Starting a new journal? Which hosting platform is the best one for your journal?
Check out this comparison of 10 different open-source platforms, and their workflows, features and costs:
An excellent piece Judith. We, at ScienceOpen, also offer a publishing solution (not open source though) for journals AND all content is available immediately after publication in the Search & Discovery environment of over 88 million records - Stuart
I came across this from a Google search, because I had no idea this community existed. We (iacr dot org) spent quite a bit of time evaluating the different open source platforms in 2023. We also talked to some commercial publishing services, but ended up being dissatisfied with the options so we decided to build our own. Part of the reason was because we use a nonstandard peer review process that synchronizes submissions four times per year, and uses a committee to perform peer review. This kind of peer review is more and more common in computer science, but no publisher offers such a thing. Luckily there is an open source solution (HotCRP). We are also in computer science and mathematics, so authors submit LaTeX. This turns out to be an advantage because the author is already submitting a structured document and it’s easy to style it for the journal without a lot of human effort.
The system is nearly complete now, and consists of three parts:
- a reviewing system (HotCRP) with minor modifications
- a production and copy editing platform in python. This is similar to ACM TAPS.
- a hosting and indexing platform in python.
The total for #2 and #3 is under 15,000 lines of code in python. HotCRP is 160K lines of PHP code. By contrast, OJS is 1.4M lines of code and is still lacking features we need. I plan to make our system open source this year once I work out some licensing and modularization issues. In my opinion the three pieces of a publishing workflow should be plug-and-play, with well defined data requirements to pass between them.
I can only imagine the pain felt by fields that are dependent on Microsoft Word documents in OJS. I came to the conclusion that anyone who wants to build a new system for open access publishing should focus on minimizing the amount of human effort. This produces inefficiencies and it’s where the money goes in publishing.
The link returned 403 Forbidden
Do you mean the link to the article I shared? I can access it ok today. How about this link to the PDF itself? https://hcommons.org/deposits/objects/hc:59232/datastreams/CONTENT/content
OK, thanks for the pdf
I discovered the link is now active