@richardsever gotcha. Yup, I think that's right. Might also ask of self-hosted solutions what commitments the maintainers have made to stay around. And I think maybe the summary is: all software is debt.

@richardsever would love to discuss more at some point (am back in Brooklyn!), but I hope folks don't over-learn from this and don't apply it too broadly. In my view, the cautionary tale here is about sustainability, and it applies to any vendor and any type of hosting. We were on the path until the rug got pulled out. If the community wants open next-gen they can have it, but they have to invest in it. Has little to do with the tech or type, imo.

@mike hahahahaha yeah no. No one in this industry does it for that. One time a (different) funder said “thanks for falling on the sword of public interest tech for me.” Ouch.

@mike all good. I also don’t want to be too harsh on them. They’re good folks who do good work. We put them in a hard position. I get that. Frustration warranted. And I think we earned a bit more than that.

@mike heh, yeah, I don’t envy that. I’m glad you’re able to make it work, though.

@mike yeah, that’s basically where I am. I of course have a huge amount of empathy - that’s my job as a PM, and the last few years have been quite painful. It’s the insinuation that we didn’t care that gets me. They wrote this after an early demo that didn’t show a lot of what we’d planned/half-built for Legacy users. The bad communication is on us, I’ll own it. But they clearly didn’t trust us despite a lot of history, and couldn’t abide any change, and that’s on them.

@mike It’s also worth noting that the self-hosting vs centralized issue they describe poses a particularly difficult business challenge. We did so many analyses over the years. The market probably just isn’t big enough for a Wordpress.com-like approach to centralized hosting (particularly at those prices) unless it’s subsidized by another revenue stream. In our case, the least-bad option was custom dev. But folks still expect the pricing and service of a billion-dollar company. Anyway.

@mike I appreciate it. And not sure my response was particularly moderate. You’re right, of course, that relying on infrastructure you don’t own is always a risk. But what strikes me about their post in particular is that the bad options they describe are essentially our theory of the market and what were trying to triangulate out of, because we actually are guided by users. Ironically, that was in some ways responsible for the funder’s decision, because they wanted something else.

@mike Do they have valid critiques? Could we have done things better, and more transparently? Of course. But far from the claims made, we put everything we had into trying not to abandon people who relied on us but didn't have resources, when the "smart" move would probably have been leaving folks behind. I'm sorry COPIM didn't like it, but the options the community gave us were go up-market or perish. We tried to make it work for everyone. Sometimes reality doesn't live up to the ideal world.

@mike Sorry, I can't let this one go. The actual moral is don't rely on things you won't pay for. COPIM donated below the suggested amount once — far less than they'll spend on their Wordpress design — for years of usage. I'm grateful for it, but come on. I put almost a decade of my life into building PubPub. You think I wanted this? We were forced to go in this direction because funders abandoned open infrastructure and when we asked users to pay for it, not enough people were willing to.

If you're one of my academic publishing folks: sadly, it's true. Due to a funder's unexpected decision to pull support, we've had to make the incredibly difficult decision to wind down PubPub over the next 18 months and regroup to figure out how to best serve our mission.

I'm sure I'll have more to say, but today I'm feeling both gratified and saddened by the overwhelmingly supportive responses to our announcement.

I urge you to learn from our mistakes: https://www.knowledgefutures.org/updates/2025-06-update/

Wherein I actually read the studies (and invent a philosopher).

https://buttondown.com/gabestein/archive/has-anyone-actually-read-the-generative-ai/

Anyone in the scicomm/policy world working on efforts like described to resist cuts to U.S. biomedical research? Imo, Josh is right that activating patient communities is likely the most effective way to demonstrate the political costs of cuts to legislators. Lmk if you know of any. https://talkingpointsmemo.com/edblog/a-path-forward-to-save-american-bio-medical-research

You can and should admit it: yesterday was deeply, absurdly funny. https://buttondown.com/gabestein/archive/laughing-to-the-bank/

Wherein we oscillate between personal time zone choices, political silliness, and the lengths people went to entertain themselves when history ended. https://buttondown.com/gabestein/archive/greeland-standard-time/

Do we actually make things? https://buttondown.com/gabestein/archive/do-we-actually-make-things/

I did an AI, and it convinced me that building actually valuable AI products is a very specific, very difficult skill. https://buttondown.com/gabestein/archive/i-did-an-ai/

I am "launching" a newsletter that will live upstream of my blog. Here's the first one, about how our metamodern culture truly is upstream of politics, what that means, and how we can use that understanding to move through and beyond the darkness to come. I would love if you checked it out and subscribed so I can get off all the social platforms. https://buttondown.com/gabestein/archive/the-metamodern-miasma/

@jonny @roaldarboel @felwert @lschiff co-sign. “Good” PDFs and integration with these systems are almost the only thing that matters specifically for the folks who have resources, and that’s no accident. They have resources because they’re good at playing that game. We’ve been told that a reason we lost a large paying user was a board member complained that our auto-generated PDFs didn’t look as good as hand-made ones at a cocktail party.

@jonny @roaldarboel Agree. There’s plenty of good open tech. It’s not that. But unless societies have a biz model independent of publishing, they can’t afford to use it because it doesn’t pay them. Simple as that. It’s why we had to shift to building bigger custom platforms. When donors started universally saying we needed to reach self-sustainability (another story) we asked users for help and they just didn’t have the ability, especially to support tech salaries.