As a co-founder of ScienceOpen, it gives me great pleasure to curate my first Collection of 23 Open Access (OA) research articles from different publishers entitled “Perspectives in Scholarly Publishing”. I created it using our recently launched tool for pulling together research articles around a specific topic, ScienceOpen Collections . Researchers with the role of Community Editor (available to those at all career stages. including earlier), can pick any paper from the ScienceOpen platform, which currently consists of nearly 1.5m imported OA works authored by almost 2.5m researchers and add it to the Collection. They can also fill any content gaps by calling for more articles, are encouraged to write an accompanying Editorial and comment on why they chose each article. What is the basic idea behind this new approach? In essence, it is to explore new ways to reinvent the classical journal concept. Journals were created at a time when the whole publishing workflow was based solely on print. They frequently had (and in many cases today, still have) a narrowly defined scope and their Editors had a “gatekeeper role” in terms of rejecting submissions that didn’t fit their niche interests. When the need for printed issues was reduced and even eliminated in the digital age, some innovative organizations created so called “mega-journals”, pre-print servers and repositories. Their popularity grew partly because one size of journal could now fit all subject areas. As the director of the Humboldt University Library in Berlin, Professor Andreas Degkwitz, said in a German newspaper article last year: “there is no longer a need to confine knowledge into the 'silos' of certain journal topics”. Meanwhile, more than 20,000 scholarly journals in the field of scientific, medical, and technical research (STM) publish about 1.8 million new articles per year - more content than any one person could ever follow! Sadly, fewer and fewer libraries can afford the subscription fees for these mostly pay-walled publications. Given that this trend has been accompanied by a rise in the amount of cross-disciplinary research, it is true to say that we need an alternative approach to gathering content together, preferably one that can flexibly assign articles to specific topics or themes of relevance to modern research. Moreover, we have to create a new process to help determine which articles from the crowd appear to be of higher relevance than others. Every one of us who has been working in research for a certain period of time has their own shortlist of the most useful articles. ScienceOpen Collections enable us to share that knowledge with others. We can also combine these articles with our comments about why we chose them, rate them and undertake post-publication peer review on them. We can encourage others to do likewise by inviting them to “Follow” a Collection. They will then be alerted to new activity and invited to take part in the resulting scientific conversation. In this way, ScienceOpen Collections are dynamic, not static and new articles can continue to be added. I would like to invite any of you to follow this Collection. Those who would like to experience the role of Community Editor are welcome to email me and express your interest. Together we can explore a new concept that is better suited to scholarly publishing in the digital age.