Communities

Writing
Writing
Codidact Meta
Codidact Meta
The Great Outdoors
The Great Outdoors
Photography & Video
Photography & Video
Scientific Speculation
Scientific Speculation
Cooking
Cooking
Electrical Engineering
Electrical Engineering
Judaism
Judaism
Languages & Linguistics
Languages & Linguistics
Software Development
Software Development
Mathematics
Mathematics
Christianity
Christianity
Code Golf
Code Golf
Music
Music
Physics
Physics
Linux Systems
Linux Systems
Power Users
Power Users
Tabletop RPGs
Tabletop RPGs
Community Proposals
Community Proposals
tag:snake search within a tag
answers:0 unanswered questions
user:xxxx search by author id
score:0.5 posts with 0.5+ score
"snake oil" exact phrase
votes:4 posts with 4+ votes
created:<1w created < 1 week ago
post_type:xxxx type of post
Search help
Notifications
Mark all as read See all your notifications »
Meta

Please gray out the usernames of automatically-imported accounts on the main Users listing

+4
−0

Hi! I'm an SE user and I'm here because I'm curious about what's happening here and because I support the goals of this site.

I have one quibble with what's being presented here, that I hope you understand and can act upon. I also want to couple that with some background on the same issue cropping up on a previous build-an-outside-site experience.

Specifically, I'm concerned about the main Users listing,

enter image description here

which contains several accounts that were auto-imported from Stack Exchange without direct input from their authors (generally because they're part of a bigger thread that got imported because someone requested another post on that thread to be imported). The highest-scoring such account on the listing is currently Chris Sunami.

Now, the profile pages themselves are clearly labelled with the text

This user was automatically created as the author of content sourced from Stack Exchange.

The original profile on Stack Exchange can be found here: https://writers.stackexchange.com/u/10479.

that explains what's going on. However, this is not visible from the main listing, and I would argue that this is essentially in breach of the CC BY-SA 3.0 license, specifically the provision in §4.c that states

You may not implicitly or explicitly assert or imply any connection with, sponsorship or endorsement by the Original Author, Licensor and/or Attribution Parties, as appropriate, of You or Your use of the Work, without the separate, express prior written permission of the Original Author, Licensor and/or Attribution Parties.

Now, I know this is being done in good faith, but I think that if a site like this aims to improve on SE and to get everything right, it should do better.

For some context, a previous build-a-parallel attempt, PhysicsOverflow, got this pretty catastrophically wrong when they started (It's still not great, but I guess it's OK); you can see some of the clashes involved on this Meta PhysicsOverflow thread and this Meta MathOverflow thread. There were a fair number of people whose accounts used their real names and who were not happy to see those real names being used to "bulk up" the users listing on a site they previously knew nothing about. Which, I would argue, amounted to the implicit implication of sponsorship by the author of the new collection, in contravention of the CC clause quoted above. I hope this site can show that there are better ways to handle this, in the same way that it has shown that there's better ways to handle the 3.0 / 4.0 mix than what SE did.

Now, I know that the current software is temporary, and that as such it doesn't necessarily make sense to alter it too much. But please take this into account when the new software comes online.

(Also, I'm not very familiar with where the conversation about building the Codidact software is happening, but I'll be happy to post this concern there, if it hasn't been posted already, if you show me where to do so.)

Many thanks in advance for listening!

History
Why does this post require attention from curators or moderators?
You might want to add some details to your flag.
Why should this post be closed?

0 comment threads

2 answers

You are accessing this answer with a direct link, so it's being shown above all other answers regardless of its score. You can return to the normal view.

+0
−0

So this is kinda tricky.

On one hand: what you said. I don't believe it's a breach of the license, because an attribution necessarily involves citing the author's name (or username, in this case) - §4.c is about false endorsement, not simply "this person wrote this". Something that implied that imported users wrote content specifically for this site would fall afoul of it, but a use of a username doesn't. That said, I can understand the concern about folks not wanting their names used without their permission on a site they know nothing about.

On the other hand: firstly, I'm required to display those names. The footer of every Stack Exchange site says "content licensed under (license) with attribution required". Part of the blog post that links to says:

So let me clarify what we mean by attribution. If you republish this content, we require that you:

  1. Visually indicate that the content is from Stack Overflow or the Stack Exchange network in some way. It doesn’t have to be obnoxious; a discreet text blurb is fine.

  2. Hyperlink directly to the original question on the source site (e.g., http://stackoverflow.com/questions/12345)

  3. Show the author names for every question and answer

Emphasis mine. Point 3 there is unambiguous: I have to use the names.

Secondly, if we assume for now that this display does fall afoul of BY-SA 3.0 §4.c: graying the names out has no effect on whether or not that's a breach. Color can assist user recognition and experience, but it has absolutely no legal standing for this sort of thing. It'd need a different user display entirely with affordances that clearly lead you to find out that it's not a native user. Which brings me neatly on to...

Thirdly: it's technically difficult. This is deliberately third because that shouldn't be an obstacle where it can be avoided, but this is the real world and I have to balance ideals with practicality and the time available. There's no easy way to find out whether a user is "native" or SE-imported: it's doable in small numbers, but going over a large dataset as is necessary to show the user list, it's horribly inefficient - and that's already an inefficient page. Likewise, adding a separate display for native vs. imported users adds an extra cost - a relatively small cost, but nonetheless an extra thing to develop, test, and maintain, and an extra thing to process.

So - it's not the answer you wanted, but short of someone finding a unicorn solution, I'm going to have to leave it as is for now. I hope that gives you some insight for why, at least.

History
Why does this post require attention from curators or moderators?
You might want to add some details to your flag.

0 comment threads

+0
−0

This was originally intended as comment on ArtOfCode's answer, but got too long for that.

Listing a name on the site's Users page IMHO very clearly claims that person is a user of the site. Note that this is not about the name displayed under imported posts (those are marked anyway), but about the user listing.

And while I don't know the code base, I can't imagine it being that difficult. After all, the information is there, or else you couldn't display it on the individual user's page nor support claiming of the posts.

If the information is there, but not very accessible right now, the solution should be simple:

  1. Add a new column “is_imported_user” to the users table, which defaults to false. Set it to NULL for all existing users. Do not yet act on that field (that is, the field exists in the database, but QPixel ignores it). Making it default to false means that newly created users automatically get the right status, without touching the QPixel code.

  2. Run a background process that looks for NULL entries in that column of the user table, checks for each such user whether it is an imported user, and sets the column accordingly. It doesn't matter if this process is slow (it will certainly have low priority so that it doesn't impede normal use). Maybe it will finish in a week, maybe in a month. Since during that time, nothing visible changes, it doesn't matter too much.

  3. After the process finished, you've got a simple boolean field in the user table that you can act on. Therefore now you can grey out the corresponding users with minimal effort and very little runtime cost. Note that from now on, the background process is no longer needed.

History
Why does this post require attention from curators or moderators?
You might want to add some details to your flag.

0 comment threads

Sign up to answer this question »