Peer Review Anonymity on Trial <Pine.3.89.9601150030.A23284-0100000@mistral.ERE.UMontreal.CA>
Ferland Louis H.
ferlandl at ERE.UMontreal.CA
Tue Jan 16 01:58:53 EST 1996
On Mon, 15 Jan 1996 U27111 at UICVM.CC.UIC.EDU wrote:
[snip] (yes, one line at a time)
> >I don't think bad apples CAN be identified except in the most
> >obvious cases of DEMONSTRATED fraud, etc.
> Personally... I can walk into a cell culture lab and tell you if it
> is a bad one or not just from how it looks. It's not that
> As for other types of labs... it doesn't take long (2-3 days) and
> I know.
I wish there were more god-like people like you around: Our earth would
be so much better!
> If I was an inspector... I could do it in less time... look through
> notebooks (assuming they even keep ones?), look into refrigerators,
God (him again...) forbids!
> And if we can somehow figure out a way in which proficiency testing
> can be done in research labs - even easier. If the lab doesn't
> give back the right results - they loose accreditation.
More seriously, I'm not sure how that would work in research, where by
the very nature of our endeavor, methods and their applications are in
constant flux. Of course, there are some basic techniques that are pretty
well established and widespread, but these are precisely the ones that
are the least likely to cause problems *because* they are well
established and widespread.
One other point: there are good ways of doing things and there are bad
ways of doing things. We must make sure people use good ones, but I think
we shouldn't standardize beyond that. Research is an act of creation. If
you standardize everything (including the innumerable variants of any
given technique THAT WORKS), you kill creativity, and you kill science.
> Proficiency testing of the labs would help tell us who is doing it
> properly and who isn't (or, change the ones who are not doing it
> right to do it more correctly?).
> And this isn't something which one can do for just a one shot
> (test) sample. If you have to do internal testing once a month
> (like a clinical lab) and keep records of such testing... and then
> get an outside unknown every 2-3 months... you ALWAYS have to be
> running your samples correctly to pass all these tests.
And a few paragraphs below you say you don't know that this would be
costly? Gee, last time I didn't realize you were advocating going to THAT
extent. I think this would become more expensive than conducting the
research itself, let alone taking scientists' and staff's time AWAY from
their research. We need FEWER hurdles, not more.
> And as for new PhDs... many of them very little about how to do
> proper bench work. This is not what they are being taught.
It is what I teach my student, and the same with pretty much all of my
colleagues, past and present, where I can form an opinion. You tell some
pretty scary stories below (OOPS! in a snip!). I have seen some of these,
except in less serious, but they were exceptional, in my experience. The
very vast majority of scientists I have been in contact with are, in my
judgement, qualified, dedicated, hard working, and doing good work.
> [snip big time.] Now, that was fun! :-)
> But, if we require certain lab design standards as a norm in the
> future... then newly re-modeled labs would have a blueprint to copy
> off of.
There are already a lot of books that suggest good floorplans for labs,
etc. We don't need imposed strandards.
> Do you see what I am talking about here?
> We need to set up standards for current work.. as well as standards
> for future labs.
Yes, sadly, I think I see what you are talking about here. The entire
scientific community as automatons working in standard labs, with
standard machines which they operate according to standard procedures in
the course of their standard experiments which, in turn, will yield the
standard results expected of standard research. If you dare try something
non standard (i.e. do research), you probably "get phased out"? Early
retirement is starting to sound good (I am 36 years-old... and losing my
enthousiasm for this thread).
> FERLAND, two postings ago
> >Also, I'm NOT SURE AT ALL that the cost of implementing such
> >certification for every individual AND every research group (in
> >the world?)
> These societies/organizations which already requires a membership
> rate would just have to charge more.
So, the cost of this would come out of the scientists' pockets!
> As for overall accreditation... that's would cost. And we would
> have to see how much this costs in a clinical setting to compare it
> to the costs here. But I think if they could do it... so can we.
> >(which Berezin wants to be limited in size to three people (!),
> >see earlier postings, though he later conceded he would accept 5
> >or 6 people under exceptional circumstances if I recall) would be
> >anything but an extremely heavy and costly process.
> ??? I don't know.
Dr. Louis H. Ferland
Centre de Recherche, Hotel-Dieu de Montreal
Dept de Nutrition, Universite de Montreal
Phone: (514) 843-2757 FAX: (514) 843-2719
More information about the Bioforum