a fan of clickbaiting in general, and very much against it in science. At ACS Omega, we do not explicitly prohibit these words in titles, but our co-Editors and peer reviewers routinely do comment on (and object to) their improper use. A title is the best mean to make your paper retrievable through search engines (which is how 90% of readers will come in contact with it): if you use unspecific words, nobody is going to read your work.


Hello!

I'm currently pursuing BSc. Chemistry and I wanted to know two things:

Have you ever been in a situation that you had a reject a simply amazing paper because of language and/or grammar issues or something similar that couldn't be resolved? How did it feel?

What's the one tip you want people who are publishing for the first time to know?

Chai4You

1) No. Generally, if the paper is written very badly (reject-level badly), then the science is spotty as well. If there are language difficulties but the paper is very good, most editors will make suggestions (commensurate to the amount of time they have!), refer authors to language editing services, or tell authors to ask peers to take a critical look at language in the paper.

2) Compare, compare, compare. There is a lot of published research out there, chances are someone else has done work related to yours. Cite those papers, and tell your readers what is the improvement/step forward/increased understanding deriving from the bit of research you are describing in the manuscript.


How to find and select a journal for publishing a paper? How much should a researcher care about journal metrics such as impact factor?

mhrn110

Submit your research to a journal you actively read and that is appropriate for the scope of your paper: chances are many other scientists in your field will read it too, and engage with your research (which is the purpose of publishing a paper, by the way).

Metrics: many authors have direct incentives to publish in high(er) impact factor journals. Frankly, choosing a journal over another JUST BECAUSE its impact factor is .001 higher (real life story) is downright silly. Overall, metrics only measure certain things, and those things are not necessarily the ones you care about.


Right now, not many studies can be conducted solely to reproduce others' results. For one, they are less funded, and additionally they are perceived to have a lower chance of publication. How do you think publishers can begin to encourage studies trying to independently verify others?

Hrothgar_Cyning

Publishers already have taken steps to improve the chances of such studies being published: there are all sorts of journals now which judge research on quality as opposed to impact, and one side effect is the possibility to publish papers which show that previous literature is incorrect. At ACS Omega, we welcome this type of manuscripts.

As I mentioned in another comment above, scientists can take steps to improve the reproducibility of the works they publish, by pairing peer review with additional services tailored specifically to this end.

Research is expensive, and it is up to funding agencies to figure out how to incentivize their own researchers to publish more of this sort of results. Fostering a culture of openness among researchers will definitely help incentivizing disclosure of negative results.


What is your opinion on anonymous post-publication review websites like pubpeer?

ekpg

Pre-publication and post-publication peer review should be integrated. I do not believe in post-publication peer review as a standalone. Getting time out of the day of GOOD peer reviewers is hard, and if the referees are either not good or asleep at the wheel there is not much peer review going on, unfortunately. Good reviewers only lend their time if there is an incentive for them to do so. That said, if you are doing pre-publication peer review, you should: A) do it quickly B) allow publication in preprint form, as ACS Omega does.

Points A & B negate, in my view, the need for standalone post-publication peer review.


Additional Assets

License

This article and its reviews are distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited.