Comment by dnautics | Hacker News Reader

Comment by dnautics | Hacker News Reader

I think the general reputation among the, say, postdocs (generally intelligent and creative but cynical and past the naivete of the grad student) in synthetic biology whenever Professor Church makes an announcement, is that yes, that was something we had mused about but we know already culled it as a bad idea or an idea not worth pursuing at the moment ("I can't believe he actually had the audacity to propose doing that"), and a generally feeling of pity for the grad students and postdocs who are obligated to work on it. Of course that limits the pity to the seen; think of all the poor souls tirelessly working on the project that either totally failed or were a just a cog in the machine on a giant effort that would probably have taken less time and less effort if only the community had waited for the tools to become marginally more sophisticated.

Of course when you are as politically connected and as well known as Professor Church, you can afford to burn people on these impractical efforts, which of course cements your reputation as being successful at doing the marginally possible (or unuseful but with a wowworthy headline), which attracts grad students and postdocs who either blindly or cynically chase working for someone who will improve their reputation, and also makes it possible to further secure funding for crazy projects that anyone else would be laughed at for pursuing (if they were unsensible enough to do in the first place).

I worked in George's lab as a postdoc and have helped start a couple of companies with him. This is a ridiculous and slanderous characterization of the Church lab. I'd recommend talking with folks in the lab before spreading stories like this.

Right... I don't know if he's even the worst though (personally I think Craig Ventor is way worse the Chruch).

I don't think he's all that bad overall. I think he helps direct effort to interesting problems. Much academic work is something of a moonshot these days, and your job is to also carve out some useful incremental advances.

People who put too much stock in what he says are likely to be more to blame. For example, if you see him on the board of a company or as a scientific advisor I wouldn't think much of it. That's because he's on the board of practically everything.

But personally, and from people I've spoken to, I've found him to be friendly and helpful and basically a good guy. That's about as much as you can ask.

Can you expand on "Much academic work is something of a moonshot these days, and your job is to also carve out some useful incremental advances."?

My own assessment is that there is too much low-risk low-reward work in medicine (and its intersection with biology), and I want to hear/read about how I'm wrong.

You need to make grand claims, state grand objectives to get funding.

The easiest route to publication is then incremental advances. Ideally with a story around them that makes them sound like part of a grand plan.

To drop back to epistemology, how do you think people learn new things without trying things that are expected to fail?

It's perfectly fine to try things that are expected to fail, that is not the argument in the comment.

To give a hyperbolic example of what I'm talking about: Let's say you propose to build a giant hadron collider that rings the moon. This is, in the grand scheme of things, a reasonable proposal, within the realm of possibility, and probably something we should get around to doing. Would it be reasonable to pitch starting the construction project today? Or should we maybe wait a little bit? Should a person that jumps the gun maybe by about 20 years, and completes it, say at mabye 100x the expected cost be called a visionary?

Yes.

Someone who waits until things are easy before trying them is not pushing the world forward.

A false dichotomy. Nobody is saying "wait till things are easy". Moreover there is always plenty of low hanging fruit that goes unseen for some reason or another, and I have seen and celebrate the many people go after those things and be wildly successful and world-changing.

You're acting like progress in science is inevitable just by sheer effort alone. If impractical efforts are self-funded I have no problem with them, but when they are sponsored by the taxpayer coffers I balk a little bit. Henry Markram's Human Brain Project is one effort where the technology clearly isn't there and EU money could be better spent on experiments using less sexy/flashy, but better established methods. This is not progress- this is regression.

Yes! A great contrapositive example is Peter Mitchell who believed so hard in the chemiosmotic effect and couldn't find public funding because nobody else could bring themselves to believe it. He had to find independent funding to start a lab and proved it using decidedly non-sexy/flashy methods. Eventually won a very well-deserved nobel prize.

When George and Robi Mitra were working on polony sequencing [1], it seemed like a ridiculous idea that would never work or scale. I'm glad NIH and DOE weren't as shortsighted as me at the time. The efforts here and by others in the lab over the next decade set the stage for the next-generation sequencing revolution.

[1] https://academic.oup.com/nar/article/27/24/e34/2902337/In-si...

To be fair, J. Craig Venter's use of shotgun sequencing was poo-pooed by the NIH, which provides another example where the maverick had merit. It's kind of a difficult position to be in as a funding agency- you want to fund all of the cool things, but you need to balance that with practicality, which leads to risk aversion. It'd be quite a different world if our federal budget were a bit different and everyone were awash with cash. :)

Actually Craig didn't think it would work either (he's not the sharpest knife in the kitchen), and it was ham smith that convinced him that they should do it. Ham smith is legitimately smart. On one project he wrote a Fortran program to find promoter regions... This was in the 2010s.

Reply on news.ycombinator.com