I'd joined Ello early[1] on and watched its journey with interest, frustration, and ultimately sadness. That included an on-site visit in October 2015, which was genuinely enjoyable. I'd largely written off the site around 2018 when it was sold and information on either the status of the social benefit corporation (SBC) or new ownership was exceedingly scarce. Andy Baio's account corresponds strongly with my own.<p>The community was small (I'd estimated total accounts at perhaps 10--15 million, and actives at roughly 1% of that, generally confirmed during my visit), but vibrant in its own way, and as with numerous other social networks, I miss numerous connections I'd made there, though a small handful have resumed at the Fediverse. There really were some notable gems among the group I connected with, including Paul Mason (journalist, briefly), poet Trenton Lee Tiemeyer, authors Ksenia Anske and Bruce Sterling, SVG guru David Dailey, and others.<p>Most disquieting was how Ello died with neither notice to its members <i>or</i> commentary elsewhere. TFA here is among the very few accounts I've seen of its demise. (I've written a few times about it at the Fediverse myself.)<p>The other concern, going far beyond Ello, is how little protection or guarantee either its manifesto or SBC status ended up providing. Going forward, I'm going to give similar attestations vanishingly slight weight.<p>Having participated in online communities since the late 1980s (Usenet, mailing lists, Slashdot, Google+, Diaspora*, Reddit, Ello, the quite short-lived Imzy, amongst others), I'm reminded of the song "dumb ways to die": there are many ways for a social network to die, most of them depressingly uncreative.<p>The flip side is that it's also difficult for a social network to <i>survive</i>, let alone thrive, and survival. Commercial pressures from every thing I've seen worsen this, though even noncommercial / non-ad-supported sites may see failure modes.<p>Of the best communities I've encountered:<p>- Early Usenet, when access was strongly predicated on academic affiliation.<p>- 1990s-era mailing lists, most especially those focused around Unix / Linux and Free Software topics. (Broader-appeal mailing lists ... tended to function poorly.)<p>- Google+, in patches. The fact that the social network wasn't ad-dependent, and had strong representation from the tech community were strengths, and some of the best engaging online conversations I've had were from there. Diaspora*, though vastly smaller, saw a substantial portion of my G+ community engage there and had similar dynamics which also made for some good discussions, though not quite as often.<p>- The Fediverse. Not quite as rewarding as G+, but I strongly suspect it will prove far more enduring.<p>- Hacker News. As with G+, not directly intended as a revenue-generating platform. Not as good as it could be, but far better many other options, and <i>astonishingly</i> consistent over a very long lifetime (approaching 20 years), which would beat out Usenet by quite a margin.[2] I'd chalk up HN's durability to a few factors: unsexy tech and appearance, stellar moderation, solid design principles, and a viable founding cohort and active community. There are of course criticisms of the site and I've made a few myself, but relative to much the rest of the Net, still running strong.<p>________________________________<p>Notes:<p>1. The Internet Archive has some captures: <<a href="https://web.archive.org/web/*/https://ello.co/dredmorbius" rel="nofollow">https://web.archive.org/web/*/https://ello.co/dredmorbius</a>><p>2. Usenet's birth date was ~1979. By 1999 it was well past its sell-by date, though not <i>completely</i> dead.