Development Abstraction

An astute reader sent me this link and I thought it might be fun to do a spin-off for the biosciences.

Basically, this guy Joel talks about how the ideal infrastructure is one you don't have to even know about because it works so seamlessly. In his case, the ideal situation allows software programmers to just, you know, write programs.

In our case, what would we want?

1. Unlimited funding. 'Nuff said.
2. Unlimited instantaneous ordering, the sort of thing where you think of it and (some invisible and speed-of-light person orders it for you without asking you ten times exactly which thing you want because they already KNOW what you mean because they're ACTUALLY TRAINED TO DO THIS full-time) it's just THERE.
3. Unlimited instantaneous video conferencing from our laptops all around the world, so any time we wanted to ask our collaborators something, they would just be there and answer our questions.
4. Equipment that is always maintained and working and only replaced when the replacement is actually BETTER. If something gets broken WE NEVER KNOW ABOUT IT because it's someone else's job to order the replacement or better yet, they always make sure we have replacement parts on hand and keep up on whether the equipment is not longer being manufactured anymore...
5. Everything is up to code and stays that way and if it's not, WE NEVER KNOW ABOUT IT because it's someone else's job to make sure the biohazard waste and the radioactive waste and the chemical waste are all taken care of.
6. People who actually train students so we can work with them without having to teach them everything they SHOULD HAVE LEARNED IN SCHOOL.

Feel free to add to this list. What are the major sources of our headaches?

Gee, in the course of writing this, here is my thought: if Joel is right and the Roman army had a ratio of servants:soldiers that was 4:1, how would we be doing if for every actual bench researcher in the lab, we had that many people just doing support staff work (maintaining equipment, keeping things up to code, ordering) full-time?

Can you imagine how PRODUCTIVE we would be if we didn't have to do all this stuff ourselves?

And that's where we get back to the unlimited funding dream. Would we really need more money to do this? Or just allocate it differently by changing the career structure? The current academic system is set up this way for two main reasons I can think of (both valid):

1. Doing the maintenance and ordering yourself teaches you how it all works and gives you an appreciation for how much things cost.
2. It's cheaper than hiring people to do it, especially since theoretically there wouldn't be enough for them to do to keep them around full-time.

But what if neither of these things mattered because cost was irrelevant (say, if we could figure out how to make science fund itself instead of being a welfare state)?