[kliao]

I really like this idea and try to apply it as much as possible. In my experience, for any kind of teamwork, the faster we get to the "I don't knows", the faster we solve problems, while stating half-truths/guesses-as-facts spreads confusion and kills productivity. I've also noticed that the type of person who is afraid to admit knowledge gaps is usually also the one who looks at me funny or expresses disapproval whenever I use the phrase. It's as if they are confused as to why I would ever admit such weakness.

[belovedeagle]

As one of those people who rarely says "I don't know", let me offer a different view: saying "I don't know" is negative and gets you nowhere. Instead of saying "I don't know", say what you do know. Then you're a step closer to understanding than anyone who said "I don't know"; and you're likely to be several steps ahead by the time they get over that hesitation.

Now, guesses should never be presented as absolute facts, but having a guess puts you much closer to the truth than having nothing. If you really, truly have nothing, no relevant knowledge---which should rarely happen if you're well-educated and/or experienced, in which case you're a novice to the subject, where somewhat different rules apply---then say so; otherwise, go with what you know and learn the rest on the way. Will mistakes sometimes happen? Will you sometimes head down dead-ends? Absolutely, but that's part of the learning process and we shouldn't be afraid of mistakes, as long as we're in a process which is tolerant of them. (If you're pushing code to production, for example, then that's a time either to know or not to know, not to guess. But that's the end of the process, not when you're asking questions.)

So if you said to me, "I don't know", I might look at you funny because it's probably not true: you probably do know, at least partially; it just seems like you're not willing to acknowledge that.

[watwut]

I hate that. And after I worked as analyst, I started to hate that even more.

Just say "I do not know" and do not waste your and my time with monologues full of guesses and sort-of-related-but-not-relevant info. It does not help. It just wastes time.

If you say, "I do not know" we can work on it. We can file it as an open question until you know. We can ask somebody else or search for information.

And btw, the answer to "Will mistakes sometimes happen" question is: guesses cause them very often and they slow analysis a lot. You just do not see it, because I can not go to you and show you all your wrong guess mistakes. I have to smile and pretend that I trust you while triple checking everything you say which again, slow down analysis.

Working with people able to say "I do not know" is way easier and faster.

[belovedeagle]

You're treating knowledge as a binary, which it isn't. My way deals more gracefully with thinking you know and being wrong: since you always say what you know, but allow that it may be just a guess, you're always prepared to back off and rethink it. But if you treat knowledge as yes/no, then when you think you know and you say that "yes", you're ready to go full-steam ahead and you're less likely to turn back, since you already "decided" that "I don't know" wasn't the answer.