Web Sites That Are Independent Are Also Like Software Projects (Sometimes Literally So)
Roll out your own 'stack'
Let us state, upfront in fact, something that ought not be controversial. It's also something that we alluded to in passing several times in the past, as did other bloggers. That stance is worthy of becoming a consensus, inspired by past thinkers and veteran authors.
Like Professor Larry Lessig expressed it in relation to law, words can be code and code can be words. Code can manage law; code can manage words. Literary works can be dealt with like projects with version control. The relationships are there, even if they're not deeply intertwined. Code is logic. Words convey logic. Law gets interpreted by human logic.
Independent, investigative news sites must be censorship-resistant, at least to the extent that is practical - i.e. pragmatic given existing limitations (ISPs have rules and transnational barriers) - and generally feasible. Authors quickly learn their boundaries and "speech landscape" in their country. There are no universal rules, but legal protections vary based on the chain of liabilities and economic obligations (if you pay nothing, you're not a customer but a worthless tenant). That means that self-censorship (typically out of fear) is minimised and publishing whatever no other sites dare - despite it being true - won't result in very negative consequences, such as suspension. Yes, that means social control media is immediately out the door. Don't even think about it. Merely having presence there alters one's speech (adapting to other people's rules, expectations, ideologies; those can change at any time, not just due to ownership changes).
A lot of that (above-mentioned) independence boils down to technical details, which are either acquired from the outside (that can be very expensive) or acquired 'in-house'. Thankfully we are very technical people, so even when we don't know how to do something we can learn fairly quickly. The site itself is technical, so it makes sense for it to be run and authored by technical people. My parents aren't technical, but despite that their interest in this site grew a lot in recent years and it's not limited to them; our audience grew a lot this past year and we're more fearless than ever. Nobody can stop us, even if some people still try (and fail, repeatedly).
For sites to do independent journalism or coverage of "difficult" (likely to draw heavy backlash) topics they must be sure they are in control of their destiny, not just financially but also technically. They must gradually develop tools that best suit (or work for) their routine workflow and have those adapt - internally rather than externally - as the requirements change/evolve. All sites are different and unique; if they're not, then they probably have not much to offer. Sites can also develop tools for research, for writing, for discussions, peer review etc.
We've got all that. It's improving over time. We're proud of what we've accomplished and we look forward to the coming phases. We attract many whistleblowers, who help us cover what no other site is trusted enough to do (because the sites can betray sources or have no track record with sources, rendering them a risky enigma, potential backstabbers).
We're not claiming that sites run by non-geeks will perish, but typically they need to pay someone from the outside to help them repair technical issues or build things for them. If they rely on some third-party bloatware, then they probably aren't quite in control of their destiny. For about a decade we used a long-term support version of WordPress (from a decade earlier), knowing WordPress was becoming horrendous, especially at the back end. So in 2023 we finally eliminated this dependency, months after support had run out. Automattic could no longer bother with patches; it was "old"...
An associate explains that sites should use their own software or at least consider using a fully FOSS stack. "If the latter, then active participation in the software community at least."
In simple terms, one must ensure that components aren't stale, poorly maintained, too complex to fix/improve oneself (if the need arises years/decades later), scarcely documented etc.
Consider this: Automattic's grip was tightening; forks barely existed, not effective one (when I was new to WordPress, more than 20 years ago, there were some active forks, predating Automattic and its financial motives).
Our stack uses many libraries, but no "mega-packages" like WordPress or Ghost. What we have is simple enough to modify and build upon, e.g. using some wrappers and automation. If some package gets abandoned or becomes incompatible, we can likely replace it with a drop-in alternative; things are modular and atomic. We plan to announce (some time soon) a release of a new curation/research tool (R.R.R.R.R.R is accessible via Gemini Protocol, still lacking installation instructions). It's years-old but fast-evolving - a work in progress this month. Maybe by year's end we can also release the search tool for this Web site. It already works, but it's not accessible to the public yet.
This site became a software project out a need, not a mere desire. █