Is PDF the MP3 of Research?
Exploring the parallels between the digital transformation of music and research, and why research hasn’t yet reached its ‘Spotify moment’.
My Mondays at work always start with Spotify’s Discover Weekly. Kudos to Matthew Ogle and his team for creating what is perhaps the most seminal example of how content personalization can truly empower listeners. In a world where the abundance of music can lead to choice overload and hidden gems often go unnoticed, Discover Weekly stands out. This personalized playlist offers users a curated selection of 30 tracks each week, tailored to their listening habits. It’s a masterful blend of Collaborative Filtering for personalization, NLP and audio analysis for categorization, and a dash of serendipity to keep things fresh and surprising.
It’s not perfect, and that’s part of its charm. Some weeks, the playlist might feel irrelevant or dull. I often skip a handful of tracks, but I usually add a few to my Liked Songs, and I find myself returning to the playlist multiple times throughout the week. The secret to Discover Weekly’s success lies in its clever mix of algorithmic content personalization with principles of behavioral economics, such as the First Start principle, Varied Rewards, and the Endowment Effect. Launched in July 2015, Discover Weekly quickly became Spotify’s most popular feature, boasting 40 million active weekly users and inspiring spin-offs like Daily Mixes and Release Radar. Clearly, I’m not the only fan (just liked a new track: “ Praia — Yuksek Remix”).
Spotify is today the giant of online music streaming, the culmination of the digital revolution in the music industry. The MP3 format, introduced in the early 1990s and standardized by the end of the decade, revolutionized how music was consumed, shared, and pirated. With its ability to compress audio into manageable sizes without sacrificing too much quality, MP3s opened the floodgates, shaking the music industry, its traditional models, and its stakeholders to the core. The MP3 was a game-changer — a true disrupter.
I’ve closely followed this disruption as a music lover, academic, and later, entrepreneur. In 2012, along with students of mine from the University of Thessaly, we co-founded two interrelated music startups.
With Christopher Varakliotis, Argyris Stergiou and Christopher Dimitriadis, we launched Noodio.fm, a music streaming service that functioned as a social radio, supporting emerging musicians and enhancing music discovery for listeners. Using an innovative algorithmic voting system, Noodio.fm highlighted talented artists, helping them reach local audiences and organize events.
In parallel, with Eleni Koumara, we created StageDiving.gr, a platform to help bands promote their gigs to the right audience using augmented online posters. These posters combined the aesthetics of traditional print with additional multimedia information about the artists, such as bios, photos, and videos.
Both Noodio.fm and StageDiving were genuine efforts to help musicians navigate the evolving music industry landscape, and we learned a great deal from their rise and fall. Ultimately, it was Noowit — the flagship of my entrepreneurial endeavors — that was acquired by Atypon/Wiley and transformed into Scitrus. Noowit was the first web app to generate intelligent web magazines tailored to individual user interests. Scitrus became “Noowit for Research,” or as Georgios Papadopoulos, the founder and CEO of Atypon at the time, described it: “a Flipboard for Research.” Scitrus was an AI-powered app for personalized paper aggregation, designed to help researchers cut through the noise and discover relevant research. It included a full-fledged reference library that could be shared among researchers and even featured a Chrome plugin for annotations and comments.
When Christos Spiliopoulos, Christopher Varakliotis, and I moved to Atypon and anchored its Thessaloniki office in January 2017, the digitization of research was well underway. The number of Open Access journals was rapidly increasing, posing a challenge to the traditional license-based revenue models of academic publishers like Elsevier, Springer, and Wiley. It quickly became clear to me that music and research were following parallel paths, albeit with research lagging a decade or so behind. Just as MP3s brought music to the masses, PDFs have done the same for research. The Portable Document Format (PDF) became the standard for compressing and disseminating academic content, lying at the heart of the transformations in research publishing.
As I’ve transitioned from music and news to research publishing, I’ve noticed striking analogies between the journey from vinyl to Spotify and the shift from print journals to Open Access, preprint archives, and beyond. These analogies between two seemingly unrelated paradigms offer valuable lessons, and I’d like to share them with you.
From Vinyl to Digital
In the beginning, there was vinyl. Vinyl records, introduced in the late 1940s, dominated the music industry with their tactile, high-fidelity experience. Owning a record was more than just listening — it was about the satisfaction of collecting, admiring cover art, and the ritual of playing the music, creating a deep connection between listener and sound.
In academia, print journals held similar prestige. For decades, they were the gold standard for research dissemination, symbolizing scholarly achievement. Libraries meticulously cataloged these journals, and researchers cherished their physical copies. Like vinyl, print journals represented an era where access was exclusive and highly valued, with the physical medium deeply influencing the experience of the content.
Let There Be CDs & MP3s
In the 1980s, vinyl records evolved into CDs. These new discs were smaller, lighter, had greater storage capacity, and offered improved sound quality — though some audiophiles might debate that last point. One thing was clear: CDs drastically reduced production costs, from $5.00 to $10.00 per vinyl unit to just $0.50 to $1.00 per CD. This cost efficiency propelled music distribution into a new era. By the end of the decade, the shift from analog to digital was in full swing, with over 200 billion CDs sold globally. Yet, the true digital revolution would only come when the music stored on those CDs transcended its physical form and found its way onto the web.
The MP3 format was officially released in 1993, as part of the MPEG-1 standard, which included video and audio compression technologies. By early 2000s MP3s have taken the world by storm. With their small size and decent quality, MP3s made music infinitely more shareable. The internet, particularly through platforms like Napster, exploded with music swapping, much to the dismay of the record industry. Suddenly, anyone with a dial-up connection could access a world of music for free — legally or not.
That same year, Adobe launched the PDF (Portable Document Format) as part of Adobe Acrobat 1.0. Research began its slow migration online, and PDFs quickly became the standard format, preserving the look and feel of print journals while offering the convenience of digital access. As journals began digitizing their archives, PDFs became ubiquitous. Researchers could now easily share papers via email or academic networks, bypassing traditional paywalls. Like MP3s, PDFs democratized access but also opened the door to illegal sharing. In the academic world, Sci-Hub became the research equivalent of Napster. Founded by Alexandra Elbakyan in Kazakhstan, Sci-Hub offers free access to millions of academic papers, circumventing paywalls and raising serious ethical and legal questions.
As downloading music became popular, many — including myself — would organize their downloaded tracks into folders, create backups, and transfer them to USB drives for the car stereo. Tools like Winamp were essential for managing these collections. Similarly, researchers would download PDFs, create local libraries, and meticulously organize them using reference management tools like Mendeley or Zotero. Remarkably, this practice remains prevalent in the research community today.
The parallels between MP3s and PDFs extend beyond their compression, ease of distribution, and accessibility. There are additional commonalities and differences to consider when it comes to music and academic papers:
Commonalities
- Both music tracks and academic papers require significant mental effort and time to produce, unlike many news articles. This is reflected in the copyright and distribution laws governing both types of creations, and it may explain why music and papers have a longer lifespan compared to the ephemeral nature of news.
- Both are categorized according to evolving, multilayered taxonomies — genres in music and research fields in academia.
- Research publishers act as gatekeepers, similar to music labels, ensuring quality by reviewing and selecting works for dissemination.
- Just as creating a music track often involves multiple contributors, writing a paper typically involves many co-authors.
- The authority and brand of authors and journals in academia are similar to the popularity of musicians and labels in music, both serving as indicators of quality that often bias the choices of readers and listeners
- Both formats face limitations in interactivity — MP3s are primarily for audio playback, while PDFs are often printed out and read offline.
Differences
- The scope and mode of engaging with music are quite different and generally less demanding than fully reading an academic paper.
- You can listen to about 12 music tracks in an hour, but reading a paper often takes several hours.
- Music tracks and their creators don’t explicitly reference each other in their work — there are no citations, unlike in academic papers.
As the digital era unfolded, MP3s and PDFs revolutionized the sharing of music and research. However, this newfound freedom posed significant challenges for labels and research publishers who had long controlled content distribution. The surge in free, often illegal, downloads disrupted both industries, sparking a crisis that required innovative solutions. The MP3 revolution led to a significant decline in traditional music sales. The first graph in this great piece on the “The History of MP3” tells 1000 words about the drop of physical, over digital, and ultimately streamed music distribution.
Similarly, in research publishing, the ubiquity of PDFs created turmoil, as traditional publishers with long standing licensing agreements with libraries and institutes, and with millions of papers behind paywalls, had to react to the rise of Open Access journals, pre-print repositories and platforms like SciHub or even ResearchGate, where researchers can directly share their paper PDFs, by passing potential paywalls.
The future depended on finding legal, ethical ways to access content while preserving the value of creative and intellectual work.
Online Music Stores and Open Access
The first real reaction of the music industry to the MP3 liberation, after exhausting the lawsuit avenue, was the launch of iTunes in 2003. Apple’s platform offered a legal alternative to piracy, allowing users to buy individual songs or albums digitally. It was a compromise — a way to embrace digital while maintaining a revenue stream, for both the creators, and the distributors. Notably it was not put together by one of the existing players of the music industry. It was put together by a technology company, that did not own the content, but had the devices and … the listeners. More online music stores like Amazon MP3 (2007), Bandcamp (2008) and Google Play Music (2011) emerged. By 2010, digital music sales had grown to $4.6 billion, representing one third of the global market
In the research world, the industry’s initial solution to the (illegal) access issue was a bit different. Traditional, established publishers license their digital journals to university libraries for access by researchers and faculty members. Access to the general public was restricted behind paywalls.
Open Access (OA) publishing turned the coin upside down, by asking the authors or their institutions to pay for each publication an Article Processing Fee (APC), to make it publicly accessible to everyone. At first glance, this model contradicts what is happening in most creative industries. With the exception of self-publishing of books, or the self-funding of a director’s own movie, authors, musicians, photographers, painters, etc. get paid for their work, rather than having to pay themselves.
But for researchers, career development depends heavily on number of publications and citations. OA access makes publishing both faster and easier, with Open Access journals having on average double the acceptance rate of non-OA journals. Also OA means in general more citations, as publications are globally accessible to everyone. It soon became a win-win arrangement, with more publications and citations for researchers and a direct revenue stream for publishers, that scaled with the number of publications. The average APC is typically reported to be around $2.000 to $3.000 per article, but it can go up to $5.000 for more prestigious journals. In 2010 approximately 270.000 OA papers have been published, accounting for 15% of all publications. By 2020 this number grew to more than a million, a third of all publications. Do the math for a rough estimation of the OA market in this decade.
As I further discuss in my Tribute to Information Overload, OA combined with the COVID pandemic and the need for a fast reaction on behalf of the academic community, had a profound effect on the already growing volume of publications. But with quantity and abundance, comes a price as quality naturally drops. And as quality drops, the main value proposition of publishers, as gatekeepers of quality and prestige is also put in doubt.
There was also another price, as more paper submissions, meant more work for researchers, who are also responsible for reviewing peer submission and editing the journal. It was estimated that in 2020 alone the monetary value of the time researchers spent for reviewing papers, was a few billions. This may also explain the huge profit margins of publishers, which is even larger than traditional internet companies like Google.
When I was interviewing researchers for UXR (User eXperience Research) purposes, there were plenty of times that the interviewee would refer to publishers as the “publishing mafia”. This was the overall sentiment. I guess, if you take all of the above into consideration, PDFs allowed publishers to scale paper production and OA to monetize on it, with the noble help of the greatest minds in the world. The end result, so many PDFs that their value gets inflated and finding relevant quality research becomes harder and harder. More and more work for researchers. As a study published in Nature identified, while the number of publications increases, innovations and breakthroughs have become less disruptive. More and more publications remain unnoticed, including work that could lead to breakthroughs.
So overall, the digitization of research with PDFs combined with OA provided publishers with a first solution to the problem of illegal downloading and a way to grow their business on top of the abundance of papers, but this approach was neither researcher driven nor researcher oriented. It still relies on an archaic reviewing and publishing process and researchers’ hard labour and the millions of papers that it produces are not a sign of innovation and progress, on the contrary. I think that Scott Aaronson’s, almost sarcastic review of The Access Principle by John Willinsky, captures these irrationalities of academic publishing and OA in particular in the most graphic way.
Music Streaming and Research Repositories: Access Over Ownership
I hope that it is becoming clear by now that as we move from analog to digital, from Vinyl to MP3 and from Print Journals to PDFs, we are progressively moving the needle from ownership to access. Collecting and organizing albums or journals, became downloading and storing MP3s or PDFs mainly for accessing them easily and anywhere. As further discussed in my “Tribute to Information Overload”, when multiple copies of a music track or paper can be generated and disseminated for free, these copies become worthless (Kevin Kelly “Better than Free”). Also as the number of music tracks or papers that we have access to grows, the possibility of coming across or revisiting a certain piece diminishes mathematically.
As Kevin Kelly (founding executive editor of my beloved Wired magazine) remarks, when owning (copies of) MP3s or PDFs does not have value, it is what can’t be copied that become scarce and valuable. “When copies are free, you need to sell things which can not be copied”. This lies at the core of the emergence and dominance of music streaming services. Instead of selling individual music tracks or a papers, they sold continuous, personalized, access for a monthly subscription fee.
Rhapsody (2001) was probably one of the first web companies to offer a legal music streaming service for a monthly subscription fee. My beloved Last.fm (2002) followed soon after with additional social discovery features. But it was Pandora (2005) that added “personalization” to the mix. Using the Music Genome Project, Pandora could recommend songs similar to what users liked, paving the way for algorithm-driven music recommendations. Here comes Spotify (2008), which combined access to a huge library of licensed music, with data-driven recommendations, an intuitive interface and a freemium model, to quickly gain popularity and become the leader of music streaming services.
All major web giants followed suit with Google Play Music (2010), Apple Music (2015), YouTube Music (2015) and Amazon Music Unlimited (2016). By the late 2010s, streaming had become the dominant way people consumed music. In 2020, streaming accounted for 83% of the music industry’s revenue in the U.S., according to the Recording Industry Association of America (RIAA). A new era for the music industry has risen, with the old gatekeepers being pushed in the background of music production alone and new technology platforms taking over.
In academia the shift from ownership to access continued with the rise of of research repositories and preprint servers like arXiv and bioRxiv. These platforms allow researchers to make their work accessible to everyone, freely and immediately, often before it’s even peer-reviewed. So while music streaming services provide users with access to a vast catalog of music without owning the tracks, open repositories and preprint servers provide access to large archives of freely accessible research papers and other research material.
ArXiv (1991), founded by Paul Ginsparg at Los Alamos National Laboratory in 1991, allowed researchers to share preprints — papers that had not yet undergone peer review — with their peers freely and rapidly. It’s wide adoption led to the replication of the concept in other domains, such as BioRxiv (2013) for life sciences and MedRxiv (2019) for medical research, which became especially popular during the COVID-19 pandemic, as researchers sought to share findings as quickly as possible.
PubMed Central (2000), launched by the U.S. National Institutes of Health (NIH) was the first in a series of Open Research Repositories, that had a similar purpose to pre-print servers, the free deposit, archiving and dissemination of research work. Throughout the 2000s, universities and research institutions began establishing their own open repositories to archive and disseminate the work of their faculty and students. Unlike pre-print servers, that specialize in the sharing and discussion of research at its early stages, institutional repositories include published peer-reviewed papers, theses, dissertations, and other faculty publications.
There are today at least 78 preprint servers (according to Wikipedia) and around 6000 research repositories (indexed by the Registry of Open Access Repositories). Are they the Spotifies of Research? There are clearly commonalities but also pronounced differences.
Commonalities
- Access over Ownership: Music streaming services offer access to vast music catalogs, while open repositories and preprint servers provide access to scholarly work without requiring traditional subscriptions.
- Democratization of Content: Music streaming makes music accessible to a broader audience, just as open repositories and preprint servers make research freely accessible, especially benefiting those with limited resources.
- Technology Driven: Both music streaming services and research repositories or preprint servers were enabled by advancement in web technologies and were not created by the traditional gatekeepers of the respective industries (labels or academic publishers).
- User Oriented: They both cater to the needs of individual listeners or researchers, helping them to find and easily access music or research content respectively.
Differences
- Purpose: While music streaming is designed primarily for entertainment, open repositories serve as long-term archives for final versions of research, while preprint servers focus on the rapid dissemination of early-stage, non-peer-reviewed research.
- Revenue Models: Open repositories and preprint servers are typically non-profit organizations, funded by institutions and governments, aiming to promote the free dissemination of knowledge. Music streaming platforms on the other hand operate on a subscription-based, for-profit model but they share part of their revenue with the creators.
- Content Curation and Personalized Recommendations: While the success of Spotify and similar services rely heavily on algorithmic personalization and curated lists, to help users discover relevant content, open repositories and preprint servers merely archive, index and categorize scholarly work, so users can find it through search or navigation. There is no editorial control, user curation or personalized recommendations.
- Central vs Fragmented Access: Music lovers can satisfy their listening needs with a single subscription to a music streaming service, enjoying access to a vast catalog in one place. In contrast, accessing research is much more fragmented, requiring navigation through thousands of open repositories, preprint servers, online libraries, and individual publisher or journal websites.”
- User Experience: You only need to visit a few academic sites, repositories or preprint servers and then open Spotify to realize that there is no comparison whatsoever when it comes to their design and functionality.
The Spotify of Research
So if open repositories and preprint servers are not the “Spotifies of research”, as they are clearly lacking in terms of universal appeal and user engagement, is there a “Spotify for Research”? Here are a few candidates:
- Google Scholar is widely used by researchers for finding relevant papers mainly via direct search or email alerts that deliver papers about specific topics or by authors the researcher has defined. This is how far Google Scholar gets when it comes to personalization. Also although Google Scholar is very comprehensive, it doesn’t provide full access to all articles. Many papers are behind paywalls.
- ResearchGate and Academia.edu function as social networks for researchers, enabling users to share their work, follow other scholars, and receive updates on relevant research. However, their catalogs are limited to user-uploaded content, and they often face challenges related to content legality and intellectual property. Their initial appeal largely stemmed from enabling users to access paywalled papers by directly requesting copies from the authors. Their recommendation capabilities are also limited and not the primary drivers of researcher engagement on these platforms.
- Mendeley and Zotero are popular reference management tools that help researchers organize and share papers. While they offer social features, search capabilities, and various text analysis tools, they don’t provide a discovery experience comparable to Spotify’s music recommendations. Their primary focus remains on organization rather than discovery. Mendeley, now owned by Elsevier, has integrated more social features, while Zotero, an open-source alternative, continues to be widely used for its flexibility and community-driven development.
- Semantic Scholar, Consensus, Elicit, Iris.ai, Scholarcy, and ResearchRabbit are AI-powered platforms designed to enhance research discovery through features like semantic search, summarization, analysis, and graph-based navigation. While these tools help users explore research more efficiently, they primarily require proactive engagement, lacking the fully passive, personalized recommendations typical of platforms like Spotify. ResearchRabbit offers email alerts and markets itself as a “Spotify for Papers,” improving recommendations based on user preferences, but this is the extent of their personalized recommendation capabilities. Overall, these platforms excel in facilitating active exploration rather than delivering passive, personalized research content.
- Scitrus, ResearcherApp and RDiscovery are personalized paper aggregators designed to help researchers stay updated on the latest and most relevant publications. Their main premise is to aggregate a wide range of published research and create a personalized feed based on the user’s selected topics of interest, with algorithmic personalization further refining the content based on the researcher’s reading history and past publications. Unlike the platforms above, these apps are specifically discovery-oriented. However, despite their innovative approach, none have yet achieved widespread global adoption, on the contrary.
In 2016, Wiley acquired Atypon as part of Wiley’s broader strategy to adapt to the digital era, which included a significant shift towards Open Access (OA). This shift was marked by the acquisition of Hindawi and its portfolio of OA journals, along with a series of other companies aimed at streamlining the publishing pipeline to increase paper volume and, consequently, revenue.
In this context, Scitrus became redundant, as it wasn’t clearly aligned with the core production processes that drive revenue. Wiley officially shut down Scitrus in April 2022. Unfortunately, it seems that ResearcherApp might be facing similar challenges, as its growth appears to be slow. Its revenue model, which relies on content targeting campaigns based on user interest data, puts it in direct competition with more established social networks that publishers and institutions already use heavily for marketing. In contrast, RDiscovery seems to be performing better under the umbrella of Cactus Communications, a company with deep insights into the needs of researchers. RDiscovery also benefits from offering complementary services for authoring and illustrating papers, and it’s the only platform to introduce a freemium model, with premium services — such as the ability to listen to, translate, and collaborate on papers — available under a monthly subscription tier.
Despite all these so interesting efforts, from OA to preprint servers, researcher networks, dedicated search engines, AI analysis and summarization, specialized Reference Management tools with social functionality and Personalized Aggregators, I can still argue that:
There is no Spotify for research yet
Will there ever be? I am not sure. I am still a bit disappointed by the bitter end of Scitrus and not only and it appears to me that personalized aggregation by itself is not enough for researcher engagement. Researchers want to stay on top, but they also want to stay connected and stay at the forefront of their field. It is a lot about personal branding and leading the way.
That said I am more than convinced that research is more than ripe for disruption. The rise of illegal downloading, the inefficiency of academic publishing to produce innovation, the researcher’s frustration with a system that relies heavily on their free labor and the existence of all those approaches and apps are some of the signs pointing towards something big happening. The turmoil will lead to a new attractor for researchers and research.
A central point for all researchers will arise and it may not be just like Spotify, but I predict it will have some of the characteristics below, that are missing from the current landscape, at least combined.
Beyond PDF
With streaming, the music industry has moved beyond MP3s to higher-quality formats like WAV or FLAC, but this shift has become transparent to the listener — most people, especially younger generations, don’t even know what an MP3 is anymore. Similarly, researchers of the future shouldn’t need to worry about PDFs. PDFs, after all, were designed for print and are not truly digital. They’re neither responsive nor interactive, which limits the user experience. We need to evolve to a point where all research papers offer an impeccable, interactive reading experience across devices. Data, figures, and relevant code should be accessible separately and interactively.
There have been efforts in this direction, such as Manuscripts.io, an online editor dedicated to authoring papers, which was also acquired by Atypon, it was directly integrated with Scitrus, but unfortunately had a similar bitter end. Such innovations cannot be easily adopted, as they conflict with the traditional reviewing and editing processes. The entire system needs an overhaul.
But, we need to move beyond PDFs in another, more radical way: the research paper should no longer be the foundation of scholarly communication. Instead, collaboration and contextualized discussions between researchers should become the primary focus, with papers, data, code, and other resources contributing to these discussions and driving collective innovation.
Evolving Scholarly Communication
Publishers have traditionally acted as the gatekeepers of quality and integrity in research. However, this role is increasingly being questioned, particularly as the surge in Open Access (OA) papers has overall led to a decline in quality, thereby diminishing the prestige of both journals and their publishers. The recent Hindawi and “rat paper” scandals exemplify this troubling trend.
As a result, many researchers are seeking to bypass traditional publishing routes in favor of faster, more open methods of dissemination. The future might see a shift away from publishers as the central arbiters of scholarly communication, with new platforms emerging where the research community itself plays a larger role in ensuring quality and integrity. Peer review could become more transparent and collaborative, with researchers engaging in ongoing discussions and assessments rather than relying solely on static evaluations at the time of publication.
In the music industry, platforms like Spotify demonstrate a model where gatekeepers (like labels) play a reduced role. Anyone can upload their music through various distribution services, as long as the audio file meets technical standards and the metadata and artwork are correctly formatted. It is community-based algorithms that handle curation and recommendations. Could this not be the case for research as well?
Researcher First
It’s striking that some of the greatest minds in the world continue to navigate such a fragmented landscape of content, workflows, and user interfaces. Researchers invest billions of work hours annually into an archaic system that struggles to innovate, and they often pay out of their own pockets (via OA fees) to produce millions of papers — further hindering breakthroughs and adding to their workload.
Instead of being revenue-first, the “Spotify for research” should be designed with a researcher-first approach, offering a User Experience (UX) comparable to leading web apps like Spotify, Netflix, and Airbnb. This platform should support the entire research process, from ideation and literature review to dissemination and collaboration. It should serve as an indispensable partner in a researcher’s daily activities, helping them stay organized, discover relevant work, and connect with peers effortlessly — making their lives easier and their work more impactful.
Community Driven
A successful “Spotify for research” will be community-driven, with researchers actively contributing to shaping the platform and ensuring the quality of its content and its recommendations. This means fostering a sense of ownership among users, where they feel empowered to influence the platform’s evolution and contribute their expertise to the broader community. Community-driven platforms encourage collaboration, transparent peer review, and the open exchange of ideas, all of which are essential for advancing knowledge and innovation.
Subscription Based
Just as music and video streaming services have become household staples, with users willingly paying a monthly subscription for access, a similar model could be applied to research. A subscription-based model would not only be fairer but could also promote quality over the quantity-driven approaches needed to support ad-based or OA publishing models. Consider this: the average Article Processing Charge (APC) for publishing a single OA paper (around $3,000) could fund 25 years of a $10 monthly subscription for a single researcher or cover a year’s subscription for 25 researchers.
The subscription model should be transparent and fair, ensuring that the platform remains accessible to all. I imagine a free guest account complemented with a paid membership tier, for full access to the community, its content, its discussions and all the AI tools built on top of them to turn individual, into collective intelligence. More importantly it should reward researchers implicitly but also explicitly, through revenue sharing, as it happens in all creative industries.
Non-profit
Researchers are not motivated by profit; they are idealistic, passionate for discovery and in love with their vibrant academic communities, their conferences, the trips around the world and the different beers that come with them. To truly serve the academic community and align with researchers’ mentality, the “Spotify for research” should ideally operate on a non-profit basis. This would ensure that the platform’s primary focus remains on advancing knowledge and supporting researchers, rather than generating profit. A non-profit model could also foster greater trust, as the platform would be seen as a neutral entity dedicated to the common good, rather than driven by commercial interests.
Knowledge Economy
The future of research dissemination should be rooted in a knowledge economy where contributions are recognized, shared, and rewarded. This means moving beyond traditional metrics like citation counts and impact factors, and instead valuing a broader range of contributions, including data sharing, code development, and collaborative work.
I envision a credit system similar to those used by platforms like Reddit, StackOverflow, or Kaggle. The algorithms underlying this system should be transparent and shaped by the community to ensure fairness. Such a system could generate an authority score for each researcher based on their contributions, while also governing content sharing to prevent spam and ensure quality. By fostering a knowledge economy, we can incentivize a more open and collaborative research culture that accelerates innovation and benefits society as a whole.
What’s next?
I’ve spent the last eight months in stealth mode, working to crack the “Spotify for Research” puzzle. Here’s what has emerged from countless discussions with researchers and people I respect (thank you all for your invaluable help!):
Akanaba envisions revolutionizing research by blending the personalized discovery of Spotify, the engaging reading experience of Flipboard, and the collaborative power of Slack. By combining AI with personalized content discovery, Akanaba seeks to empower researchers and foster a community centered on collaboration, quality, and shared knowledge.
This is just the beginning. I’m eager to hear your thoughts and explore how we can shape this vision together.
