I interested. I have some research in the field of nonlinear control theory. I used Julia to create simulations and try out ideas. Julia helped me a lot. I need some time to write the actual papers, but I really interested to submit those research, and celebrating Julia 1.0.
Thatās right! Iām an editor for Ecography, which is a fairly high-impact (IF ~5) journal focusing on (macro-)ecology, and Iāve been thinking about for a while putting together a special issue on Julia packages for Ecology. Itād be a nice idea to have several special issues in specialized journals - e.g. I doubt anything that would fit in Ecography would also be relevant for āAdvances in Engineering Softwareā. But weād need at least 4-5 different papers on packages that havenāt been explicitly presented in earlier papers. I also know that the BioJulia crew has talked about approaching PLoS Computational Biology with an idea for a special issue on bioinformatical tools in Julia (maybe they talked to you, @tpoisot ?)
They did not ā but Iād be happy to assist if they did. I would be curious to see the reception of a Julia special issue in Ecography.
But for the purpose of the Advances inā¦ special issue, I think a discussion of the outstanding computational challenges ecologists face, and how Julia can help, may be relevant.
The special issue is essentially greenlighted by the EiC, but at the end of the day itāll depend on us having a critical mass of high-quality weighty contributions in order for it to merit a full special issue. Because of this Iād be quite keen to try to aim any ecology contributions towards Ecography. Unfortunately I think EcologicalNetwork and BEFW already have MEE papers attached to them, right?
I also have an invitation to write an editorial on the possible impact of Julia on how we work on data analysis in Julia, but it has to be submitted quite soon. I was planning to invite you to co-author - will follow up over PM.
My recollection is that we started talking about this and then it petered out because none of us felt like what we had was baked enough for a publication yet. A lot of the effort in the last few months from @Ward9250 and @bicycle1885 has been on the modularization of BioJulia, which will make development much nicer going forward but doesnāt really have user-visible impacts in the near term.
2 Likes
While I am supportive of this, potential authors should be aware that like most Elsevier journals, there is a hefty open access fee (USD 2500), otherwise there is an embargo period of 2 years before you can make the final version publicly available (though it appears you can make the pre-review preprint available): see Open access information - Advances in Engineering Software - ISSN 0965-9978.
Also, I would like to encourage authors to post citations of Julia-related publications to https://julialang.org/publications/
8 Likes
These are all valid remarks.
Additionally, nothing prevents the author from developing the sources to the submitted article as a public repository on github. I think this would be very much in the spirit of developing with Julia!
2 Likes
Sounds great.
Precisely. Which is why it is very unlikely that one would find in the wild pure-language codes to compare.
I do hope a remark like this will not end up stopping momentum for a great project like this one, which has real potential for advancing the use of julia in scientific applications.
Just to make it clear, because it is easy to get that impression from the remark, Advances in Ecological Engineering is not some predatory journal charging unusual fees. The non-profit journal PLoS Biology charges $2900 per article, and that journal runs at a deficit due to the high quality of its publishing services (it is kept alive by mass-publishing outlet PLoS ONE).
There is a big current discussion about the best format for academic publishing, unfortunately rarely discussed at a level that appreciates the complexity. The biggest discussion is between a pay-to-read model or a pay-to-publish (leadingly called āopen accessā) model, both with fairly large inherent problems (that Iām happy to discuss, but not in this thread).
If the julia community is going to take a stand in this debate I do hope it will happen after a discussion that is informed and presents the relevant stakes. Perhaps a short debate here on discourse could be beneficial, but in another thread. Letās remember it has big signal value what Julia āStewardsā remark here.
I know that Dutch-based Elsevier has a very bad reputation in the US, but most other major publishing houses (Wiley-Blackwell, Springer, Nature publishing group) have identical business practices (the for-profit āopen accessā journal Nature Communications charge $5000 per paper - but then they salary their academic editors).
(Iām not saying this in defense of my suggested special issue in Ecology. Ecography is published by Wiley and publish Software Notes free for both author and reader, in recognition that we require software to be both open-source and non-commercial. I strongly pushed to implement this policy. The publishing costs are indirectly paid by the other papers, of course, so this model is not open for an only-software journal).
If julia wants to become the software of choice for scientific applications, a well-curated special issue in a key journal is indispensable.
8 Likes
Sorry, my tone was perhaps a bit harsh, but it was not my intention to derail the endeavor, just ensure that any authors were aware of the policies. Obviously it is in the interests of the project to encourage to have more high-quality papers about Julia in different domains, and special issues are indeed a fantastic way to achieve that.
2 Likes
As long as weāre discussing paywalls knowledge is a public good and the only long term solution is for high quality open access journals to be funded by the public
2 Likes
Much of my work with BioJulia the past few months has been both organisational, as well as in work demonstrating how use of BioJulia packages and algorithms can really speed up many of the very basic operations required by anyone studying population genetics with biological sequences, in some cases way over a predicted 64 fold. There is still a question on how to publish this, and I have to talk to my supervisor, as itās a question of, well does this go in a āBioJulia paperā - one introducing the project as a whole, or a specific paper about a specific application, and so on.
1 Like
The Journal of Open Research Software appreciates discussions of scientific software architecture and organization.
2 Likes
Hi, I would like to contribute for Article under title "Use cases of Julia Programming for āData Science as Serviceā which will be inspired from following article:
That is an interesting idea. Just keep in mind that Advances in Engineering Software is a research Journal.
I think the purpose of papers in AES is to present your own original research. If you develop a new algorithm or a tool making it possible to solve a known issue in an innovative way, and implement that as a registered julia package, you can use the journal format to publish a paper describing your advance. This will also give other researchers a citable reference if they build upon your research (or just use your tool).
What youāre descibing there represents the research of the developers of julia, and could be described by those developers as novel research, but not by others. It sounds like it could make a very interesting blog post, though, or paper for a popular journal.
3 Likes
I have some questions about publications in AES. I never make publication in Elseiverās journals.
I have 4 original research. Those are in the field of Fixed Point based Non-Linear Adaptive Controllers. The hart of this type of controller is the Transformation function. I developed four new Transformation function. Each working on a SISO system. Then I generalized it to MIMO systems.
Some places the rule is to put the SISO and MIMO system into one paper. Some places prefer one paper for SISO solution and one for MIMO solution. In AES what is the preferred way?
If some places allowed to have exact same parts in two separated papers in the Introduction part (most are copy/paste). It is trivial the main part has to be original.
Some journals it is not allowed, this flagged as self-plagiarism (the reason is, the plagiarism checked by purely with a computer). What is the case in this Journal?
Can you approximately tell me the deadline? (One weak, One monthā¦etc)? Basically, how many time I have?
Usually, in scientific papers, we donāt put any āimplementationā part,
because this will be a Special Issue to celebrate Julia 1.0, maybe it is a good idea to put some key points into the paper, make a short section, how easy to do it in Julia.
There is some restriction on the size of the paper? (max, min value) ?