Is anyone familiar with a scientific publisher that can host a large static html file and provide a DOI for it.
I looked in many places, and so far found only a few publisher that can handle html files, yet those all have one of the following limitations:
The publisher is limited in scope to AI related papers and cannot expand to other fields
The publisher will store the html file, yet will not serve it so people can download the file, yet clicking the link will not render the interactive content
The publisher cannot assign a DOI that will link to the paper.
Hopefully someone knows of a publisher that can provide a service. The reason I am looking is since I have a COVID-19 model paper that i interactive - meaning that the user can interact with the figures. The file is static, meaning you can load it in your browser without needing internet. Yet it seems no scientific publisher can handle this.
I contacted CrossRef to get a DOI and self publish the interactive paper and they referred me to this forum. Hopefully someone here has an idea in mind.
Jacob Barhak
–
Jacob Barhak Ph.D.
Sole Proprietor, Software Developer, and Computational Disease Modeler
Zenodo will archive the html file and give it a DOI, yet when you search for the DOI, you will not be able to see the html paper - you will see the html code at best and need to download the paper file to your computer and load it from there.
This is the second problem I mentioned in the list of problems.
I contacted Zenodo and they confirmed that they will not serve the html from security reasons.
However, the html file I create is harmless - it does not even connect to the internet unless you click on a link of a reference that is processed through DOI for most of the references- it just create interactive plots using HoloViz technologies that can create those interactive graphics. I know it is safe, yet I cannot find a publisher that can publish it - I have been looking for weeks. Here are venues I checked:
distill - it is the best choice yet they focus on machine learning and cannot handle topics like modeling - they tried in the summer, yet had editorial issues.
visxai - it is conference base that happens once a year one has to wait and it is based on AI…
parametric press - it is a magazine that requires a specific format. They are pioneers.
f1000research - they limit the interactive graphics to plotly - cannot handle other html technologies
zenodo- Will archive the paper yet will render the html - will show the html code instead. It is set up as a preprint rather than a scientific peer review journal - although I recently learned they provide options for peer review.
figshare - will assign a DOI, yet they store the html and did not reply to my query when I asked about actually serving the html - I did not see any example that shows the paper. It is not setup as a scientific peer review journal
Netlify and GitHub.pages will store the html and serve it, yet will not give a DOI and those are not set up as a scientific peer review journal. They do not intend to provide DOI soon according to this discussion
SimTk that is NIH funded can give a DOI yet will not store or serve the html file and is not set up as a scientific peer review journal
I also tried some traditional publisher like Springer/Nature, MDPI , yet they only allow supplementary material at best and I want interactive figures imbedded in paper.
It seems my only option is to self publish, yet CorssRef is reluctant to allow me to register and sent me to this forum to see if there is a solution. Hopefully you will realize that a new type of publisher is needed and allow me to register as a publisher for a singe DOI.
Hopefully the need will be understood and CrossRef or another entity will allow publishing this paper in a proper manner.
That’s quite the comprehensive list! What about getting a citeable DOI by depositing a a PDF at https://www.biorxiv.org or https://www.arxiv.org, and then the HTML either as supplementary materials to the PDF (if they will serve those as web pages) or at one of the other sites you listed, linked from the PDF? It would be nice to have the DOI and the HTML bound together at a single site, but failing that, seems like a DOI that links to the HTML is a good fallback.
First, readers in this list have to know that you are in charge of Holoviz - it is appropriate in this public forum.
I followed your advice in the past and many journals provide the ability to store supplementary material.
However, I decided that this paper I am trying to publish should be published interactively so that the link would point to the article with the interactive content directly.
Many readers may not be aware of supplementary material and may not visit the link in the paper. And if the paper has to be downloaded to be viewed and does not appear immediately with the interactive content, some readers may be confused. In this particular paper the interactive content alone tells a story that may be much better than the text. I want the reader to be transported with the DOI link to a paper that they can scroll down to the interactive images and look at models and results visually quickly. They say one image is worth a thousand words - I believe an interactive image is worth much more.
The current technical difficulties of the scientific publication industry to implement such capabilities should not stop scientific publication and indexing of interactive papers. The HoloViz technology your team develops already allows it - the current barrier of publishers to host it can be easily fixed with minor effort.
My request from Crossref to allow me to register as a self publisher for the sake of a single DOI assignment is intended to open this possibility. I want to publish an interactive paper and no willing publisher can currently handle it as far as I know. It seems there is a need for a new publisher that can handle interactive scientific papers.
Hopefully this discussion will lead to a quick solution.
As far as I know this is the first interactive paper that was published in the modeling field and there was no venue that can publish it.
I have to thank flashpub.io for providing the DOI service - they went out of their normal operation mode to provide a DOI for this manuscript since there was no other solution available and they understood the value of being able to interact with a manuscript. .
I am sending this to this forum so that other publishers will open up and support publishing interactive papers - in many cases one picture is worth a many words, and an interactive figure can be worth so much more.