gyuserbti

I agree with the OP a little. Central repository models have advantages but they've always seemed like short term benefits in exchange for long term costs.

Things like the app store, in for profit scenarios, seem like ways to slip in monopoly control. Brew is an attempt to circumvent it.

I don't want to come across as suggesting they're a bad idea or don't have advantages, just that on balance I've always had a sense there had to be a better way.

Editor and former professor here.

Pay to publish imho always seemed to have a perverse incentive structure to me. Reminds me of the "advertisement" journals you used to see.

The way I see this headed is academic nonprofit orgs self publishing using open source software etc. and recouping costs through membership fees. Journals are already often closely affiliated with orgs, and the members do all the work except copyediting and editorial portals. Academics are also used to paying academic org membership fees for other reasons.

I've already had conversations in this direction with colleagues.

It's either this or eventually things like the preprint services, maybe with sugar on top.

The killer app in this area would be an open source service / server software for running a journal imho. If it were done well it would be hard to beat.

That said, I think traditional publication models are going to be around for awhile, maybe with much more open access or lower pub charges, just because the publishers do provide a service, even if access and price are distorted, and there isn't a uniformly better model at the moment (maybe too big a conversation for typing on my phone).

I agree but there are many varieties that could meet and have met the criteria for public sales, except for current trends in appearance or taste.

It's really remarkable how canalized apple markets have become, when you're exposed to alternatives. The fads become somewhat self perpetuating because they shift context of perception (people judge based on what they know).

It's not so much currently popular apples are bad, just that there's lots more out there.

With apples too you have the interesting case of the cider market, hard and soft. You can kind of get a sense of the effects of fads because cider apples can be really different in appearance but otherwise meet other marketability criteria. You end up with different sets of apples, which speaks to what people like when appearance isn't an issue.

I acknowledge texture is an issue in eating vs drinking, and that cider apples are blended for their charactistics, but many are also varieties that used to be popular eating or cooking apples that drifted in popularity but are still used in cideries because they've stood the test of time and customers like the flavor.

So sad to hear it's gone. There might have been some interest in propagating it, if for no other reason than cross breeding.

Curious about the taste.

Wolf River apples can be very large, but they're not dark, at least not the ones I've seen.

Speaking from personal experience, it's impossible to find anyone who wouldn't be labeled as unbiased in this area.

There are too many turf wars, government sanctioned monopolies, and rent seekers for anything but that to be the case. There's biases in maintaining the status quo too.

I suppose you could have someone with a pure economics or public health background but in my experience they tend to avoid these topics, in part because they have no incentive to fundamentally change healthcare delivery structures.

Lack of real competition among providers is a real underecognized problem in discussions of healthcare in my opinion. Much could be deregulated in a very beneficial way, but discussions almost always focus on payment instead. When deregulation is raised as an issue, it's almost always done in a way that focuses on easing obstacles to large pharm corps, without addressing other forms of deregulation.

In every case that I'm aware of, increasing scope and practice of providers only has net public health benefits. The only losing group is physicians. It's has always been that way, all the way back to dental practice.

This standard sentiment seems to me to be a machiavellian justification for the status quo and for not trying to improve anything. The costs of the problems are the problem, not whether they are eventually resolved extremely ineffiently.

Interesting read that loops full circle to classic AI debates.

I was frustrated a bit by the dismissal of probability theory, though, as if Bayes theory solved it, and by extension, and probability as a whole could be dismissed.

A lot of the issues the author raises are limitations with Bayesian (at least classical Bayesian) theory. The author's critisms dovetail with some areas of probability theory (cf Jaynesian or algorithmic probability literature); I suspect their concerns are one in the same at some level as some of the concerns discussed there.

The problem is uncertainty to various degrees is fundamental to reasoning, so probability must be involved at some level. An integrated approach is needed. I agree that Bayesian theory per se isnt the end of the story, but something involving probability will be part of it (and because Bayesianism is a big part of that, probably that too).

Isn't this basically what syncthing is (same functionality as wormhole, with gui)?

I also thought I've used a gui dat protocol client to transfer files but maybe it was only in the terminal.

I came here to say something similar. You're right to point out baking powder, because it's a mess. Think it's all the same chemically? Nope. Different brands use very different chemicals that can affect lift, and it can vary over time for the same brand.

Just about everything is like that. Whole wheat flour? Different bran and gluten percentages create really different loaves.

Pretending weights will solve everything is nonsense too. If humidity affects mass per volume, it's going to affect chemical identity per mass too.

Finally, everyone has different tastes. You might like a finer crumb to your bread; your spouse might like it more holely.

Recipes are like guidelines. Always. Experience matters because it reflects knowledge of how variations on the recipe matter.

Finally, if there's one thing I've learned, it's that a lot of things in cooking get passed down as critical when they're not, or even worse, are detrimental. This goes for traditional knowledge handed down, as well as half-baked "scientific" approaches to recipe development that are bad science and driven by a desire to establish authority above all.

Plausible deniability must be assumed with morally controversial actions on the part of institutions, especially government security institutions.

One way to deal with this is context. In this case the pattern of Russia with regard to electronic free speech is pretty clear, so it's reasonable to assume these actions are at least partially motivated by something other than emergency preparedness.

As a native English speaker with Dutch colleagues, I've always thought Dutch was remarkably close. Once you kinda figure out pronunciation patterns a lot of it is very similar. Written Dutch is the only language I intuitively perceive the cognate rrelationships with.

The untold story in these things is the role of forced EMR regulations. I've worked in healthcare for some time, as does my spouse and my family.

These record systems used to all be developed in-house systems, fully self supporting, with little outside involvement. Then in the 2000s gov regulations started mandating adoption of EMRs.

Especially here this seems like a no-brainer but in actuality for many hospitals they were unnecessary and introduced with massive cost overruns as hospitals were forced to buy them from a limited pool of vendors that were approved by deadline rather than by intrinsic need.

How does this relate to the ransoms? Because if the EMRs were adopted organically, my guess is it would have happened more gradually, with more diversity of systems, more open source, more testing, and more emphasis on security, backup, and self-reliant reliability.

It's hard to overemphasize the change in records infrastructure in hospitals due to mandated EMRs, and a lot of it has been for the worse. EMRs would have been implemented eventually without the mandates, but at lower cost and greater security probably.

It's just another example of how overregulation in healthcare that sounds good but in practice ends up creating unnecessary costs and causing problems. It also once again doesn't get attention in healthcare price discussions, because it's structural, indirect, and removed from the immediate billing.

I blame these types of ransomware attacks in part on EMR mandates and those who encouraged them. Should bthe federal gov, which encouraged this mess, pay the costs, either of the ransoms, or the cost of not paying?

Are you suggesting dd is about local minima sort of? Like if you extended the risk: parametrization curve out you'd start to see overfitting again?

My initial intuition is there's limitations in test samples that are used, in the sense they only have so much information. At some point overfitting is likely to manifest not in test risk per se, but in random variations over alternate test samples. Eg overfitting would evidence in susceptibility to adversarial regimes not cross validation risk.

I've always been skeptical of cross validation based inference though and admit it's a fascinating phenomenon in the paper.

It just seems, informationally speaking, to be proposing something akin to free energy: that more data is worse and if you just increase your model complexity you can magically infer truth. It seems more likely to be an error in the inferential paradigm.

The vaccination debate makes me sad and pessimistic. I am an avid proponent of vaccination in general but also of people's right to refuse them.

It's remarkable that things like Tuskegee are so quickly forgotten, and upsetting to me how quickly the issues become so oversimplified. I feel alienated in believing education, transparency, and rewards for prosocial behavior are the best way forward.