There is a book called "The Checklist Manifesto: How to Get Things Right" by Atul Gawande. It has been in the top 1000 books on Amazon for almost 2 years now.

In my favorite summary of it, Malcolm Gladwell ( author of 'Blink' and 'Tipping Point' ) says:



Gawande's idea is a very practical one. Checklists are very useful. I don't even do something simple like going to the grocery store, much less something complicated like doing taxes or planning a vacation or building a foundation for a building without one. Or playing a game of go.
The underlying idea of why checklists are necessary is even more important, IMHO. The contrast of ignorance vs ineptitude describes much of the modern world. It occurred to me as I reviewed the things that I do on a daily basis, and realized that, for any task more complex than using a microwave or digging a post hole, there is lots of information out there about how to do it better. Even the seemingly simple things like driving a car have multiple books. The more complex the task, the more there is. For serious tasks, you can bury yourself in data that would take you a lifetime to sort through.

I think that ignorance vs ineptitude also describes the point at which most of us find go to be hard, and maybe also the point at which we stall in our quest to get stronger. In other words, 'stalling' may be the point where we must make the change from learning more to rearranging and cataloging what we know.

I recall that hundreds of years ago, the Japanese houses would hoard go knowledge. Now, one would need competence in at least four languages and 36 hours in a day just to keep up with what is available. So a different way of phrasing it is when you suffer from ignorance, you are a classical go player; when from inepitude, a modern one.

_________________
Help make L19 more organized. Make an index: https://lifein19x19.com/viewtopic.php?f=14&t=5207

I suspect there's an issue concerning the nature of the task. Simple procedures can be done well by any monkey's uncle in possession of a checklist. Other procedures, like surgery, can't be done by anyone holding a checklist, but still have precise tasks that must be performed at particular steps and can be itemized on a checklist. Reaching a bit, since I haven't read the book, I suspect that there's a lot that can't be itemized on the checklist because it requires too much subtle judgment. Nonetheless the checklist saves you from a lot of huge errors. The checklist might be the biggest thing you can do to improve your performacne. Then there are activities where checklists seem extremely limited. I struggle to imagine a useful checklist for writing a professional philosophy or math paper. I've seen checklists on those subjects, but really they tend to be either heuristics, very small tricks like avoiding particular phrases, or very subjective (the point of the checklist is that you know that you've followed it, right?)

I wonder where Go stands.

_________________
Occupy Babel!

Vaugly related...
We often hear of failure rates near the 50% mark for large software projects...

There is a discipline called "Personal Software Process" which tries to fix the problem from the bottom up by teaching the individual engineer to make checklists and then use those checklists during most phases of development. In addition, there is an emphasis on defining metrics and then gathering data to evaluate against the metrics. (of course, being software nerds, we can't help but geek out and heavily apply statistical analysis...)

Clearly, there are some similarities.

I think you more or less hit the nail on the head with the first part of you post, the idea of the checklist is to avoid a gross error with the mundane / routine part of the task. For an academic paper, the checklist would include things like "verify all relevant sections are included," "Ensure all figures are properly labelled," "Ensure all citations are present and properly formatted." A checklist obviously isn't going to prevent you from drawing the wrong conclusion from data, or stop you from performing the wrong experiment, however it will prevent you from doing things like referring to reference #12 and only have 11 sources listed, submitting a paper with a format that would not be acceptable for the publication, or leaving a section titled as "<Placeholder name>" (things that don't require an expert in the field to catch, but can greatly hurt your chances for getting a paper accepted). The point is to catch the things so simple you may have stopped looking for them consciously.

I am also not familiar with the exact book, but I think I have read some articles pertaining to similar studies. If I recall one of the biggest gains they made in relation to surgery was making sure the doctors properly followed their sterile/scrubbing procedures and also convincing them to stop wearing ties. Not that I have personal experience, but I would imagine when you are going into a complicated surgery it's easy to be thinking about something "more important" than the minutiae related to washing your hands. Skipping a small step there might increase infection risk by 0.5%...but if your hospital does 1000 surgeries a year that's 5 people who will be happy you had a checklist.

In the case of surgery, can you improve the performance of good surgeons by implementing checklists? That's the impression I got.

Either way, I think the difference is this: while there are academics many academics absentminded enough to turn in papers with sections missing (one of my professors was stopped by the police because he had previously been unable to find it and had reported it as stolen...), I don't think ensuring that professors had all their sections would substantially improve scholarship. It would save time, which would help, but that's pretty small. In contrast, it seems like the book is claimining surgical outcomes can be substantially improved by the use of checklists.

_________________
Occupy Babel!

Again, I can't speak for the book, but looking up what I think was the same study:




With an estimated 230 million surgeries performed in a year, that could potentially save about 1.5 million lives per year...given the cost of implementations is virtually nil, that's nothing to sneeze at. In fact it would seem that even just conducting the one trial of the study referred to in the article (which focused only on part of the checklist) likely saved 20-30 lives.

IMO one reason why checklists work with surgery is that in most cases it's a low difficulty high cost of failure problem. It's typical to get careless when facing such problems which the checklist aims to prevent.
On the other hand I don't expect checklists to be very useful with problems where most errors come from the inherent difficulty of the problem, and a high error rate is acceptable.

_________________
Sanity is for the weak.

Yes, the question is whether checklists only help bad surgeons or whether they help everyone.

_________________
Occupy Babel!

Two thoughts.

One: a few weeks ago we had a very similar thread on checklists and go, where people seemed to think that if they just executed their checklist faithfully, they would play better go. Maybe. But I think it makes more sense to focus on forcing yourself to do one or two things per game. After a few dozen games, those two things will become second nature and you can start acquiring a new skill.

Two: magnitudes confuse people. 1.5 million deaths is a lot of deaths. But 230 million surgeries is also a lot of surgeries. Let's say each of those surgeries has a team of, say, five people, and going through the checklist takes five extra minutes. (The checklist has twenty-two steps, one of which is for all of the members of the team to introduce themselves. I think five minutes is conservative, but ianas.) And who knows how much we're paying the average member of the surgical team - maybe $100/hr. So it turns out that the checklist will have cost us $10 billion. That's $10k per life saved, which is a great value (assuming, and this is a big assumption, that the results of that study are robust, and that the people who are saved are not disproportionately about to die anyway - serious epidemiology measures qualys, not lives), but Gawande doesn't go around saying "there's this point on the cost frontier where we can prevent deaths for 10k", he goes around saying "when you don't wash your hands, you're killing people!" That is to say, he portrays the gains from checklists as a free lunch, when they're not; the checklist costs time and resources. It's only a bit of time each time you go through it, but you're going to go through it hundreds of times before you see any benefits. This checklist is probably a net benefit, but that doesn't meant that every checklist that lowers mortality is a net benefit. Hospitals probably already force doctors to go through checklists that are a net loss.

I think I'll try using a checklist into my Go game. However, I have my doubts that it will actually help. I don't need to look at a list to remember that I need to take care of a group in danger, or question whether what I'm trying to save is big, or whether the move is sente or gote. You don't need to be a dan player to ask yourself these kinds of questions in a second nature kind of way, certainly without a list.

However, I did use a checklist of sorts (BO list) for SC2 which helped quite a bit.

$10k for life is about as free as lunch gets.

_________________
Occupy Babel!

Back to the topic at hand: are you ignorant or inept at go? I think I have reached the point of nearly 100% ineptitude.

_________________
Help make L19 more organized. Make an index: https://lifein19x19.com/viewtopic.php?f=14&t=5207

Joaz - I think a certain level of ineptitude creates ignorance about how ignorant you are. Lately I've been slightly less inept, and thus become aware of yet more things for me to be ignorant of.

But there has to a more optimistic way to pose this question. I'm reminded of "weakness trends". Maybe, are your go strengths tied to breadth of knowledge, or flawless fidelity to what you do know?

Regarding the issue of checklists, there are two separate types. One is the type posited in other threads, in which the users seeks to remind themselves to consider certain issues that they feel that they need to internalize. This is different from the type suggested for surgeons in this thread. This type of checklist is not designed to prevent errors such as the slip of a scalpel, but rather to remind the surgeon to remove the scissors before sewing up the patient. A surgeon doesn't need to be reminded of the importance of washing their hands. The study suggests however that they sometimes, for whatever reason, neglect to do so.

For a go player, this is analogous to a blunder. It's one thing to lose a game because you got outfoxed and outplayed, and it's another to lose because you blundered away a group that easily could have lived. A blunder takes place not because we don't fully comprehend some abstruse principle, but because we failed to do something we know is necessary. This type of preventative checklist is the one Araban suggests (and sort of rejects).



There is a difference between surgery and go that makes Araban's checklist appear even more useless to Araban than the surgeon's checklist appears to the surgeon. While they both are designed to prevent easily preventable mistakes, a surgeon need be reminded only once to wash their hands, Araban's list demands that the entire list be checked before every move, again and again and again in order for it to be effective.

How does such a checklist relate to the question of ignorance vs. ineptitude? When Joaz says that he is feeling 100% inept, I doubt that that is because he is missing ataris. We feel inept because we have imbibed such a huge amount of knowledge, and nonetheless practically never know for sure what the best move is. With time, we manage to eliminate certain bad moves, but no checklist in the world will eliminate all of them.

Go knowledge is dynamic and ephemeral. Typically, a surgeon knows what type of surgery he is going to perform. He goes in knowing exactly what needs to be done. Not so the go player. When a go player slices open his patient, he might discover a heart problem,or perhaps the innards of a different species altogether. Maybe he finds himself looking at a transistor board or a machine translation of an instruction manual. Under these conditions it is hardly proper to speak of ineptitude. No disrespect to doctors, who also know a lot, but if we ignore the fact that go is "only a game," go is harder.

_________________
Patience, grasshopper.

I don't know, and I don't know how I'd know.

_________________
Occupy Babel!

As ineptly as I play, I'd think it's a large percentage, but when I really consider it I have an awful lot of ignorance too (=

I will say this, I much prefer to lose due to ignorance instead of ineptitude....If I feel I played well, but lost due to ignorance, it means I got outplayed and have an opportunity to learn about a new hole in my game. When I lose because of a mistake that I should know better, it is very frustrating...going back over the game and thinking, "If I were reviewing this game for someone else, I would immediately call this questionable....why on Earth did I play this?!"