I gather that ask.fedoraproject.org site has a new interface.
But, what happened to the old posts. I used to store the URLs of posts I was interested in.
Where are they ? For example below was a URL I stored for future reference. This URL is dead now.
Is this post archived somewhere ?
I gather that ask.fedoraproject.org site has a new interface.
Sadly they’re going to vanish. For now you can still see them here in read-only mode:
It isn’t just a new interface, it’s a whole new platform. We are now using Discourse instead of Askbot.
If there are any questions that you think are worth keeping around, please consider adding them to the quick-docs (or opening tickets there so someone else can do it if you can not).
Shouldn’t the presumption be to try to not break links, rather than have end-users reestablish them? So many pages widely distributed across the www point to old ask.fedoraproject.org pages that it’s become frustrating to even bother doing google or other searches for fedora questions.
This just compounds the badly degraded state of http://docs.fedoraproject.org/. Between these two, it is terribly hard to find up-to-date info on how to solve any fedora problems.
I agree with this. Off the top of my head, I’d say take questions from 28 or 27 onwards and try to import them to discourse, maybe on a category of its own, “Archives”, if you may. It’s really confusing searching for something in the site only to realize now you have to do a workaround to visit the old site.
I’ve said this already: there’s no straightforward way to migrate the information from askbot to discourse. We really did think of all this when the move was being planned. We didnt just set up a new platform without any considerations
There are many presumptions that can be made, and none of them are based on any data since askbot is hard to get metrics off: one of the many reasons why we decided to move in the first place. The presumption/plan that was simplest to implement with the resources available to the community was that queries common through releases should be investigated, reviewed, and added to quick-docs. So if you think imortant information is missing, please open an issue or a pull request to add it to quick-docs. In addition, the other assumption was the majority of queries that are release specific will be asked again, on the new platform where users can engage with each other and the community.
What degraded state? (It just moved to a new software pipeline to make contributions easier). Have you filed issues or pull requests, to improve these issues that you’ve found? What problems have you been unable to solve? Please open topics here and others will help you out.
If you look at the forum, we’ve answered all these questions again and again, but really, if things arent as good as you think they should be, please join the docs team and help them out instead of documenting your frustration on community forums. The latter doesnt help: we still cannot migrate information from askbot to discourse, no matter how much time you spend saying it should have been the way to go
Wow. What a dismissive response. I didn’t realise I had found my way onto reddit. Clearly a blunder on my part. Newcomers clearly not welcome here.
Dismissive? Right, i apologise if thats how it sounds. I have tried to:
- explain the idea
- asked for information on what shortcomings youve found
We try to please and clearly we dont always succeed all the time. We’ve received responses from “awesome, thanks!” to “this makes no sense”.
In the meantime, will you identify the issues you referenced so that others can try to improve them? Youve not done that- still only a comment that doesnt help on that front.
My answer is not really an answer, but a comment / follow-up question :
why was it not possible to keep the askbot installation around in some kind of read-only-mode and advice everyone to look on discourse for newer topics?
This would keep old links working and new questions or requests which are not up to date could be reissued on discourse… ( advance: no migration of questions required )
EDIT: I just noted, that https://askbot.fedoraproject.org/en/questions/ IS a read-only versions as suggested above. So the chosen solution seems to be already the one suggested in this post. So why are the old links broken, if everything else ( at least seems to ) be still there / available?
I don’t know if this works often enough, but if you are really desperate ( I don’t know if / how long https://askbot.fedoraproject.org/en/questions/ will work ) :
have a look at archive.org with the links you collected, perhaps you can reach them via their service.
That’s because we reused ask.fp.o given how well known it is, and moved the older instance to askbot.fp.o. So until the search engines update their indexes, some links will point to non-existent pages on ask.fp.o. We’re working on redirects and a better 404, but given that askbot will be turned off at some point (a read only instance still costs), it isnt high priority.
According to this post the plan is to make an HTML-only version of askbot. I really like that idea, because that means:
- no content gets lost
- and you can (if possible) still redirect 404’s to the old archived versions
I think that is probably the problem. It would have been better to fade out the old one to a read only / static version and introduce the new one on a new URL.
I understand the reasoning for the way things went, but it really did break quite a bit of the fedora internet.
Because I upgraded an ancient machine to 30 yesterday I had to google a few times to solve problems. At least five of the links I wanted to read ended up in 404 and I had to fall back to the awful stackoverflow.
Maybe just something to consider if we are migrating other tools to new ones in the future.
- Why you are going to undertake a task like that and invest hours in this? , I will explain to you What you’ll find there:
- Accept with or whithout EOL (This is Important, Fedora make important improvement in every release, sometime and answer is not the best or there is another choice, flatpak coming and so on…)
- Open Question without answer
* too subjective
* not a real question
* question is not relevant or outdated
* too localized
- Quality of the answer -> Another important point here, we don’t want to duplicate this behaviour here, because we have some disappointment related to this in the old instance.
- Wiki, Sticky (Currently moved to quick-docs)
Why keep all that stuff…??? make sense or not…?
It’s all about impact in the infractructure:
- Changes in Anaconda Banner.
- Get fedora Web page.
- RPM fusion.
- and so on…
What will the next task when askbot close? search all post that give 404 and delete from google…
That is something that we have to front now or later…
The big answer has been given right before your post:
We have many resources out there and some actually are not outdated. And getting 404 links for all that is not nice.
After all you can put a warning banner at the top saying content can be outdated and redirect to the new instance or so. But I guess many visitors come there thanks to search engines. And these need content.
Your list of questions that aren’t worth keeping includes those that were rejected by the Moderators for various reasons. However, those posts were deleted (If not all of them, those that were marked as Spam certainly were.) so they’re not around any more to waste people’s time.
I don’t know how to elaborate – hyperlinking is so fundamental to the web that not breaking links is a hard requirement. You just don’t do it. Asking for more detail is like you poking me in the eye, and when I complain you ask “I need more detail! What precisely about that poke don’t you like”.
Anyway I suppose it’s too late now so I’m just adding more noise. It’s obvious to me the original content (still linked from documents all over the web) should have been left up in some sort of static or read-only mode. Then a new domain or subdomain or path should have been used for the new system. Never mind, it’s done now.
Firstly, bear in mind I (like any new visitor here) have no idea of the personnel or institutional arrangement behind all this. I was referring to docs.fedoraproject there, and don’t know whether the same people are behind docs and ask, or whether they talk to each other. I have zero idea what ‘askbot’ is.
What I meant by the ‘degraded’ state of docs is that almost literally every time I search for information on how to do something in Fedora 29/30 (and, like everyone, this is via google or ddg), this is the routine:
- find something that looks relevant on ask.fedoraproject, but the link’s broken so I move onto docs.fedoraproject
- the article that seems to address my issue is from some earlier Fedora version (IIRC 14 & 17 seem common for some reason)
- Flail around with the article to see if I can shoehorn its instructions into what’s on current Fedora, but can’t
- Browse the Fedora 29/30 docs to see if there’s an equivalent article. There isnt.
- Back to StackOverflow
This happens nearly every time. It seems lots of stuff was documented for earlier versions that just isn’t done now. Again bear in mind an outsider doesn’t know how all this stuff works - it may be all done by volunteers, in which case not keeping docs up to date is understandable. As I wrote in my comment, the problem is that it compounds the missing ask.fedoraproject stuff.
I, like 99.999% of humanity, am not an OS hobbyist. This is a pretty unrealistic suggestion for other than a minuscule minority.
Honestly, I find looking for solutions to problems with Fedora enormously frustrating. Perhaps there are better or more productive ways to express that (short of joining the docs team!). But as it slaps me in the face at least once a day, it would be require powers of restraint mightier than mine not to vent at least one time. But I’ll keep any further comments I make around here productive.
Again, I reiterate----if the data could’ve been migrated, we would’ve done so.
It is not, and I would very much appreciate you not making such statements. There’s place for hyperbole, but this is not it. Please take a moment to read the code of conduct and conduct yourself accordingly in any and all participation in the community.
Again, we thought about it, and we’d rather that the URL ask.fedoraproject.org, which is extremely well known and is mentioned in many places as the current support forum. point to the current support forum.
Where you are adamant that links provided by search engines should not be broken (they’ll go away over time, by the way), we preferred that links that say “Head to ask.fedoraproject.org for help!” remained functional. Askbot will not remain up forever—it requires community resources of which money is one, and we cannot keep paying for two platforms especially given that askbot only contains information on past fedora releases----a majority of which is now outdated.
This is perfectly OK—we do not expect new users to know how the community functions. However, what we would like is for them to ask questions to find out and learn more before airing their frustrations.
And why not ask the fresh question here on a platform set up by the Fedora community for Fedora related queries instead of trying solutions for older versions of software and asking around on general forums?
And the simple solution is to ask a new question if you are unable to find the information. This platform is not documentation. Yes, some questions have been asked before, and their solutions can be re-used, but given how software changes, it is safest to ask a new question. If it has been answered, someone who knows will point you to it, otherwise they’ll help you out afresh.
That number is made up …
It isn’t about being an active hobbyist at all. It’s about being slightly more understanding of other’s decisions. Even if this were a support forum for a product you bought and paid for, airing your frustrations without establishing how and why things are so is still not the best way to go about it.
We appreciate it. I would, as a start, request you to also ask yourself “am I being excellent to everyone” before posting. That always helps us all ensure that our conversations remain friendly and productive.
I would ask what is wrong with dedicated forum on stackoverflow, that everyone understands and love. And there will be no problem with exporting data when something will collapse… And now we are doing what we are doing before, we were trying to convince people to use poor tools like askboot, that simply doesn’t work at all. And now we are trying the same…