matthewmarkus

Oh boy! I'm going to say some unpopular things here, so please remember you asked for feedback.

I immediately click on Funds->Animal Welfare since my company, Pembient, deals in wildlife and I've had a lot of (negative) interactions with the conservation industry. I see that the fund targets "farmed animals" and even then carries a warning:

"Risk-averse donors might choose not to focus on animal welfare because the evidence base for the most effective interventions is not yet as strong as in areas like global health and development."

Excellent! But then I look at the manager's background. It seems he has extensive dealings with The Humane Society of the United States (HSUS). Further, at least one of the grants from the fund goes towards a HSUS project on broiler chicken welfare. OK, that's fine; however, digging deeper I find that at least 5% of the grant covers administrative costs (i.e., $50k). Now, money being fungible, that means money has been freed up for other activities. From my perspective, that would include HSUS's interference in The Black Rhino Genome Project:

https://experiment.com/u/yHldOQ

And that's the problem with donating to large non-profits! They have so much going on that it is hard to track all the externalities.

Let me add that this issue, to me, is the biggest problem with Effective Altruism (EA). It says, "You're 'smart,' go out and become a wealthy hedge fund manager. That's the best use of your time, and then donate the proceeds to things that matter." I would counter and say that if you're truly 'smart,' you should be directly working on problems that matter. Doing something ancillary and then giving to a group of people who might be viewed as less 'smart' because they didn't follow the EA path is internally inconsistent. I think it reveals what EA is: A moral salve to justify forgoing work on hard problems to accumulate wealth for selfish purposes. Not that there is anything wrong with that per se, I just don't like the marketing on top of it.

I said I would be harsh. Apologies, if I've been too harsh.

Thank you for your positive thoughts.

Despite this awful incident in France, there has been some good news on the rhino front. In the five years before our announcement in 2015, rhino poaching was growing an average of 49% or so year-over-year, significantly more than the approximately 1% or less year-over-year growth it has experienced since then. Obviously, correlation doesn't imply causation, but whatever brought about the anomaly has bought the rhinos some time. In fact, South Africa, home to the majority of African rhinos, recently announced its second year of absolute declines in rhino poaching [1]. Hopefully, this trend will continue!

[1] http://www.traffic.org/home/2017/2/27/south-africa-annual-rh...

Honestly, my view is that demand reduction [1] is a monoculture that has taken over conservation and produced very little positive results. In fact, it could be argued that demand reduction has even increased wildlife consumption via boomerang effects [2]! I would like to see another school of thought, harm reduction [3], challenge the demand reduction orthodoxy.

TL;DR: I can think of no campaign that can reduce demand; although, I can think of several that might accidently increase it.

[1] https://en.wikipedia.org/wiki/Demand_reduction

[2] https://en.wikipedia.org/wiki/Boomerang_effect_(psychology)

[3] https://en.wikipedia.org/wiki/Harm_reduction

I'm the CEO of Pembient, one of the companies working on this problem [1]. I would be happy to answer questions on what we're doing. Additionally, it is worth pointing out that a lot of the initial research done on rhino horn usage by conservation NGOs has been debunked by Yufang Gao, Karl Ammann, the Wildlife Justice Commission [2], and others. That is, the use of rhino horn as medicine is not what drives poaching.

[1] https://sosv.com/why-pembient-is-confident-in-synthetic-rhin...

[2] https://wildlifejustice.org/cn/overlooked-rhino-horn-demand-...

An expert with a 50% error rate is a coin. A scientist with a 50% error rate is still a scientist, and perhaps an eminent one at that.

When tens of "expert" climatologists with an error rate of say 5% in their field testify in support of a policy that eliminates thousands of expert miners, geologist, etc. with error rates of say 0.01% in their fields, you're going to have resistance. That's a huge perceived inequality in terms of ramifications and consequences.

If a racing driver loses too many races, she'll eventually have to leave the field. If a casino security person presides over too many unexpected losses, he'll eventually have to exit his area of endeavor. Their expertise is under constant pressure. A scientist isn't exposed to this same pressure, nor should she be, because failure is part of the experimental process. Thus, I think "scientific expert" is a contrast in terms unless it specifically refers to one's ability to conduct the scientific method (i.e., controlled experiments) within a given domain.

For instance, consider James Hansen. There is no doubt that most climatologists consider him an expert in climatology. He ticks off many of the boxes on your list. That's fine, but the climate models he presented to Congress in 1988 don't match observations since that time [1]. That's not to say climate change (née global warming) isn't happening, but that it is happening differently than originally predicted. This nuance is lost when "expertise" is accorded to the modeler. It is lost even more when we start talking about a 97% "expert" consensus that humans are causing climate change.

Perhaps treating "experts" like predictors and somehow attaching a coefficient of determination to each one would ease my concerns.

[1] https://www.skepticalscience.com/Hansen-1988-prediction.htm

You might want to take a look at Hopes & Concerns:

http://www.hopesandconcerns.org/

I'm not super excited about "expert" opinion. An expert, to quote Niels Bohr, "Is a man who has made all the mistakes which can be made, in a narrow field." A jet engine mechanic is an expert if none of the planes she works on experiences engine failure. A scientist is, almost by definition, not an expert. This is especially the case when experiments cannot be carried out. For instance, we cannot instantiate 1000 earths, vary their CO2 content, and then observe what happens. We can only model, and modeling is based on assumptions, or "alternative facts." Almost everything controversial today (e.g., drug policy, economic interventions, climate change, etc.) falls into this quasi-scientific realm. I'm not sure that there is a quick way to reach a consensus on such issues other than by having all of the interested parties slug it out in public over an extended period. I also believe we must be careful about overzealously applying the precautionary principle to the point that policies that are objectively not working become sacred cows (e.g., the War on Drugs).

DB: So, control theory is model-based and concerned with worst case. Machine learning is data based and concerned with average case. Is there a middle ground?

BR: I think there is! And I think there's an exciting opportunity here to understand how to combine robust control and reinforcement learning. Being able to build systems from data alone simplifies the engineering process, and has had several recent promising results. Guaranteeing that these systems won't behave catastrophically will enable us to actually deploy machine learning systems in a variety of applications with major impacts on our lives. It might enable safe autonomous vehicles that can navigate complex terrains. Or could assist us in diagnostics and treatments in health care. There are a lot of exciting possibilities, and that's why I'm excited about how to find a bridge between these two viewpoints.

https://www.oreilly.com/ideas/machine-learning-in-the-wild

Yes, and? Smoking has never been outlawed; although, many have tried. I applaud demand reduction in lieu of prohibition. Once prohibition is in place, look out!

Thanks for the mention! We're mainly focused on horn at present, but I do mention ivory here:

https://news.ycombinator.com/item?id=13288580

Burn tests, as yyhhsj0521 points out, are done. If the object is made of keratin, it will smell like burning hair. I've never burned ivory, but I imagine it would also give off a unique smell (e.g., think drilling in a dentist office). Of course, as biotech advances, DNA tests and the like will be cheap and prevelant, and those will be used to establish authenticity.

There have been some substitutes developed for ivory, but they're readily identifiable. These substitutes include the tagua nut [1] and nitrocellulose plastic [2].

[1] http://animals.oreilly.com/the-other-ivory/

[2] http://io9.gizmodo.com/the-earliest-plastic-billiard-balls-h...

For ivory, you would essentially have to grow a tooth in vitro for less than $1500, retail. Technologically, we're not there yet.

International trade in rhino horn has been banned (via CITES) since 1977. China banned the domestic trade of rhino horn in 1993. A similar ban is in place in Vietnam. You literally cannot ban it any further.

If you want to ban the "keratin" market, make sure you promptly dispose of your woolen sweaters.

Not to besmirch SRT and IRF. They're good organizations that know a lot about rhinos and ecology. However, I have my doubts about their grasp of markets, psychology, history, and statistics among other things.

For instance, from 2008 to 2014 they pushed the message that rhino horn was being used for medicinal purposes. As early as 2012, some intrepid investigators started to make the case that that wasn't what was driving the poaching crisis. There is now a growing consensus that Traditional Chinese Medicine (TCM) was a red herring all along [1, 2].

I believe they're wrong about biofabricated horn too. If you're interested, you can read my rebuttal to their objections here:

https://sosv.com/why-pembient-is-confident-in-synthetic-rhin...

TL;DR: We have Nobel Prize-winning theory (Akerlof, 2001) underlying our proposed intervention.

[1] https://twitter.com/savetherhino/status/798885174520016896

[2] https://wildlifejustice.org/cn/overlooked-rhino-horn-demand-...

Calling public awareness campaigns "extremely effective" is a vast simplification of the matter. Public awareness campaigns are good at pushing a substance towards being banned. Once a substance is prohibited, it is subject to the "forbidden fruit" effect, which I've mentioned below [1]. Additionally, if public awareness campaigns continue under a ban, they're likely to induce a "boomerang" effect. For instance, the National Youth Anti-Drug Media Campaign may have promoted drug use and pro-cannabis attitudes in the young [2]. Crazy, right? Well, California did just legalize marijuana!

My guess is that parent is obliquely referring to that phenomenon.

In my chosen domain of rhinos, it is starting to look as though the public awareness campaigns run from 2008 to 2014 very well could've accidently spread the message that rhino horn is rare, risky, and desirable.

[1] https://news.ycombinator.com/item?id=13287393

[2] http://findings.org.uk/docs/Ashton_M_1.pdf

I feel somewhat qualified to speak on this since my startup, Pembient, biofabricates rhino horn. Basically, rhino horn has been banned in China since 1993. Despite that prohibition, much rhino horn still ends up in China, where it is used primarily for decorative purposes and not medicinal purposes as is widely reported [1].

From a policy perspective, bans often don't work because they induce a "forbidden fruit" effect [2]. Or, in the words of Walter Lippmann [3]:

"We find ourselves revolving in a circle of impotence in which we outlaw intolerantly the satisfaction of certain persistent human desires and then tolerate what we have prohibited. Thus we find ourselves accepting in their lawless forms the very things which in lawful form we repudiate, having in the end to deal not only with all the vices we intended to abolish but with the additional dangers which arise from having turned over their exploitation to the underworld."

It is especially strange seeing wildlife policy push further towards a "war on drugs" stance at the same time that many are claiming that war has been lost [4].

It will be interesting to see how things play out, but I believe policy in this area is fundamentally flawed and will remain so for the foreseeable future. Hopefully, the animals will survive our blundering.

[1] https://wildlifejustice.org/cn/overlooked-rhino-horn-demand-...

[2] http://www.independent.org/publications/tir/article.asp?a=33...

[3] http://www.unz.org/Pub/Forum-1931feb-00065

[4] https://www.weforum.org/agenda/2016/11/war-on-drugs-legaliza...

I might go with something like this:

"We, corporation X, hereby declare as a matter of public record that our counsel of choice in any criminal proceeding pertaining to a NSL is the EFF. This notice constitutes prior informed consent to any advertisement featuring us as a client of the EFF should we chose to retain them."

Well, I guess at some level, I think NSLs violate the spirit of the law. Thus, it is worth examining the issue.

I particularly like the above structure because it separates the hiring of the firm from the disclosure of the NSL to the firm. In fact, you might choose to never actually disclose the NSL to the hired firm! Since the firm doesn't necessarily know the reason they're being engaged, they're not making a disclosure either.

I imagine an organization like the EFF could make it a policy to publicly disclose when they've been hired or terminated by a client.

I find it all very interesting...

I have no idea if it would work. As you state, it isn't very well-defined. I imagine it would eventually depend on a clarification of the phrase "your choice."

That seems like a "disclosure" as defined by the NSL.

How about company Y declares they may hire law firm X if they receive a NSL and law firm X subsequently releases a press release stating that they've been hired by company Y?

Nice! That's very simple and elegant.

I don't necessarily agree with your reasoning; however, it looks like one's right to counsel is restricted if one is not charged with a crime:

https://supreme.justia.com/cases/federal/us/352/330/ (1957)

Further, it seems that Miranda only applies to custodial interrogations.

https://supreme.justia.com/cases/federal/us/384/436/ (1966)

I do think that if one is charged or restrained that the process might work. Of course, that's a pretty big escalation!

Well, if the counsel of your choice was predefined as the one that bid the lowest in an open auction, then why would you lose that right? Could the court further restrict that right down to the specific lawyer they want representing you?

I do agree that the gag order would extend to any lawyer you contacted after the public bidding process concluded.

Well, I know if you're charged, you have a right to be represented by counsel of your choice under the 6th Amendment.

https://en.wikipedia.org/wiki/Assistance_of_Counsel_Clause#C...

I wonder if that extends to witnesses not charged with a crime in some way. Or can you be denied counsel of your choice under questioning?!

I've always wondered if there is a side-channel that could be used instead of a warrant canary. More specifically, my understanding is that witnesses have Miranda Rights. Therefore, couldn't a company promulgate a process whereby they'll publicly solicit bids for legal counsel should they receive a NSL? If they start asking for bids, you can assume they've received a NSL. If not, then no NSL has been received.

Any lawyers in the house?

* Edited slightly for clarity.

Google Alerts has brought this conversation to my attention. If you haven't read our recent thoughts on the issue, I invite you to do so. You can find them here:

https://sosv.com/why-pembient-is-confident-in-synthetic-rhin...

As for IndieBio and the prospects of its class members, I'm sure it is too early to tell. We're flattered, though, to be included in this discussion.

Thanks! Tusks are similar to teeth, so there is a nerve root, etc.

https://www.thedodo.com/community/Elegirl/the-truth-about-tu...

Taking just the tip of a tusk might be the best option for evading a DNA test.

You can get DNA from Ivory.

http://www.pnas.org/content/104/10/4228.full

Bioengineering wildlife products is something my start-up, Pembient, has been working on for a while now.

We're actually working on this at IndieBio!

http://www.pembient.com