Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Users should not be able to add entries to other languages' glossaries by default #8807

Open
tjhietala opened this issue Feb 17, 2023 · 11 comments
Labels
enhancement Adding or requesting a new feature. undecided These features might not be implemented. Can be prioritized by sponsorship.
Milestone

Comments

@tjhietala
Copy link
Contributor

Describe the problem

Currently, if a user has the rights to add a glossary entry, they can add it for every language by checking the "Terminology" check box. I don't think this is good default behaviour for Weblate. Yes, the unneeded entries can be deleted, but it's an unnecessary annoyance.

Describe the solution you'd like

The "Terminology" check box should be disabled by default, except for managers (or possibly also reviewers, if deemed necessary).

Describe alternatives you've considered

No response

Screenshots

No response

Additional context

No response

@github-actions
Copy link

This issue has been automatically marked as stale because there wasn’t any recent activity.

It will be closed soon if no further action occurs.

Thank you for your contributions!

@github-actions github-actions bot added the wontfix Nobody will work on this. label Feb 28, 2023
@comradekingu
Copy link
Contributor

Has this ever been a problem?

@nijel nijel added enhancement Adding or requesting a new feature. undecided These features might not be implemented. Can be prioritized by sponsorship. and removed wontfix Nobody will work on this. labels Feb 28, 2023
@github-actions
Copy link

This issue has been put aside. It is currently unclear if it will ever be implemented as it seems to cover too narrow of a use case or doesn't seem to fit into Weblate.

Please try to clarify the use case or consider proposing something more generic to make it useful to more users.

@tjhietala
Copy link
Contributor Author

@comradekingu Yes, on The Document Foundations's Weblate this happens every now and then. I just deleted some unwanted entries yesterday. It's not a huge problem though, but IMO what I propose would be a better default.

@comradekingu
Copy link
Contributor

@tjhetala Pretty sure I am one of the people doing it for the source language.
The idea that users can't edit all languages has cost the TDF Weblate dearly.

Similarly, between good entries and bad entries in the glossary, the benefit greatly outweighs the detriment.
If you really wanted to do some damage, resetting all of them would be it.
In so doing, better functionality in resetting bulk changes or per-user changes for a given timeframe would be better.
Banning people instead of limiting functionality is also a better idea. In time there is vested interest and a purified pool.
TDF Weblate is hard to get into, and much harder still to get any changes made upstream in.
It would be wise to not model default functionality on the TDF Weblate.

@timothyqiu
Copy link

timothyqiu commented Aug 4, 2023

There are two situations that could be a problem.

First, some glossaries only make sense in a specific language.

For example, German users might create two glossaries for "lion", one for the male version "der Löwe" and one for the female version "die Löwin". This does not make sense for languages that don't have such grammatical gender.

Second, "Auto-adjust context when an identical string already exists." is on by default too. So I regularly see different versions of the same string with auto-generated context like "1". That's is so frustrating to maintain. I guess it's because some languages created a non-terminology glossary:

  1. Language A user added a non-terminology foo.
  2. Language B user added a terminology foo, but it got adjusted into foo (key 1)
  3. Now Language A users have both foo and foo (key 1).
  4. What's worse: only language A has foo, and language B-Z has foo (key 1)
    • You can't delete foo (key 1) because these 25 language might already translated this string.
    • Deleting foo feels wrong because you don't actually need that auto generate context key 1.

@tjhietala
Copy link
Contributor Author

Apparently it's currently also possible to add the "forbidden" flag to other languages' glossaries by default, see: https://translations.documentfoundation.org/translate/libo_ui-master/glossary/fi/?checksum=6a157b0604755bd8

IMO this should not be allowed by default either.

@tjhietala
Copy link
Contributor Author

@comradekingu I don't see how banning users would be preferable over setting good defaults. I think the people doing these changes are by and large doing it by accident, i.e. they don't know they're affecting other languages' glossaries.

@comradekingu
Copy link
Contributor

comradekingu commented Mar 15, 2024

@tjhietala I disagree that limiting potential for getting work done is a good default to prevent harm.
For how often malice occurs, it is exceedingly rare compared to the upside of just having stuff done well, or in a well-meaning manner.
There are no easy ways to revert changes from a particular user over a given time-period are lacking for every possible type of change.
It is already possible to fine-tune what permissions different groups of users have.
Are we trying to find something that works for Hosted, or are we trying to shoot everyone in the foot because
people with special needs can and do run their own instance?

Weblate is IMO a better platform because of default openness and good functionality to deal with that,
and I like the one-big-stick method of dealing with the few problems that occur.
The alternative, as demonstrated by platforms that suck, is a bunch of authority-by-role, built on the idea that assignable blame is a good way to organize. It is a trashy way to mitigate functionality that doesn't work.
First-come basis is not a measure of quality, as the better translators often arrive later.
The task of picking a team of people with more editing rights is the same premise as being able to do all the work yourself in every language. I have never seen it done well.
Over time one may end up with a good team, and I pretty much know who to trust for most languages now.
I still don't, and people making their first project especially don't.

If the problem arises from it being difficult to gauge what dictionary the user is editing and why, that is the problem to solve. Having the same problem for fewer users is largely the same problem.
I care about correctness over time.

If malice is what I want to do, changing the glossaries when the translations are wide open without quality-reaffirming tools is not what anyone clever enough to do real damage would attempt.

Where are the examples of what went wrong, and maybe we can work from there to avoid the confusion?

@tjhietala
Copy link
Contributor Author

@comradekingu I don't think there's any malice involved. Like I said, my guess would be that these users think they're modifying the glossary for one language and don't realise they're doing it for other languages as well.

@fitojb
Copy link
Contributor

fitojb commented Apr 27, 2024

I spend a lot of time removing unwanted entries from my glossaries added inadvertently by other translators not in my language teams. They use other capitalization conventions, and I have to go through them one by one. So yes, it is annoying.

@nijel nijel added this to the 5.6 milestone May 7, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement Adding or requesting a new feature. undecided These features might not be implemented. Can be prioritized by sponsorship.
Projects
None yet
Development

No branches or pull requests

5 participants