aug 2021 / greg goebel

* This weblog provides an "online notebook" to provide comments on current events, interesting items I run across, and the occasional musing. It promotes no particular ideology. To keep up with new postings, follow gvgoebel on twitter.

banner of the month



* YET ANOTHER ARPA? In 1958, US President Dwight Eisenhower established the "Advanced Research Projects Agency (ARPA)" -- an office dedicated to "blue sky" military research that the armed services saw as too speculative to fund themselves. The Americans feared falling behind the Soviet Union technologically; Eisenhower, who saw that service chiefs often lacked vision, wanted to stay a step ahead. DARPA -- the word "Defense" was tacked on in 1972 to clarify its military focus -- would prove highly successful, notably to lay the groundwork for the global internet. More recently, DARPA provided funding to develop mRNA vaccines that have proven vital in dealing with the global COVID-19 pandemic of 2020:21.

As discussed in an article from NATURE.com ("The Rise Of 'ARPA-Everything' And What It Means For Science" by Jeff Tollefson, 08 July 2021), the success of DARPA led to the US government establishing clones:

Japan, Germany, and Britain have all launched ARPA-like efforts themselves. The UK effort, the "Advanced Research & Invention Agency (ARIA)", was launched in 2020, with the equivalent of over a billion USD in funding provided to get it running. However, Laura Diaz Anadon -- who heads the Cambridge Centre for Environment, Energy and Natural Resource Governance at the University of Cambridge in the UK -- warns: "The ARPA model has been successful, and we've learned a lot. But ARPA is not a magic bullet that will apply to everything."

Researchers who have studied DARPA and its clones say it works if executed properly, and if the problems are suited to the approach. Of course, it's not trivial to do. It demands competent management that can set up and run grant programs, with a broad freedom to put together research teams, and follow up speculative ideas.

DARPA functions differently from other major US science funding agencies. It has a relatively modest budget, about $3.5 billion USD a year. Its roughly 100 program managers, borrowed for stints of 3 to 5 years from academia or industry, are given plenty of leash in what they fund, and are encouraged to be hands-on managers. In contrast, projects funded by agencies such as the US National Institutes of Health (NIH) typically see little engagement between program managers and the researchers they fund, beyond annual progress reports. Projects funded by the NIH and such also tend to be conservative, with a disinclination to fund risky ventures that could have big payoffs.

William Bonvillian -- a policy researcher at the Massachusetts Institute of Technology in Cambridge who has studied DARPA -- says that the DARPA model doesn't work if program managers aren't given the freedom to fail, which may be a problem with necessarily risk-averse government bureaucracies. That was the problem with HSARPA, which has never amounted to much. Bonvillian says: "If you don't get the culture right on day one, you have got a problem."

Researchers also point out that DARPA works because they have a customer, the Pentagon, for the technologies the agency develops. DARPA program managers are inclined to seek backers in the military who are interested in production systems based on DARPA projects. ARPA-E enhanced the relationship with possible end users by grant recipients to develop plans for commercialization from the outset -- which DARPA has now adopted as a practice.

ARPA-E got the formula right, and has been a notable success. An office inside the US Department of Energy (DOE), ARPA-E has invested $2.8 billion USD in nearly 1,200 projects, which have attracted another $5.4 billion USD in private-sector investments and led to the creation of 92 companies. ARPA-E has also generated large of patents and research papers.

Researchers have had hesitations over Biden's proposals to create ARPA-C and ARPA-H. Some have suggested it makes more sense to simply expand the scope of ARPA-E, and not bother with ARPA-C. Others have worried that the plan to put ARPA-H inside the NIH is misguided, that the NIH is too stodgy to do the job right.

Others worry that ARPA-H's proposed mandate is too broad; there's plenty of private investment in new drugs and medical therapies for prevalent diseases, and so ARPA-H might be better off investigating neglected diseases common in poor tropical countries that don't get much funding. Advances in treatments of neglected diseases could easily have applications elsewhere. In sum, ARPAs are a good idea, but they're just like anything else: they can be made successful, but won't automatically be successful on their own.



* THE WEEK THAT WAS: The significant news of the week was the formal unveiling of the House Select Committee on the Capitol Riot, under the leadership of Bennie Thompson. It kicked off with emotional testimony by Capitol Police, describing the events of 6 January in the Capitol Building.

I knew the Thompson Committee would be a big deal, but it is shaping up to be an even bigger deal than I thought. Initially, the committee will establish the facts through testimony from friendly or cooperative witnesses, and once that is done, then work their way up through Trump enablers -- both inside and outside of Congress. It appears that the Department of Justice will enforce Congressional subpoenas on those outside of Congress; how subpoenas to Members of Congress will work is another question, but it seems likely that House Speaker Nancy Pelosi already knows how that will happen. It will be educational.

I suspect that the agitation of Trump supporters in Congress will become muted once they get hit with subpoenas, in effect asking them to back up their wild charges. It is somewhat more interesting to wonder if Fox News talking heads like Tucker Carlson or Sean Hannity will be called as well. Why not? They're traitors, too.

The last person to be called to testify will be Trump himself -- who will be confronted with a mountain of testimony against him, and will fail disastrously. That's why the Thompson Committee is so necessary: Trump has to be taken down in full public view and be discredited, reduced to nothingness. After that's done, it hardly seems important that he be convicted in court -- though he will be, presumably through plea bargaining, a trial being problematic. In any case, the Thompson Committee hearings will be the biggest show in Congress since the Army-McCarthy hearings of 1954, that brought down Senator Joe McCarthy.

Speaker Pelosi has played her cards well, putting in motion a machine that will take down Trump, and neutralize his supporters. House Minority Leader Kevin McCarthy was carrying on about the Biden Administration's backtracking on COVID-19 measures, in the face of the highly contagious Delta variant now causing a new wave of infections. A reporter asked Pelosi about McCarthy's comments, with the Speaker replying: "He's such a moron!"

That was startling, if not all that inaccurate. It appears that Pelosi has, having neutralized McCarthy, written him off. Trump's influence may be headed for steep decline.

* In the meantime, the emergence of the Delta variant of SARS-CoV-2, said to be as contagious as chickenpox, and the reluctance of many Americans to be vaccinated, has led to a step back in the USA's recovery from the pandemic. The White House is trying everything possible to encourage people to vaccinate, but with limited success. Many people are very hostile to the idea.

... so hostile, in fact, that in Missouri, pharmacists and others administering COVID-19 vaccines report that they are often giving them to people who seem to be attempting to disguise themselves, and plead that no one should be told about the vaccination. It appears that those have come to their senses and decided to get vaccinated are worried about the reactions of family, friends, and colleagues who are opposed to it.

It is very discouraging. I was thinking of taking a road trip from here in Colorado to Seattle in September, but now that's out of the question. I'm going back to masking every time I go shopping again; I don't feel I'm at much risk from the virus myself, but I can be carrying it, and give it to others. Mask mandates are coming back anyway.

Things aren't so bad here in Colorado for the moment, but COVID-19 is clearly on an uptrend. Everything is going bad in Florida, in no small part due to Governor Ron Desantis AKA "DeathSentence", who honestly seems to think that defying pandemic control measures is good for his career. In the meantime, the number of Americans who believe that COVID-19 vaccination should be mandatory is creeping up towards the 2/3rds mark. Blanket mandates are not likely to happen, but smaller mandates are on the increase.

This has its dark humor aspects, of course, as shown on Twitter:

pr94563 / @pr945: For all the fear that the unvaxxed have about being guinea pigs in an "experimental" vaccine, do they realize that they are part of the experiment?

They are the control group.

* Chris Cillizza of CNN commented on a rally Trump performed in Arizona, where Trump said:


The county has, for whatever reason, also refused to produce the network routers. We want the routers, Sonny, Wendy, we got to get those routers, please. The routers. Come on, Kelly, we can get those routers. Those routers. You know what? We're so beyond the routers, there's so many fraudulent votes without the routers. But if you got those routers, what that will show, and they don't want to give up the routers. They don't want to give them. They are fighting like hell. Why are these commissioners fighting not to give the routers?


Huh? What? OK, it's an aspect of the ongoing audit of the vote in Maricopa County, Arizona, with those conducting the dubious audit trying to obtain access to internet routers in the state -- in hopes of determining if Maricopa County voting machines were connected to the internet on election day 2020. This links into an elaborate and completely fabricated tale that the vote was manipulated via Italian communications satellites. The ARIZONA REPUBLIC wrote back in May:


Senate liaison Ken Bennett has said [the routers] are needed to check whether the county's voting machines were connected to the internet during the election. But a county spokesperson said that the auditors already have the information and machines to perform that check, and a previous independent audit commissioned by the county proved they were not.


Allowing the auditors to obtain the routers poses several problems:

So ... it won't happen. Trump is just making noise and muddying the waters as usual. The pandemic is bad enough; Trump has become something like a pandemic on top of it. How much longer do we endure? I suspect things will head for a resolution next year. There is the question of how long Trump will remain in business, those who saw videos of him in Arizona saying that he looked unwell. I will say no more about that.