@Natty report askubuntu.com/a/1028696/22949

[ Natty | Sentinel ] Link to Post Low Length; No Code Block; Low Rep; Contains Whitelisted Word - You can; 1.0;

thanks for that! hmm... I should save that comment somewhere. I could make it possible for myself to find it in chat later, by seeding some searchable word into the transcript... hmm "seeding" will do

###Here's the comment:

Manually editing `sources.list` and doing `update`, `upgrade`, and `full-upgrade` sometimes manages to upgrade an Ubuntu system to a later release, but it is not a supported, reliable, or safe way to do so. It's also *less* likely to work from an unsupported release, since the differences are greater. It resembles the recommended upgrade procedure in Debian but most Debian releases specify additional actions in their release notes. For Ubuntu, it does not even resemble any recommended procedure. [It is also completely unnecessary, even from an EOL release.](https://a…

That reminds me. There was a question about this that I had answered that was deleted. I was considering trying to use the material for something else, but I don't think I ever did.

@EliahKagan I don't think just saying that the repos are gone so upgrade is an answer. You need repositories to upgrade, so it also makes little sense. I flagged that NAA.

So, the same author of the post about editing sources.list and using apt to effect a release upgrade has apparently recommended it in other answers, too. :(

thanks for the sauce! :)

I'll comment on that other one. If there turn out to be a lot of them then I don't know what we should do. I'm not going to intentionally downvote a whole bunch of posts by the same user.

We can actually flag and delete that other post as NAA, but I'd like to post a comment before we do to clarify the situation. But because it will likely be deleted soon after any flags or delete votes are cast, the author will likely not have a chance to respond, so it's important that I take extra care to avoid making false technical claims of my own. So I want to verify that 16.04 does actually offer to upgrade to 17.10.

Back when 16.10 was EOL but 17.10 was supported, 16.04 offered to upgrade to 17.04 and it was fully supported to do so. I have not personally checked yet if it will upgrade from 16.04 to 17.10, though.

Ugh my Software Updater on Lubuntu 16.04 is taking its time doing the equivalent of sudo apt update, even though I just ran sudo apt update and sudo apt upgrade like twenty minutes ago.

It's been on "Finished" for a minute or two...

Does anyone else here have 16.04?

I do on another machine, but now I'm going to sleep... I can try it in the morning

this was a frustrating exchange but maybe I was talking rubbish in those comments...

@Zanna No problem. I'll let you know if I can't figure it out.

that is very unlikely to happen :)

@Zanna I believe the dist-upgradeaction is still used behind the scenes when one upgrade by running do-release-upgrade.

This just describes upgrading the installed packages as "upgrading the installed packages" so I'd have to check the source code or otherwise look into it further to be sure.

Was the part you were concerned about, in your recent comments conversation, that part, or another part?

Regarding upgrading from 16.04 to 17.10, this supports what I am pretty sure is true and want to say, but I want a source that is more clearly reliable (or I want to be able to personally vouch for it).

Until recently, Ubuntu.com said, "Upgrading to Ubuntu 17.10 should be seamless from Ubuntu 16.04 LTS and 17.04." But that's gone now because they're no longer flogging 17.10 (the main download page just shows 18.04) and also it's a bit vaguely worded.

[ Natty ] phpmyadmin can not access phpmyadmin By: Dillyn Barber 3.5;

@Zanna Okay, it took me a little too long to figure out I could just check this way:

ek@Io:~$ do-release-upgrade
Checking for a new Ubuntu release
Get:1 Upgrade tool signature [819 B]
Get:2 Upgrade tool [1,253 kB]
Fetched 1,254 kB in 0s (0 B/s)
authenticate 'artful.tar.gz' against 'artful.tar.gz.gpg'
extracting 'artful.tar.gz'
[sudo] password for ek:

:)

[ Natty | Sentinel ] 16.04 Resume playback not working in VLC Snap By: user74364 6.5;

The releases wiki page hasn't been updated to say 18.04 is supported!

@Zanna I've commented on, downvoted, flagged as NAA (see the OP's comment and mine), and voted to delete that answer.

The misconception that release upgrades can never skip over a release seems widespread. (That's not NAA and I haven't flagged it, but I've commented.)

I've also commented on the question with the NAA answer with the hope that it will not be closed as a duplicate of the one with the wrong answer. (This is not mainly to prevent the spread of wrong information, though it should help with that; they're also not duplicates.)

@Natty fp

^^^ Though I'm glad it has a comment on it requesting it be expanded.

@Natty tp

Here's another answer that advises to manually edit sources.list to upgrade. It's accepted.

OT Mint

Unclear, I think, but what's the best way to request information?

@EliahKagan We may be able to dupe that to the canonical black-screen-on-boot question, though, as David Foerster has suggested.

I'm voting to reopen this question, even though we might end up reclosing it as a duplicate of the general EOL install/upgrade question.

What's the best question we have for this?

@Natty report meta.askubuntu.com/a/17866/22949

[ Natty | Sentinel ] Link to Post Low Length; No Code Block; 1.0;

Oh crap.

I tried to report a meta answer to Natty but it reported the main answer with the same post ID, which is not NAA nor otherwise flaggable!

Uh...

@Natty fp

@BhargavRao ^^^ Sorry to bother you again today, I just wanted to let you know about that in case something has to be done about it. I will not attempt to report any more Ask Ubuntu Meta posts to Natty (unless you tell me it's been changed to allow that). My fp seems to have registered but I'm not sure what Natty does internally with the same user fp-ing something they themselves reported. I don't want to train Natty wrongly.

Hah, that's fine not an issue.

I've not added Meta sites because the original intention was to not moderate meta.

Yes, that makes sense. I wasn't actually sure it would work, it just didn't occur to me that it would interpret it as an attempt to report a post on the main site.

@karel I think that doesn't deserve a NAA flag. We don't flag incorrect answers as NAA. (or is it different on AU?)

@EliahKagan I should perhaps change the regex to match just the sitename and not the meta site.

Thanks! :)

@BhargavRao If an answer attempts to answer a different question from what was asked, such that it would not be reasonable to interpret the question as including what the answer is answering, we do generally regard that as NAA on Ask Ubuntu, yes.

I don't know if that's technically different from how NAA is construed on other sites but I think that, in practice, we do view NAA somewhat broader here than on some other sites, in terms of using it on answers that are attempts to answer something but that are not attempts to answer what was actually asked, due to their authors having misinterpreted it.

I don't know that it makes sense for Natty to be told that such answers are NAA, though, unless there is something else wrong with them that would indicate by looking at them that they might be NAA.

Weird, it's completely different on SO.

This's the rules which we follow meta.stackoverflow.com/questions/265552/…

On our meta site we have this post which, while I think it needs to be brought up to date in some ways, I would guess won't be changed much regarding how to use NAA flags (because it seems to accord with what we usually do and what mods are generally OK with). One of the things it says is:

> If the OP is asking how to do foo and an answer explains how to do bar: flag as NaA.

Hmm, that's going to be hard to detect.

I can't think of a way.

I think we should not attempt to detect that.

Maybe special sub-cases of it, if they emerge. But I wouldn't suggest detecting it as a goal.

But a good AI/NLP project

@BhargavRao I must agree that it's weird -- even though I support Ask Ubuntu's practices on this issue -- because even the network-wide advice linked to from the help center does not list anything like that for NAA.

I don't like that practice. It is forcing the reviewers to have some knowledge of that technology. :\

@BhargavRao Also, I think it's better to err on the side of interpreting something as an attempt to answer what the OP asked than on the side of interpreting something as not such an attempt. Before flagging a post as NAA for this reason, I'll always check if someone has commented to articulate clearly why it's not addressing what the OP asked, and post such a comment myself if there isn't one.

True. Comments help the mods while reviewing. (not on SO, though)

@BhargavRao Enough to assess what the questions and answers mean, yes. I suspect that's why it's not feasible on a site like Stack Overflow whose scope is enormous (and also because the site itself is just way way bigger). As a separate issue, if mods had to do most of the heavy lifting for NAA posts, then it would also maybe be a bad idea, I think, because when NAA flags pile up and reviewers haven't handled them, it's harder for a mod to feel justified in skipping over a lot of stuff.

Note, however, that we don't regard a post as NAA just because it's utterly incapable of helping. Posts that people are confident can't possibly help and offer no value are sometimes deleted by 20k users but should not be flagged as NAA and, as far as I know, we don't have any policy or accepted norms of regarding an NAA flag as correct just on the basis of an answer being wrong, even if it is very fundamentally conceptually wrong.

(Sorry about the multiple pings. That chat message of mine was really confusingly written originally. I hope it makes some sense now.)

Ah, yeah. I guess it works on AU. Also, I think the mods would have some knowledge on almost every tech, so it'd not be that difficult for them, as compared to SO.

Ideally (in my opinion, anyway), such an NAA flag on a post that wasn't a clear enough case of the author having misinterpreted the question would get resolved as "disputed" through the LQP queue before mods had to do anything with it.

In effect, I think it would be accurate to say that what we do with NAA flags parallels the subjectivity of Unclear what you are asking close votes (and their corresponding flags from lower-rep users that feed into the Close queue). Figuring out if a question is unclear requires an assessment of what the OP is trying to ask, which often involves the application of technical knowledge. Similarly, trying to figure out if an answer is NAA involves figure out what the question is asking.

That is if the reviewers are fast.

@BhargavRao The answer wasn't flagged as NAA by Natty and I didn't true positive comment it for being not an answer, I true positive commented it for being an incorrect answer therefore destructive. Commenting that a wrong answer is wrong is a legitimate action. What do you think Stack Overflow would be like if wrong answers were allowed to propagate unchecked.

Oh, is this for that post?

Natty can't detect what answers are correct -- as I understand it, the effect of tp is that Natty will report more mosts that resemble that post in ways Natty is able to look for.

[ Natty | Sentinel ] sound Bluetooth speaker pairs but won`t connect By: Matheus Abner 5.0;

Btw that answer currently has no votes (neither up nor down). Do you still regard it as harmfully wrong (it's been edited, so I don't want to make assumptions)? If so, you might want to downvote it, and possibly add a comment if you judge the one on it to be insufficient.

@Natty tp

@karel yeah, that was my bad. I didn't know that NAA flags are allowed on incorrect answers on AU. Sorry! :(

No it's for a different answer, you have to follow the link in the preceding comment that was addressed to me to find it. btw that post is technically NAA because it doesn't answer the original question, which is about an upgrade, not a fresh install, although none of the reviewers of the question noticed it which maybe elevates the NAA to AA due to slackness in reviewing the question.

@karel Oh, sorry.

@BhargavRao They're not allowed on incorrect answers. At least not in the sense we think of when we say that. They're allowed on answers that attempt to answer a different question from what was asked, due to misinterpreting the answer.

It's this post sentinel.erwaysoftware.com/posts/86479/feedback

Ah, like a python answer on a java question? Even those aren't flaggable as NAA on SO. :\

@BhargavRao Like a doing Y with Java answer on a doing X with Java question, where X and Y can't be reasonably considered to overlap.

Okay, those are not flaggable as well. SO's NAA rules are pretty strict. Good that you guys told me, learnt a new thing today.

@BhargavRao Wait... that is the one I linked to.

Erm, are we 3 are confused? I was talking about that post as well.

@BhargavRao I think that message of mine characterizes the distinction we make. It's not authoritative, though; it's just my attempt to restate the usual norm and practice as I've observed and sometimes participated in it. If I find something clearer and more definitive that is written about this specifically, I'll let you know.

I don't know about you but I've got the code open in the IDE and I try one answer after another in Stack Overflow and they all fail like clockwork and in the end I have to debug my own code to save time. Deleting reproducible wrong answers at Stack Overflow would save me a lot of time.

@EliahKagan Cool, thanks.

@karel But I think it's only useful to report things to Natty, or give Natty positive feedback on, posts that help it know what to tell us about in the future. So that's NAA, VLQ, and, when it comes up (though Natty isn't really for this), spam and rude/abusive. Feedback to Natty doesn't affect what happens to a post--the only way it would cause a post to be deleted, or deleted sooner, is if someone in the room sees the message (and in that case, just a message to the room would work).

It definitely helps to train it.

@karel Why?

That's what I thought those tp comments to Natty were for is to help to train it to improve the accuracy.

NAA (and the "solution" as per the external link is something like a hardware problem after losing OEM charger and then using a 3rd party charger.)