I'm tired of seeing "blinded" questions, e.g. I have a sequence/stack of sequences, should/how can I do X and Y and Z?
Some posters neither explain why/what they really want from our advice nor give metadata relevant to the question, such as how many sequences, by what methods and from what organism. In such cases its difficult to tell if we are dealing with an honest neophyte who's supervisor has left them in the dark or an experienced commercial operator who wants to pump us for expertise while leaving us in the dark (but one can usually divine the difference)
I don't know what sticks or carrots we can offer to ameliorate this situation, but IMCO more transparency in this forum would really boost utility all-round (this includes declaring affiliation details in profiles).
It was discussed before
guiding for posting better questions
Lack of explanation can have multiple reasons, such as language barrier, total lack of understanding of the subject, careless,... I agree, some questions could use more information. But it's not possible to enforce that. I don't understand why total transparency including affiliations is going to solve this and everyone should be free to share what they want.
I've never had that impression, do you have any evidence of this?
Lastly
It's fine to have an opinion and make suggestions, but no need to tell us what we really need according to you.
Yeah theres a big difference between "this is who i am and..." and "this is my data and...". I don't think any of us care about the former as WDC says. It would be good to know what people are trying to do, but often that just requires someone to ask :)
If people used something like log.bio, we could see their exact process. If they share their data with something like SeQC, we could see their data.
log.bio however was a failed project (in its current iteration) because setting up a private logging database was/is a bit of a chore, and there was no mechanism to do some things privately, and something publicly. That has to change before it will see usage. There were some other backwards-compatability-breaking improvements that needed to be made, so the whole thing was shelved until after the PhD when i'm allowed to work on it.
SeQC had the opposite problem. The data for SeQC is stored in local SQLite/Postgres SQL databases, and sharing data from inside a private institute to the outside is not particularly easy. There needs to be at least 1 public-facing server to do the firewall tunneling. This costs money, which is also something i wont have until after the PhD.
And those are just the two tools I made - there are plenty of others that do solve the technical limitations you describe, it's just that they're not used, most likely because they do not adequately address people's privacy requirements. I've heard every excuse under the sun for why someone cannot share their data with me, but do need my help working on their data. It's generally not malicious as you hint at - it's usually just fear. Fear of being scooped. Fear of being berated by someone higher up. Fear of external scientists discovering something in their data before they do. Fear of looking stupid. Lots of fears. Most of which are unfounded, but you have to lead by example on these things.
Yes, the idea of surfacing the data relevant to the question is a good one. I would have suggested Figshare but you can't delete afterwards! DropBox public? Note since this would be time-stamped it could diminish the fear factor
Might work, but I'm not going to download files from random dropbox links shared by strangers :-P
For whatever it's worth, this general issue of people being bad at asking for help isn't unique to this site, it happens everywhere (including when our post-docs pop by with questions).