= 27d567bc2d48d9f40e9ccc20ea370b15)
Should effective altruists care about the US gov't shutdown and can we do anything?
For those who haven't heard, NIH and NSF are no longer processing grants, leading to many negative downstream effects.
I've been directing my attention elsewhere lately and don't have anything informative to say about this. However, my uninformed intuition is that people who care about effective altruism (research in general, infrastructure development, X-risk mitigation, life-extension...basically everything, actually) or have transhumanist leanings should be very concerned.
The consequences have already been pretty disastrous. To provide just one, immediate example, the article says that the Center for Disease Control and Prevention has shut down. I think that this is almost certain to directly cause a nontrivial number of deaths. Each additional day that this continues could have huge negative impact down the line, perhaps delaying some key future discoveries by years. This event *might* be a small window of opportunity to prevent a lot of harm very cheaply.
So the question is:
1) Can we do anything to remedy the situation?
2) If so, is it worth doing it? (Opportunity costs, etc)
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
Comments (111)
No to the second part. Certainly not without abandoning the "effective altruist" label. The US government is something that powerful entities already have huge motivation to influence. Your motivation to change it is laughably trivial. Comparative advantage.
Can we please not engage in public hysterics?
Under which definition of disaster have the consequences already been "pretty disastrous"?
I dunno, I think the interrupting of scientific experiments all across the US is pretty disastrous, in terms of long term effects. The positive downstream effects of scientific research should not be underestimated, and a large scale disruption of that seems bad.
As I said, I'm not terribly well informed about this. Is there something I'm not considering about the extent of the interruption?
Here at a major research-focused university, work goes on as if nothing happened for now.
There are however possibly going to be snarls in the grant-application process if this goes on for a while and much of my lab's funding does come from the federal government in one way or another, causing further problems in the event of truly long showdowns. I don't know the grant distribution schedule off the top of my head.
Luckily we have all our new expensive equipment on hand as of a few months ago and ongoing costs are for things like yeast extract and disposable test tubes and tiny bits of custom DNA. And all our pay, of course...
EDIT as of 10-5-13 It would appear that those of us who have grants paid out for multiple years as lump sums from the NIH are doing okay. Those of us who charge things to an external account are having a difficult time. And those of us who get yearly infusions of money and get the infusion in the fall are doing particularly badly. My lab seems to fall in the first two categories, thankfully enough.
So, a lot of people here have been saying that my notion that there is a high probability that this entire thing has had lots of very negative effects (taking into account the exponential returns on research) are hyperbolic and silly.
When I read this, I feel like I was right about the negative impact, even if I was wrong about the possibility of effecting the outcome. (My notion was that perhaps labs which were in keen need of small funds to tide them over could be supplemented, perhaps with pay-back-if-you-can loans)
I would really like to avoid feeling that I was right if I wasn't actually right, so can I ask after your estimate of how much difficulty these people are in and what the rough ratio of effected / not effected is among those who you observe?
It's not like scientists sit around and do nothing when they don't get funds. The might not be able to buy fancy new toys but they can still use their brains. Not having to waste as much time with writting grants is also a plus. They might even start thinking about applications of their knowledge to make money by solving real world problems.
It reminds me of Bruce Sterlings book Distraction.
Sure. Those who can pick up where they left off are not going to even notice that this happening.
But there are a lot of projects that are very time sensitive, which have already had a lot of money and labor invested into them. These which might be significantly delayed or even cancelled, resulting in a loss of invested resources.
Could you give an example of such a project? I doubt that any projects will be canceled because of the shutdown.
From the linked article:
also
(A 3 year wait for something that was going to launch next month probably means wasted resources. There's also a chance that it becomes a sunk cost as better tech. comes by.)
Given the number of programs being effected, I think it's really unlikely that no projects will be shut down anywhere because of this.
I'm more worried about biology experiments-- some of them need constant maintenance. Do you know whether the folks who are feeding the animals and such are still doing so?
I did hear about this. Yes, those people were generally considered essential, so they get to work. Article link that mentions this:
http://www.npr.org/blogs/thetwo-way/2013/10/01/228208246/the-shutdown-s-squeeze-on-science-and-health
I just heard a radio report which said that animals are being taken care of, but the results of experiments aren't being collected.
Sure they can; but how many of them will be willing to?
What do you think they are going to do? Spend their time learning cooking and taking long naps?
Don't think about the first scientist you can think of, think about the marginal scientist.
I think you underrate the value of unsheduled time when it comes to coming up with new creative ideas.
I don't -- for the typical scientist. But there are a few marginal scientists out there who are torn between keeping on doing research and quitting science to take up (say) surfing, and will pick the latter but would pick the former if counterfactually the NSF were still processing grants.
Yes, but how much would they have contributed to science had they not quit science?
Actually, no. A federally-employed scientist can't "still use their brains" during a shutdown; see the Antideficiency Act.
See above. Grants have to be written eventually; not being able to work part of the year just makes the proportion of working time that has to be devoted to writing grants larger.
Huh? How does this stop them from using their brains? Nothing there is going to stop them from continuing to think about their work, mentally desigining new experiments or new hypotheses.
Admittedly no one's ever been charged under the ADA, but there are plenty of examples of people being disciplined for violating it. I've been temporarily laid-off before -- they're not joking about not being allowed to work. At all.
Even granting that our hypothetical scientist is willing to take the risk of being admonished for working during shutdown -- what exactly are they going to do without institutional support? No journal access, no computing resources, no facilities? Navel gazing only gets one so far.
Thinking about your experiments does not (in itself) involve expenditure of government money, so I don't see how they would prosecute you under the ADA for that. Yes, managers have to be very clear to workers not to use resources, just to keep them away from edge cases, but even with that level of overcaution, managers can't actually stop you.
Even if you came back and (for some reason) said, "Hey boss, I totally thought about this experiment from the couch when the shutdown was going on", they still don't have grounds unless you were using up resources. Now, they could fire you just for the defiance (maybe), but if they're that trigger-happy in the first place, then ...
How effective is the thinking that can be done if you don't have access to any of your work? I'm a gov't employee and am affected by the shutdown. All of my work is on my office computer, which I'm not allowed to even turn on during the shutdown. Yes, it's illegal for me to turn on my work computer or access work email during the shutdown.
Sure, I can think all day about how to solve the current bug in my software, but without access to the actual code on my gov't computer not much can be done.
I worked out an algorithm on paper while I was on vacation once. Once I was back, I implemented it quickly.
This doesn't seem to be that severe a grant. People go into science because they like it, not because it pays well.- for many, thoughts fields about one's subject can border on the intrusive. And as long as they come back and don't say explicitly that their new ideas were from when they were on leave, they'll be fine.
So, they can read papers they already have. They can get journal access from friends at universities. They can do computing that doesn't involve as large a scale. They can think about data they got that doesn't seem to make sense. I agree there are limits but those limits seem not that restrictive as long as the shutdown doesn't last for that long.
Context: The United States Federal Government has shut down on 18 occasions since 1976 (Source)
Additional context: only one of those shutdowns has involved a significant fraction of the government suspending its operations for more than 5 days.
Before 1980, "shutdowns" followed different rules so that they did not affect government operations nearly as much. Since 1980, every shutdown but one has been 5 days or less. The Clinton-Gingrich shutdown, which began in late 1995, is the only one to last longer (first 5 days, and then 21 more days after a brief truce).
...
Is that why you thought I was doing public histerics? I am uninformed, but that much I did know. Just because something has been happening doesn't mean it's okay to let it keep happening.
I'm not implying that society will collapse. I'm saying that research has exponential returns, and as a consequence setbacks that seem small right now might actually be pretty bad. Each one of these 18 occasions could have potentially set us back several years.
I posted to get an estimate on how bad this damage is, and how preventable it is. It might be a stupid question to someone who knows what's going on, but I really don't understand the hostility towards the fact that it was asked.
Politics is the mind killer. If you happen to be cluless about a charged political issue you shouldn't be suprised when you encouter hostility when you talk about the issue.
If you want to get an accurate estimate it's a bad idea to start by saying "The consequences have already been pretty disastrous". Rationality 101.
I suppose it could be interpreted that way. It's not like anyone wants research to shut down - everyone agrees that research should continue. There's no political faction that wants to cause trouble. We've talked about much more divisive things in the past.
If the question is clueless, I find it rather strange that no one is actually bothering to explain. I've seen much more uninformed questions talked about in the Discussion sections. I guess I seriously underestimated the whole "politics-mindkilling" thing...
Fair point.
Einstein was more productive when it comes to producing scientific breakthrough when he worked in a patent office in 1905 than when he was having big grants.
It's not at all clear whether writing grants to run fancy experiments and then publishing papers that don't replicate to have a high enough publication rate to get further grants helps the scientific project.
Bad example. Einstein was A) doing physics at a time when the size of budgets needed to make new discoveries was much smaller B) primarily doing theoretical work or work that relied on other peoples data. Many areas of research (e.g. much of particle physics, a lot of condensed matter, most biology) require funding for the resources to simply to do anything at all.
I don't think that true.
If you take something like the highly useful discovery that taking vitamin D at the morning is more effective than at the evening that discovery was made in the last decade by amateurs without budjets.
Fermi estimates aren't easy but that discovery might be worth a year of lifespan. If you look at what the Google people are saying solving cancer is worth three years of lifespan. The people who publish breakthrough results in cancer research have replication rates of under 10 percent. Just as Petrov didn't get a nobel peace price, the people advancing human health don't get biology nobel prices.
Relying on other people's data is much easier know that it was in Einsteins time. Open science doesn't go as far as I would like but being able to transfer data easily via computers makes things so much easier.
The fact that most work in biology relies on experiments suggests that there are not enough people doing good theoretical work in the field it. I don't know much about particle phyiscs but I'm not sure whether we need as much smart people doing particle physics as we have at the moment.
So there are two distinct arguments being made: one is a resource allocation argument (it would be better to spend fewer resources right now on things like particle physics) and the second argument is that in many fields one can still make discoveries with few resources. The first argument may have some validity. The second argument ignores how much work is required in most cases. Yes, one can do things like investigate specific vitamin metabolism issues. But if one is interested in say synthesizing new drugs, or investigating how those drugs would actually impact people that requires large scale experiments.
That's not what is going on here. The issue is that biology is complicated. Life doesn't have easy systems that have easy theoretical underpinnings that can be easily computed. There are literally thousands of distinct chemicals in a cell interacting, and when you introduce a new one, even if you've designed it to interact with a specific receptor, it will often impact others. And even if it does only impact the receptor in question, how it does so will matter. You are dealing with systems created by the blind-idiot god.
You are defending a way of doing biology that plagued by various problems. It's a field where people literally believe that they can perceive more when they blind themselves.
There are huge issues in the theoretically underpinning of that approach because the people in the system are too busy writing research that doesn't replicate for top tier journals that requires expensive equipment instead of thinking more about how to approach the field.
So every field has problems, but that doesn't mean those problems are "huge".
Outside view: An entire field which is generally pretty successful at actually finding what is going on is fundamentally misguided about how they should be approaching the field, or the biologists are doing what they can. Biology is hard. But we are making progress in biology at a rapid rate. For example, the use of genetic markers to figure out how to treat different cancers was first proposed in the early 1990s and is now a highly successful clinical method.
Did you, perchance, mean expected deaths? It seems to me that CDC is important iff there is an outbreak of a deadly epidemic. Then one can discuss what the delta-deaths is actually likely to be; but at any rate it does not appear obvious that losing CDC for a month is likely to increase the number of deaths in a non-epidemic (ie, business as usual) environment. So there's a small chance P(epidemic breaks out while shutdown) times a not-very-well-known but conceivably quite large delta-deaths (CDC handles epidemic versus improvised handling). The latter should likely be, instead, "CDC has a watch officer at first report versus CDC scrambles to get a response together once the epidemic is obvious through other channels".
As for doing something about it: Perhaps you could crowdfund together enough money that CDC could have a skeleton staff manning the phones? Kickstarter, for example?
NB: I would not contribute to such a thing, I'm modelling someone who thought the expected-deaths calculation above came out with rather a large number.
I am pretty sure CDC has people manning the phones...
Crucial agencies within HHS such as the Centers for Disease Control and the National Institutes of Health will still be operating (Source)
I called 800-232-4636 and verified that they are manning the phones. If you want to check for yourself without wasting 5 minutes, skip the phone tree by pressing 1 for English and then 0 for an operator.
I was thinking of more mundane things, for example the yearly flu, which kills quite a few people yearly and would kill more but for careful monitoring of strains and preventative vaccination measures.
The CDC alone isn't what I'm concerned about. It's the small-to-medium inconvenience distributed over a very large number of research facilities, and the larger inconveniences to projects which are time sensitive.
Why is a mere statement of contradiction voted up to five? Something I'm missing here? I could understand if it was Clippy and there was some paperclip related subtext that took a minute to "get" but ...
I suspect two reasons: 1) This summarizes a large amount what other people were thinking. Note that the post Gwern is replying to has had a lot of downvotes, so people who think it is obviously not well thought out favor a response like this. 2) Gwern is a highly respected user who almost never says something without fairly good data to back up his positions, so they are operating under this being a summary of Gwern's more detailed position. (A slightly more cynical version of 2 is simply that Gwern has high status here.)
You didn't provide any reasons, which is odd. Did you just want me to weigh your opinion by itself?
I see you as someone who generally knows stuff, so your opinion alone does have some weight. However, as it stands, I can't even tell whether the lack of an explanation is meant to imply that this is an obvious conclusion and I'm being silly, or whether you're just making a casual remark. Can you say how confident you are in this opinion?
Yes.
If I were to attach a probability, it would be far below 1%; even if the most prominent famous person connected to LW I can think of, billionaire Peter Thiel, were to intervene, I still would not expect as high as a 1% chance of meaningful influence on the outcome.
I think this is the important point people should be talking about; Why are you talking about politics? What possible benefit will come of talking and arguing over that which you can have no effect?
As the son of a company VP said after he observed his elders pontificating on politics:
Case in point: the weather.
In retrospect it was my mistake, although given my initial state of knowledge I don't think I did anything wrong.
1) I thought there was a possibility that something could be done. As of now, I'll take the majority opinion's word that nothing can be done, even though I'm still not sure why this is true. The opinions from people I trust to assign confidences accurately is sufficient evidence for now. I might investigate myself later if I have time, although the fact that most people think nothing can be done indicates that perhaps it's not worth the time to research it.
2) I didn't mentally classify this under the heading "politics", but under "shit, lots of labs are shutting down for a potentially preventable reason, and many smart people (on this forum, too) think that science research is the single most cost-effective good, so maybe this is a very critical time to act. Maybe if I post, other people who know more than me will think about it from an effective altruism perspective and a useful discussion will spark." It seemed a pretty non-partisan issue, since all sides agree that it's bad. That was actually a mistake - I should have realized that anything tangentially related to politics is a political issue.
Despite some of the responses to the contrary, I'm actually still not convinced that this whole shutdown isn't a really, really bad thing...but I guess calculating the harm would be a difficult fermi estimate to pull off.
I'd say the mistake was speaking about disastrous consequences (as a certain fact), when in reality you had little information to back this up.
The proper approach in such situation would be asking: "I heard about X. Do you think it will significantly impact Y?" And then the debate would be about the estimated impacts of X (instead of about your overconfidence).
The political aspect just makes it worse, but I think speaking about disasters in situations where you have little information would be bad even in non-political areas.
We should care, the likely damage from this while mainly diffuse impacts will be large. But no, there's not much one can do about this. The effective altruist community is not large enough nor influential enough to have any substantial impact on this matter.
Would it be worth attempting to fund some entity who is in a position to do something about it?
Outspending an organisation like the US chamber of commerce who lobbies a lot won't be cheap.
There also the problem of lack of information. The information that you read in newspapers about Washington is not objective but put into circulation to achieve political ends. Understanding the details of the problem enough to be able to usefully engage with the situation would require inside information that most of us probably don't have.
Encourage representatives to pass bills funding specific parts of government, e.g., passing a bill funding the NSF specifically, i.e., promote things like this on the internet in various ways could help shame congressmen into at least refunding parts of the government.
Write a letter to your congresspeople, and do it on paper to maximize the impact. Beyond that, you're looking at going up against big money, and the efficiency is gone.
Well, it might be a bit frivolous, but a large number of people are planning on trolling Congress on Friday the 11th.
I would have gone, but DoD employees were given permission to return to work today :)