That is something I would like to point out, Is we permit add categories to everyone, we can have tag’s for every-language…
WDYT?
People can already choose whatever category they wish, and similarly they can use whatever tags they want to after reaching a trust level. They can use whatever language for tags too. We cant limit a tag in a particular language to the related language category if thats what you are asking. The mods will have to do their cleanups from time time if users mix languages up.
1 Like
I honestly dont see much of a difference. Given that most forums are happy with the default icon set, this is a low priority issue that we’ll look into it after everything else is set up.
1 Like
To clarify, we are not setting up tags for users to choose from. We are letting users create and manage tags as discourse allows them to do based on the trust system. Is that ok?
2 Likes
All in all it is how Discourse works.
1 Like
I’ve deleted the language groups now since we’re not using them.
1 Like
About the old Ask Fedora instance, I also think we should either find a way to import all/most of its posts, or should have it running along side this one for a while so that good answers can be copied here if needed; or at least have a database dump or even some kind of plain HTML archive somewhere. For now, a database dump should be enough to enable later processing; e.g. creating a plain HTML archive.
I think, there will be a lot of recurring questions there. Just thinking a little bit, I can remember things like fixing grub problems, UEFI related issues like signing third party modules, nvidia driver installation tips like selecting appropriate driver, repairing duplicated packages with dnf, …
2 Likes
Im -1 for this. It sounds trivial in theory but we dont have the man power to do it. Id prefer if people asked these questions again and we rebuild our knowledge base from scratch on this new platform instead of referring to mostly old advice from askbot. If you identify common issues, please put them on quick-docs. Thats the path we are working towards.
(I.e., improve quick docs to include common howtos and tweaks instead of keeping aslbot data around)
1 Like
Well, I’m OK with the quick-docs approach; but I wonder if an archive of old ask fedora would help creating content for quick-docs too.
About the man power, that was why I’m even OK with a database dump. Someone might be able to do something with it in future!
1 Like
So we cant keep askbot running because thatll require us to keep paying for it. The database dump isnt that simple either because it will at least have to be anonymised before it is made available (and available where?).
(So, if stuff needs to be moved to quick-docs, it needs to be done now before we stop our askbot subscription.)
1 Like
I’ll look into possibility of crating a static snapshot of current Ask Fedora, that could be hosted as loads of html files.
You will probably still need to anonymise it. IANAL, so you should check with -legal.
I was thinking about getting all questions with wget, without logging in, and removing some unwanted elements (edit/login links, form elements, most/all JS, etc.) with regexes (I see that there’s some read-only mode in Askbot, which would simplify the clean-up). So everything publicly available and licensed under a CC-BY-SA 3.0 and, in case of Askbot, GPLv3. I’ll ask on legal@lists.fedoraproject.org.
On second thought what also might be useful:
- user profiles might be required to satisfy attribution requirements (also only public part);
- disclaimer of it being a read-only archive;
- some email address for complaints (hidden behind captcha);
- information about and link to Discourse and Quick Docs.
Would it be possible to put it under e.g. https://olddiscussion.fedoraproject.org? I’ve got hosting where I can create sub-accounts, but if it takes something more than some random dude’s hosting for a subdomain of fedoraproject.org, I could send ready files elsewhere.
Redirecting (with just few rewrite rules) to this new domain all addresses starting with language codes (the Askbot link structure: /en/question/…) would minimise number of 404s on Discourse (important for site’s reputation in search engines) and land people coming from web searches on questions they were looking for.
https://web.archive.org/web/20180320053101/https://discussion.fedoraproject.org/en/questions/ ?
Edit: not complete, questions arent archived
How about we think of where the data will go after youve investigated how easy it is to get it?
I still think it’s not the best use of the limited resources we have—I’d really prefer it be spent on quick-docs—but it’s your time 
As a side note, we still havent seen much action on quick docs: Pull requests - fedora-docs/quick-docs - Pagure.io
Could we move more of the important questions that we’re trying to save on askbot there please?
(I will be back to it myself next week—waiting on my new machine, the old trusty one finally passed away a few days ago 
)
@ozeszty It’d be really great if you can do it
@FranciscoD I do understand your position, but I don’t agree. I’d say that there is lots of valuable info on current ask fedora, and our limited resources are not enough to move (or even identify) the useful ones in a short time.
I for one might not be able to put some time for converting even my own answers to quick-docs in short term. And if I came back to quick docs later when there is no access to old data, I really don’t like to spend my time on something that I’ve already spent time some day.
In other words, it’s true that the content in current Ask Fedora was ‘free’, but people have spend ‘valuable’ time on them and it should not be taken lightly. It is really a waste. Even for Fedora, the time spent on rewriting data can be used to do new things.
2 Likes
Will someone please point out at least some of these bits that we want to spend all this effort keeping? Maybe others can work on moving them now?
A plan that is based on “someone in the future will move it” is not really a plan. Someone has to volunteer to lead this work, to get people interested, to convince people to help out, like any other task/project in the community.
I’ve spent quite a bit of time on askbot myself https://discussion.fedoraproject.org/en/users/, but it has been very rare for me to have answered two identical questions unless they were repetitions of common (sticky) issues. Each troubleshooting issue is more likely to be unique than a repetition. So if we are going by anecdotal evidence, mine does not suggest that the info is worth archiving. Instead, similar to how we troubleshoot issues on the IRC where there is no access to past discussion, it suggests that each new query needs to be dealt with as a new one by starting the diagnostic process from scratch, i.e., working with each user, going through the investigation (and teaching them bits in the process). The info on askbot is not going waste—it already served its main purpose by helping the users that asked their questions.
This is the last I’ll say on this topic for now. When we do have the archive, we can see what can be done with it. Until then, lets get back to getting this instance ready for public use. Are we only waiting on the start here docs? Is anything else still pending? 
@ozeszty: we’re thinking of pulling the plug on askbot and formally making this instance ask.fp.o before F30 releases in ~4 weeks. Would that be enough time to pull your archive, do you think?
@not-mattdm-at-all: is there a date when our subscription to askbot expires? Would it be possible to keep it in “readonly” mode until the expiry date—we can disable logins, and move the URL to archiveddiscussion.fedoraproject.org (or something)?