View from the Trenches
berezin at MCMAIL.CIS.MCMASTER.CA
Thu Aug 3 09:54:16 EST 1995
Why PEOPLE (and not the "ideas") should be funded:
I can't quite identify who said what in the attached
exchange, so I give a summaraly reply to this
Except of the Platonic world, the existence of "ideas"
per se has no (or very little bearing) on what we are
actually doing. Ideas, like AIDS (sorry for comparison)
exist only in PEOPLE. So, any sensible (funding) system
should primarily fund PEOPLE (with demostrated capacity
to generate ideas), so they can do precisely this - generate
and TEST the ideas.
Not being a medical myself, I would not press too
hard for the above for the bio-med sciences
(perhaps, in bio-med "ideas" more important than
But at least as far as fundamental core sciences
like physics or mathematics are concerned, the notion
"fund ideas, not people" is completely bankrupt by
history, it NEVER worked and NEVER WILL. All history
of peer review demostates that it has no capacity
whatsoever in forecasting the "best" ideas better
than random. Pre-approval of "good ideas" by peer
review is a PROVEN nonsense.
However, peer review IS CAPABLE to assess (better
than random) the ability of specific people generate
those ideas which actually can (and were) IMPLEMENTED.
Contary to futuroloical crystal balling over the
UNDONE work ("proposals evaluation"), the assessment of
TRACK RECORD is a fully feasible task
(and give me a break with people with "good ideas,
but without track record").
Anyone can generate plenty of "good ideas".
In 1726 (!) Johnathan Swift in "Gulliver Travels"
described an "idea generation machine" used by
the Laputa scientists (randomly rotating cubes
put random key words together - and it
works pretty much the same way as many actually
Once again - if we (you) want to do something
seriuos about the present NIH/NSF/NSERC/MRC mess,
the only way to do it is to fund MORE people
on (often) significantly LESSER levels than
at present. Of course, grantsmanship elite
will fiercely resist, but you have no choice
but to fight for this cause.
Ny estimate is that unless we will do the
above (reform the system from competition to
coopreation and mutual support rather than
internal grantsmanship warfare), we (scientists)
are going to be extinct as a social specie in
about 5 to 10 years. The wall handwritings are
already well too many, e.g. recent posts on
PhD overproduction crises.
You can say, I am a doomsday prophet. Fine.
Here is a case to compare:
In 1970 Andrej Amalrik wrote a book "Will Soviet
Union Survive till 1984 ?". Despite that it
was a sort of Samizdat bestseller, most (even many
core Russian dissidents) did not take it too seriously.
At least not in specific terms (..." yes, yes,
sometime next millenium ...").
Now we see that Amalrik was "optimistic" by
just 5 short years (USSR start crumbling in 1989).
Likewise, I will be happy to be equally wrong by a
few years, but highly unlikely it will stretch
beyond the phasing out of the present baby boomers
(i.e. ca 15 years from now).
On 2 Aug 1995, William Tivol wrote:
> >And methods utilized are just as repetitive as clinical labs...
> One of the best ideas from a lab I used to work in was the production
> of "Idiot Sheets"--a set of step-by-step procedures for each prep, analysis, or
> anything else done in that lab. Everyone who came to that lab was required to
> do all their preps, etc. according to the idiot sheets. At the very least, a
> lab run this way will allow anyone to find out exactly what was done (theoreti-
> cally, a tech could not follow the rules, but that's a different problem). At
> best, this system assures consistancy of process, so unusual results can be
> investigated for fewer variables.
> >I suggested you have a person working in the lab who actually knows
> >the proper way of doing one (commercial or non)... certified to do
> The idiot sheets teach anyone the accepted (in this lab) procedure.
> All that remains is a certain amount of co-ordination, the ability to follow
> the rules, and a careful attitude. If these are lacking, that person would
> not be kept on the job.
> >My notions of certification is too force the community to clean out
> >all researchers who could care less on whether they actually do a
> >Northern properly in the first place.
> I agree with the goal.
> > Now Trival [actually, Tivol] had some
> >interesting comments along these lines which I agreed with -
> >>>And what's the difference between junk bonds and investing
> >>>government monies in junk research just because it was a project
> >>>ran by somebody they knew?
> My choice for biggest part of the problem; people are now funded when,
> IMHO, ideas should be.
> >>Junk research has to pass a committee going in and the
> >>reviewers going out, then has to look interesting enough that I
> >>will read it, and conceal its sloppiness or irrelevance well
> >>enough that I will not toss it away in disgust.
> Apparently not too tough. I'd guess that there is a lot of "Well, so-
> and-so's lab *must* know how to do this procedure; they've been funded for
> years." going on in study sections. After all, only one or two members of the
> study section read each proposal fully.
> >>And all it takes to quash who-you-know style "junk research"
> >>is a sufficiently fair and open review process.
> >> Finally, the odds
> >>are that anyone who can make his way through the gauntlet is
> >>likely Who You Know because they are famous for doing some good
> >>work anyway.
> If only famous people are funded, we're in worse trouble than I
> >> Favoritism is a problem to watch, but it is NOT as
> >>dangerous as outright fraud.
> It might be in the long run. Favoritism dictating funding is devastat-
> ing to science; outright fraud will eventually come out. If fraud were common,
> a point which can be debated at length, it would be more serious than favorit-
> ism, but, AIMHO, sloppiness is more serious than fraud, because there is more
> of it, because attitudes accepting sloppiness are found in some labs, whereas
> no lab that I know deliberately accepts fraud, and it can be harder to detect.
> >Charles McCutchen, NIH physicist is also quoted in this article to
> >say, "Once you get a really political atmosphere, then appearance
> >is all important, and if you need a little fraud to maintain that
> >appearance, you do it."
> Exactly why ideas, rather than people, should be funded.
> >And how about this idea of teaching more courses in ethics? It's
> >nothing more than spitting in an attempt to put out a fire.
> Actually, if enough people spit, the fire might be put out.
> >I've said it before and I'll say it again... any system which has
> >to rely upon the honor system and self-policing is doomed to fail.
> The system must act so that it is in the best interest of researchers
> to do competent, original work. If funding were based on that criterion, there
> would be less sloppiness. This is not a quick-and-easy answer, but it may be
> the most stable answer.
> >Why do you insist on twisting this idea into a mold where
> >government sets the standards and tells us *how* to do procedures
> >and protocols?
> It may end up this way; it is all too easy to envision a government
> agency created to provide "accountability"--especially if it is perceived that
> the politically favored will be able to circumvent the standards. After all,
> how likely is it that the president of one of the certifying societies will
> apply for a grant, and how much faith can you have in the certification of that
> >BTW, can you even name me another professional field where people
> >do work who are not certified to do a specific task?
> There are some which offer certification, but do not require it. The
> cert is then worth a higher salary, or can be required by individual prospec-
> tive employers.
> >> 1) Get the entire text of all scientific journals freely
> >> accessible to everyone on-line, at every college, high school,
> >> high-tech business, library, or other Internet access point in
> >> every country of the world.
> Not yet, but coming soon, when the publishers can be appropriately re-
> imbursed. Of course, this begs the question of how those with no access to the
> internet will fare--but that's outside of this thread.
> >> 3) Expand the already extremely successful funding of
> >> biologically oriented computing solutions to allow full
> >> coverage of this part of the field.
> >And how does this help to deter fraud and sloppiness in the field?
> A good idea--at least it standardizes some part of the research. If a
> lab uses a particular program, which is known to work, at least that part of
> their research should be more easily checked. So much can be wrong with the
> statistics of a particular experiment; it would help if the calculation proto-
> col were standard.
> >> No --- just the labs who want ACS funding. If it were
> >>successful, it would catch on. If it were not successful... it
> >>wasn't a good idea to start with.
> It might just work to have (optional) certification be worth some
> points when evaluating a proposal. That might aleviate the need to prove your
> lab can do the work. That also allows more time for the evaluation of the
> actual idea being proposed.
> >You may ask, "but what did she have to lose if she tried it?"
> >It may have immediately killed her or worse yet... caused great
> >suffering. (human dosage was unknown, drug side effects was
> >unknown, etc. etc.)
> Or prevented her from getting a more effective treatment.
> U58563 at uicvm.uic.edu wrote:
> : For instance, in the case of the mycoplasmas, you feel that the burden of
> : proof, as always, is on the researcher.
> The burden of proof of the validity of the research is *always* on the
> : One specific point: Your complaint about review committees "passing"
> : proposals doesn't make sense as I had used the term. When an NIH subsection,
> : for instance, deliberates on a group of proposals, they do not have the luxury
> : of "passing" them all in actual funding --- though most are intelligent enough
> : to DESERVE a "pass" if the money existed.
> The study sections rank the proposals with the full knowledge that
> those below such-and-such a rank will not be funded--this is equivalent to
> not passing them.
> : (e.g. a researcher from the Congo probably thinks that
> : heterosexual transmission of AIDS is *REALLY* interesting, especially by
> : contrast with a gay researcher)
> Researchers from here better find heterosexual transmission of AIDS in-
> teresting--the fastest-growing population of people with AIDS is women. It is
> now clear that the appearance of AIDS in the gay population in the US and Eur-
> ope is an accident of who first brought it to those places.
> : It is easy to say that this should not be, that the reviewer
> : should have a Global Viewpoint For The Good Of Humanity, but difficult to see
> : better solutions.
> Not to see, but to implement.
> Bill Tivol
More information about the Bioforum