Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Some years ago I got a librarian (MLIS) degree, and one of the things I remember from the training was that when a patron comes to you with a "reference" question (you know, the kind of thing that doesn't happen anymore, before there was google when they'd ask a librarian what they now ask google)... what they initially tell you they are looking for (or at least what you initially hear/think they are telling you), is usually not actually what they want.

The process is a give-and-take "reference interview" where you collaboratively get to the bottom of what they are actually looking for, not just a simple process where they explain what they need and you find it.

This has been very useful and applicable to any kind of tech support, and this account reminded me of it.

(Also, btw, applicable to any kind of stakeholder expressing requirements/specs too...)



A related article about the XY problem: https://xyproblem.info/


I find it hard to correctly state what I'm looking for when asking for help. Stackoverflow is the main example that comes to mind but it can be any place.

How were you taught to get to the bottom? When saying what it's for, I often feel like I give way too much irrelevant backstory that nobody cares about. Or without it, I risk missing out on a better solution. Also, if one mentions the goal, people will often go "why don't you [redesign the whole thing and] use X" instead of answering the question, but perhaps that's out of the scope of what the asker can do for a better answer and this is more something answerers should avoid doing.


Honestly I remember the general principal more than any particular techniques!

We could try googling the "reference interview". But basically, just inquire about context: Not just what do you want, but what do you want to use it for, what do you want to do with it. And an iterative back-and-forth process of teasing it out. And basic respect for the person, meeting them on their terms not yours, even if you wouldn't want what they want. Be careful of your assumptions. Ask open-ended questions.

When I googled, here's the thing I found that I liked the best: https://opencommons.uconn.edu/cgi/viewcontent.cgi?referer=&h...


In undergrad and even in HS I learned from teachers or advisors that using librarians like this is how its done, well into the google era. Walk in and utter little more than "Ancient Greece" and walk out with four relevant books for your report and a fifth on the way from another branch.


Under different circumstances I noticed something quite similar, but there was a quirk... if they had to make a title of a few words for the request, the title often reflected the true underlying request better than the dissembling longer freeform underneath.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: