You're welcome! It's basically two entirely different philosophies: they went for the nonprofit, donation-based model, while we considered nonprofit status, I took one look at the amount of paperwork it would involve every goddamn fucking year, and my ADHD-having, executive-function-lacking brain immediately noped RIGHT the fuck on out of there, heh. They went for owning the physical hardware they run on because they felt it gave them stronger protections against arbitrary disconnections by their upstream provider, we didn't think the risk of that happening with cloud hosting was significantly more likely enough to justify the extra work it causes and the way it sacrifices your ability to add more resources right this second if you get a spike in usage. They do a ton of (excellent!) other work in addition to offering the AO3, we don't have the energy (or the time -- I'm hella disabled, Mark has two young kids and a very demanding day job) to do anything more than just DW, and lord, I wish I had enough time to do everything I'd love to be able to do with DW, even, but these past few years have not been kind to me. (We finally got my spine fixed! The orthopedic surgeon who finally agree to fix it had some choice words for the orthopedists who blew me off for four years' worth of me steadily losing more and more of how much I could sit up per day until by the time I got in to him, I was at "if I am very very lucky I can get out of bed long enough to eat and pee" level of bedbound. It was a 3 hour surgery with zero complications, I was out of the hospital in 48 hours, and 3 weeks post-surgery I was completely recovered: now I am just having to rehab all the muscle tone I lost in four years of unnecessary pain and immobility, sigh.)
Anyway, I digress! They're two entirely separate approaches, and you can kinda boil it down to: the AO3 went for the "very spread out, loads of people, governance by democratically elected board, build the organization with the extreme long view" approach, we went with the "minimal overhead, less paperwork, retaining control over the decision-making process so we don't ever risk someone who doesn't agree with our vision being the one making decisions, centralize decision-making power but show people they can trust us by being transparent in our decision-making process" approach. Neither is better or worse than the other! Both of them have their advantages and disadvantages. (For example, they tend to get bogged down in committee and debate about things forever because there's unclear structure on who has the authority to be the final word on whatever they're debating; we are really susceptible to "people who are working on a thing don't want to bother the people with the decision-making authority, aka me and Mark, because they know we're busy/overstressed/in a lot of pain right now, so things sit forever until one of us is available".)
I am getting far afield from my original point, heh. Which was: AO3 has much higher cash reserves than we do -- we aim to keep 5-6 months of expenses on hand in cash and another few months' worth in available credit limit; the credit line stays reasonably stable while the cash reserves fluctuate slightly, mostly based on how long it's been since we've had to write a check to the IRS -- but they also have much higher overhead and would have a much higher necessary capital outlay if they decided to move into image/media hosting.
From what I understand, they also have a lot of trouble finding qualified people with the necessary experience in server ops and site availability, because it's the kind of shit job nobody really wants to do on a volunteer basis unless they really, really, really believe in the cause, and because you don't want to give a newcomer access to all your everything (which server ops inherently involves) it takes a lot of time to build trust with new people. Whereas we have the advantage that Mark's entire career has been devoted to that kind of thing (you definitely have heard of the place he works for his dayjob) and Mark has known Robby and Jen since they were all just barely out of their teens, so trust isn't a problem. Again, different approaches, and they all pretty much balance out.
Hosting media is hard, expensive, annoying and fiddly to keep running, and also (in the US) opens you up to a lot of potential legal issues that text alone doesn't carry: they may have decided not to do it (or decided not to do it yet) purely on that basis alone. If I hadn't been working in the field for literally 20 years I would be way more nervous about making the necessary judgement calls myself, honestly.
no subject
Anyway, I digress! They're two entirely separate approaches, and you can kinda boil it down to: the AO3 went for the "very spread out, loads of people, governance by democratically elected board, build the organization with the extreme long view" approach, we went with the "minimal overhead, less paperwork, retaining control over the decision-making process so we don't ever risk someone who doesn't agree with our vision being the one making decisions, centralize decision-making power but show people they can trust us by being transparent in our decision-making process" approach. Neither is better or worse than the other! Both of them have their advantages and disadvantages. (For example, they tend to get bogged down in committee and debate about things forever because there's unclear structure on who has the authority to be the final word on whatever they're debating; we are really susceptible to "people who are working on a thing don't want to bother the people with the decision-making authority, aka me and Mark, because they know we're busy/overstressed/in a lot of pain right now, so things sit forever until one of us is available".)
I am getting far afield from my original point, heh. Which was: AO3 has much higher cash reserves than we do -- we aim to keep 5-6 months of expenses on hand in cash and another few months' worth in available credit limit; the credit line stays reasonably stable while the cash reserves fluctuate slightly, mostly based on how long it's been since we've had to write a check to the IRS -- but they also have much higher overhead and would have a much higher necessary capital outlay if they decided to move into image/media hosting.
From what I understand, they also have a lot of trouble finding qualified people with the necessary experience in server ops and site availability, because it's the kind of shit job nobody really wants to do on a volunteer basis unless they really, really, really believe in the cause, and because you don't want to give a newcomer access to all your everything (which server ops inherently involves) it takes a lot of time to build trust with new people. Whereas we have the advantage that Mark's entire career has been devoted to that kind of thing (you definitely have heard of the place he works for his dayjob) and Mark has known Robby and Jen since they were all just barely out of their teens, so trust isn't a problem. Again, different approaches, and they all pretty much balance out.
Hosting media is hard, expensive, annoying and fiddly to keep running, and also (in the US) opens you up to a lot of potential legal issues that text alone doesn't carry: they may have decided not to do it (or decided not to do it yet) purely on that basis alone. If I hadn't been working in the field for literally 20 years I would be way more nervous about making the necessary judgement calls myself, honestly.