“Follow the Money?”: Funding and Digital Sustainability

Humanities scholarship requires funding, whether for books or journals, database access or server space, research assistants or a trip to the archives, or simply a scholar's time to think and create. In the traditional book- and article-based humanistic workflows, colleges, universities, archives, special collections, publishers, and libraries have established labor and funding models that aim to support text-based research, writing, publishing, distribution, and preservation pipelines. But digital humanities (DH) projects don't fit neatly into these pipelines and digital humanists often struggle to work within these preexisting labor and funding models. While the differences between traditional and digital forms of humanities scholarship can cause funding difficulties at any stage of a DH project's lifecycle, these differences are particularly problematic when it comes to sustainability.

There are two common funding narratives often associated with DH projects in the United States. First, that DH projects are always grant-funded and that digital humanists are thus “cash cows” who bring in large amounts of external funding into an institution with projects that could be accomplished only due to those infusions of outside cash. And second, that the open-source, open-access, and collaborative ethos of DH makes it possible for scholars to do DH for “free,” without any institutional investments. Neither narrative is entirely false — and, indeed, the “free” project model is crucial for many contingent and independent scholars who lack institutional support — but not only do most DH projects exist somewhere on the spectrum between “free” and “fully funded,” but also each narrative obscures the financial realities that exist at those poles.

No DH project is ever free; even if no money changes hands, there are always real costs associated with the work.[1] At minimum, a “free” DH project created by a single independent scholar with open-source software on their personal computer relies on the donated labor and resources of both that scholar and the community that created and shared the software. Larger DH projects with multiple collaborators rely on the donated labor and resources of all the project team members. If the project is web-facing, these donated resources may include real money for costs such as domain name registration or server space rental. Collaborators with access to institutional resources also effectively donate those resources to the project team on behalf of their institutions — from the costs of their salary and benefits, to hardware and software costs, to office space and utilities, to interlibrary loans and database access, and more. Some collaborators further involve their students as a source of “free” project labor — exchanging student labor for course credit in a model that should involve significant additional labor from the instructor.[2] As a funding model, then, the “free” project model is actually a self-funded or collaborator-funded project model.

When projects reach the limits of what can be accomplished for “free,” or are conceived of in such a way that they can't even be attempted in such a model, one of the common next steps is for the project leaders to shift towards the other pole and seek grant funding — something that usually, though not always, requires institutional backing.[3] US university and college administrators are particularly pleased when humanists join scientists and social scientists in bringing in outside funds to support both scholarly research and the overall operating costs of the institution. However, the grant application process itself has real costs, primarily in the form of the labor to write and submit the applications, which can take from days to weeks depending on how involved the application process is. Furthermore, most humanities grants are capped at a low dollar amount compared to other disciplines and funding ratios for both major and minor grants are also comparatively low. As of 2021, two of the major US grant opportunities that support digital project creation — the National Endowment for the Humanities (NEH) Digital Humanities Advancement Grants and Digital Projects for the Public — together fund a mere 30 projects a year, with a 14-15% funding ratio and maximum awards of $400,000.[4] On average, then, an institution should expect to submit 6-7 DH grants to the NEH to get one success, translating to tens of thousands of dollars of investment with no guarantee of a return. Grant writing is thus a resource-intensive gamble that doesn't always pay off, especially for scholars playing outside a research-intensive institution who lack access to infrastructures designed to support grant writing and administration.[5]

Each of these models has implications for project sustainability, here defined as keeping the project live on a server and accessible to members of the public through typing a URL into a browser. This is different from preserving a digital project or keeping the “bits” of the project safe somewhere — such as zipping up files and placing them into an institutional repository — even if they are not easily accessible.[6] The sustainability of self-funded projects is reliant on team members continuing to invest their personal resources into — and generate internal team enthusiasm for — the project. If leadership and enthusiasm can be maintained, these projects have the potential to live on indefinitely. But this stream of donations can dry up at any time, at which point the project will begin to fall apart. Web-facing projects will even abruptly disappear at the first failure to renew server space rental or domain name registration.[7] By contrast, grant-funded projects have pre-determined end dates; they can generally be relied upon to survive to that point, but afterwards no more funds will be forthcoming from the granting body. Unless the project team successfully finds a different source of funding to sustain the project — often in the form of new grant applications — it will begin to die slowly the day its grant funds run out.

While both funding models can work reasonably well for the traditional humanities research output of printed books, neither model is ideal to meet the sustainability requirements of digital projects. Both printed books and digital projects require a major initial investment, in time and research materials, to produce. But after the book is printed, it is no longer the creator's responsibility. Instead, hundreds, thousands, or tens of thousands of effectively identical copies are scattered across the globe. And the upkeep costs are low for each individual copy placed on a scholar's or a library's bookshelves: once an appropriate space is set up to hold bookshelves, the marginal cost of each additional book added to those shelves is negligible until there is no more room and either collection items must be deaccessioned or the space must be expanded again.[8] Barring a roof leak, mouse infestation, or careless coffee spills, the book may remain in good shape for decades without much if any active curation; indeed, most books survive best by being left alone.[9] In the carefully controlled environments of a special collections vault, books can last centuries. And as long as at least one physical copy of a book survives, its scholarly contents survive as well. Books thus require an upfront, time-limited financial investment for production, after which point their survival is turned over to a distributed series of owners who pay a set infrastructural cost to sustain hundreds, thousands, or millions of these physical objects.

While digital technologies are all too new to know if any DH projects will last for centuries, right now most are struggling to even last years, much less decades [Davis 2019]. In stark contrast to books, digital projects are generally centralized to one copy, and they decay if they are not actively maintained.[10] From a financial perspective, sustaining a digital project is more akin to sustaining a car than a book.[11] Like a newly-purchased car, DH projects tend to run well for five to seven years with minimal recurring costs for server space or domain name renewal, only to begin to suffer increasingly dramatic breakdowns that require equivalently hefty “mechanic's bills” to fix.[12] And after a decade or maybe two, both cars and DH projects reach a point where they either become too much trouble to be worth fixing or they reach “classic car” status and are deemed worth saving regardless of the costs.[13] Unfortunately for digital humanists, book-based financial models falter and fail when faced with a “fleet” of digital objects whose costs are ongoing, increasing, and directly related to the number of objects that need to be sustained.

The first step for ensuring the financial sustainability of a digital project is thus to understand the technical construction — and associated maintenance costs — of complex digital objects and plan accordingly. Simply being able to accurately predict the ongoing costs of sustaining a website will help ensure that any funding plan is adequate to meet the project's technical needs for its expected lifespan. A custom-coded phone “app” will be more difficult and expensive to sustain than a website based on an open-source content management system (CMS) such as WordPress, which in turn is more difficult and expensive than a website without an underlying database, such as a “flat” website built only with HTML, CSS, and JavaScript. If the predicted funding requirements of sustaining the project end up too high, then building — or planning to reduce a project upon its completion — to the minimum viable complexity will reduce the sustainability costs of the project to a more achievable level. Once a project's needs have been identified, scholars can seek funding from a variety of diverse sources, prioritizing those that best fit the project.

Scholars associated with institutions can start by taking advantage of hard-funded institutional resources, such as specialized email addresses, listservs, or subdomains of their institutional domain, which have low marginal costs to the institution and are likely to be made available to employees who have a need for them. In institutions with more resources, IT services or the library might even supply shared server space for projects running a specific CMS, along with basic systems administration services such as keeping a WordPress installation and its plug-ins up to date. Hard funding and institutional inertia will help sustain projects built with such resources, so long as their creators remain associated with the institution, the projects continue to see active use, and nothing opens up the institution to legal or security threats — e.g. violating GDPR through listserv mis-management or using obsolete and vulnerable CMS plug-ins that open shared server spaces to hackers.[14] And in the most well-resourced institutions, scholars may have access to student research assistants, course releases, or other research funds that can be deployed in support of digital project costs.

In addition to hard-funded resources, institutions can give scholars access to a variety of irregular — but still internal — funding sources. Research-intensive institutions often provide access to start-up funds and internal funding competitions for prototyping, with the implicit or explicit goal of leveraging those prototypes into external funding to further implement those projects. Some may even have internal, endowed prizes that could be sources of unrestricted funds; while they are rarely huge amounts of money, they may still be a significant source of project funds for humanities scholars who regularly apply for external grants as small as $500 or $1000 USD. But the biggest source of irregular institutional income by far is appropriations — money set aside for specific centers, departments, or projects — from friendly administrators who see the value in promoting DH on their campuses. While these funds sometimes “roll over” from year to year, the funds are more likely to be allocated on an annual basis, vanishing at the end of the budget year if not spent. These "use or lose" funds are good targets — especially for scholars with existing high-profile projects that need a brief bump in sustainability support, such as to pay for a CMS upgrade — because they must be spent down every year for the administrative unit in question to receive the same amount, or more, in their budget for the following years.[15] The funds themselves come from a wide variety of sources, including tuition dollars, government appropriations for public institutions, endowment income for wealthier institutions, and grant overheads for research-intensive institutions.

Beyond institutional support there are, of course, external sources of financial resources. Digital humanists regularly participate in the grant economy, seeking funds from government agencies, foundations, not-for-profits, and corporations.[16] Grants might seem like the optimal solution to digital humanists' funding problems, as they entail receiving funding for a specific length of time to generate a specific project, much as other humanists are funded to visit a specific archive or come under contract to write a specific book. A variety of funders offer grants of varying sizes for the creation of digital projects, which makes it possible to align a scholarly agenda with the agenda of those funders in order to create something mutually beneficial. Furthermore, grants come with numerous additional benefits, including a peer review process that helps to establish the digital project's scholarly value prior to the investment of significant time and money; built-in project publicity from the grant funding announcement; and institutional and external prestige for having won a grant.

However, the grant-funding ecology favors the sciences and is not always friendly to humanists. Anyone applying for grant funding in the humanities will quickly recognize there are a limited number of major funders — the US has the NEH and ACLS, and other nations have their own analogous funders — and most of them have limited funding available for specific projects. These grants are not always sufficient to complete, much less sustain, a project, which means scholars must create projects in stages and occasionally creatively re-envision the project. Scope creep — adding unnecessary features to make a project appealing to a new funder — is an ever-present danger that must be guarded against. Opportunities may also be limited by funder agendas. If a project could go two directions and only one is fundable, choosing that direction is not scope creep, but it is a lost opportunity. Similarly, institutional and funder requirements may limit team members' participation — insisting, for example, that team leaders have PhDs or be citizens of specific countries — and prohibit specific project activities, such as event catering. Funders also, quite reasonably, expect grantees to spend funds on the activities they agreed to fund; pivoting the project to new, unanticipated directions thus requires obtaining funder permission.

But the biggest funding issue, from a sustainability perspective, is the fact that grant-funding is term limited. No matter if a grant is for one, three, five, or more years, there is always a planned end to the financial support. There are ways to avoid a hard cut-off of funds, for example by writing a pre-payment for hosting or other sustainability costs into the budget. Scholars should also become familiar with their institution's policies on grant overhead, a surcharge added to all direct grant expenses intended to support the general operating of the institution.[17] A percentage of these funds often "trickles down" from the institution writ large through the scholar's various administrative units — e.g. some goes to the university for general operating expenses, some to the scholar's college, some to the scholar's department, and some to the scholar themselves. For scientists, these overhead funds often help support their lab infrastructure. For digital humanists, these overhead funds can be funneled back into a project to support its long-term sustainability. However, these funds can only extend grant support so far past the grant's end date.

While there are no easy solutions to sustainability funding, the last decade has seen a growing sentiment that not everything needs to be free on the internet and that independent artists, writers, and other digital content creators should be paid for their work.[18] Digital humanists can and should view themselves as digital content creators and pay attention to the business models “indies” use to support themselves. Not all these models will be appropriate to adopt — e.g. for scholars associated with an institution, a grant will bring more prestige and fit better into institutional workflows than a Kickstarter campaign — but an awareness of these models will help scholars creatively seek and acquire funding to create and sustain their projects.

Any list of current indie business models is likely to age poorly, and it's less important to debate the benefits of Patreon vs. Kofi vs. BuyMeACoffee here than to note the existence of business models based on community support. Universities and other large institutions are accustomed to fundraising for large amounts of money, because it costs money to process donations and a $1 USD donation might actually cost more to process than the donation itself. However, there are now and will likely continue to be a variety of platforms that support tip-jar-style microdonations of $1, $5, or even 10¢. There are also platforms that allow audience members to sponsor — or provide recurring donations to — creators, often in exchange for exclusive or early access to content. Freemium models — with a large amount of content available for free and additional content or services available for money — are also popular, and WordPress.com runs on this model; one can create a basic WordPress site on a WordPress.com subdomain for free, but it costs money to remove the advertisements and have a custom URL [Wordpress.com n.d.]. The Mozilla Foundation — creators of Firefox — have similarly been exploring freemium models to reduce digital advertising and support user privacy.[19] And while advertisements can annoy users and raise ethical considerations about advertised content, they are an established model for earning revenue on free content. Podcast advertising, in particular, is a growing market segment that podcasting scholars should be aware of [Interactive Advertising Bureau 2021]. Crucially, these are all models that often are effective for people or projects with robust pre-existing audiences, which is often a key factor in determining which digital projects in a person or institution's portfolio should be prioritized for sustainability funds.[20]

With no one set path for financing their work, digital humanists need to use a variety of funding strategies across their digital projects' lifespans, optimizing strategies to fit a project’s evolving needs. Strategies that work well during the creation phase of a project are often insufficient to sustain the project after its completion, when it needs to be actively maintained for continued public access. Digital humanists need to be flexible about seeking money for their work — both from within their institutions and outside of it — and pay attention to larger funding trends among independent digital content creators. But most importantly, digital humanists need to be clear-sighted about the real costs of their project, aware of their prospective future revenue streams, and proactive in minimizing the technical complexity of projects so that their future revenues are able to cover future costs. Digital humanists need to follow the money — not in the sense of continuously reimagining their scholarly agendas to fit those of funders, but rather in the sense of understanding their project's technical needs, the real costs of that work, and the various avenues available for funding their project at every life stage. Only then can they choose paths that accomplish their intellectual objectives in a fiscally sustainable manner.