Ramblings: December 2022
22 January 2023
Below are collected Twitter ramblings from December 2022, with additional color commentary sprinkled in.
This month’s photo of the month is of Prydden Brook Falls on the Zoar Trail in Lower Paugussett State Forest in Newtown, CT. Although this is the highlight of the trail, the rugged back 4 miles of this trail ensure that you’ll get a good workout in along the way.
Brevity is the essence of business communication, even in science. It takes time to learn your audience(s) and tailor the message. These days, I’m very much of a mind that less is more. Get to the key points, and save the dumpster dive in the data for the backup slides. You can usually have as many backup slides as you want in case someone in the audience feels like diving in. More often that not though, your area expertise can/will/should be acknowledged and some details can be taken for granted. If you’re a project leader, nobody in management can pretend to know the project as well as you do.
December was the month when ChatGPT was ascendant in the public consciousness, and wow. Some of the things I’ve seen it come up with are remarkable. That said, many scientists have noted that its deep knowledge on even rudimentary science topics can be lacking. What it lacks in raw knowledge, though, it makes up for several times over in being a consummate bullshitter. To a non-expert, ChatGPT’s bluster will be indistinguishable from the truth in many cases, and if that isn’t an indictment of the decline in our critical thinking abilities as a species, I don’t know what is. So I’ll leave here the First Law of ChatGPT: if promotional materials you see in the real world are indistinguishable from something ChatGPT could write, it’s probably bullshit.
Marylanders will stop at nothing — nothing — when it comes to adding Old Bay to things. Seafood of all kinds, sure. But also: potato chips, french fries, roasted chicken, popcorn, corn on the cob, pizza, all kinds of eggs, and even beer (look up Dead Rise from Flying Dog Brewing, no joke!). But I have to draw the line at sugar cookies: that’s just gross.
This is a main reason that I object to the phrase “alt-ac”. I understand that the intention here is to agglomerate everything that’s “not academia” in a single phrase to make it easier for folks to find resources. But it also props up the notion, which I strongly disagree with, that academia is the peak career for PhDs and that doing the “alt” thing is really alternative, counterculture, or whatever. When it’s really academia that are the ones who are the minority employment sector. Especially true in chemistry, and I suspect many other disciplines as well. Academia is also the Land of Misfit Toys, which continues to put up with far too many individuals who are bullies, predators, misogynists, and the like — folks who are probably too damaged to be hired anywhere else, where the tolerance for “brilliant jerks” is low to zero.
Good enough carries the day almost every time in medicinal chemistry. Strive to make the simplest molecule that will answer the question. This is an area where medicinal chemists and our computational chemistry colleagues can get disconnected. The latter seem hard-wired to look at x-ray crystal structures and design the perfect fitting molecule with little regard for synthetic tractability. Good medicinal chemistry/computational chemistry partnerships are founded on give-and-take here, with the understanding that a “perfect” design may be morphed into something less-than-perfect in the name of synthetic expediency and getting an answer to a question sooner vs. later.
This post generated some discussion on whether or not I was being too harsh here, and it seems I was. It was (correctly) pointed out that there’s all kinds of reasons why less-than-complete publications go out the door which have nothing to do with doing right by the science: funding, personnel coming to the end of school or a contract, the never-ending need to publish, etc. We as readers need to remain alert to publications overreaching with their conclusions, but if a partial or incomplete study recognizes what it is, that’s fine.
These three posts are all connected to the end-of-year shenanigans at the American Chemical Society’s flagship publication, Chemical & Engineering News. Two highly-respected editors were let go, and it seems the society is hell-bent on turning C&EN into a propaganda mouthpiece. This is not the content that its members want; we want, and have paid for, real science journalism. I can only say I’m taking my purchasing power elsewhere. I haven’t, and won’t, renew my ACS membership in 2023 unless forced to do so for some work-related reason. ACS was already thin on member benefits as it was, and it seems they’re more interested than ever in being a publishing house raking in the dough and paying their executives a small fortune. I see no reason to continue supporting that when C&EN was the last reason I was hanging on.
I’ve gotten tremendous mileage out of Paul Prudhomme’s cookbooks over the years. These dishes are not always health food though: étoufée of any stripe in particular is drenched in enough butter to clog the arteries of a horse. But a spicy dish is sometimes just the thing as we approach the depths of winter in New England.
This is, in hindsight, Manager 101 stuff. Ask your direct reports empowering questions: who, what, when, where, why, how? Avoid closed questions that can be answered with yes/no and focus instead on questions that invite further exploration. Sometimes managers forget to ask their folks how they’re feeling about the trajectory of their career, and it’s important to normalize this conversation early in the relationship. Far better to find out someone’s unhappy and help pivot them onto a better path than assume everything’s fine because they’re not volunteering anything.
This is less true than it used to be, but back at the beginning of my career, I ran across some real doozies who got put into a leadership position solely because of their perceived scientific brilliance. Gradually, but steadily, industry’s tolerance for brilliant jerk-types has gone way down over the last two decades. It’s non-zero, but smart companies realize these days that putting such people into positions of authority can greatly undermine the company’s mission. It’s easier to absorb this lesson in biotech than big pharma. In the former, the organization is often too small to afford even one piece being out of place on the board without risking the failure of the entire enterprise. They’re also thus usually much better at cleaning house in the C-suite when necessary. In the latter, a few leadership failures can persist without sinking the whole ship. Bureaucracy that allows such leaders to hang on is not your friend.
STAR has been around for a long-ass time, and prospective hires should take the time to practice giving answers in this format before interviews. Folks are often caught on the back foot when a trained interviewer will press on maintaining this format. But it’s for a good reason: it allows the interviewer to ferret out the consequences of the interviewee’s actions. “My team had a big problem and there was a lot of infighting, so I just stepped in and fixed it myself” only goes so far. Maybe the result in that case was that the problem was fixed, but you stepped on a lot of toes and pissed off all your colleagues in the process. This is why I talk so much about “how” the work was done mattering as much as “what” was accomplished.
I for one would be perfectly happy to never peer review another article ever again. Really hard to convince anyone that the time put in is worth it these days. This post has good historical perspective on peer review, and also food for thought about what we should do instead.
Remember the med chem null hypothesis: your chemical matter and its mechanism of action are all bullshit. The burden of proof is on you to show otherwise. For anyone truly interested in clinical translation and not just mucking around in the lab with mice, that burden is a very high one indeed. Medicinal chemists in industry will tell you just how much time they spend dealing with this issue: a lot.
I mentioned this in a slightly different context a few posts up regarding brevity. Industry communication in science needs to get to the point. It’s disrespectful of everyone’s time to do otherwise, unless a thorough review was specifically requested. That old joke about academia having infinite time and no money, and industry having infinite money but no time… there’s a note of truth in there.
Possibly the single most difficult thing for new hires to the pharma industry to grasp is the concept of the critical path: the shortest essential set of experiments between you and the clinic. That’s pretty overwhelming when you’re starting from zero on a brand new project in your brand new job. This is why we often break discovery up into milestone phases like exploratory, lead optimization, and the like. Each of those milestones, while arbitrary, is a waypoint along the way to the clinic, and defines the work packages a given company deems necessary to move to the next stage. Smaller chunks are a little more manageable.
I’ve never done axe throwing before - lots of fun! And everyone came back with all their fingers and toes too.