Translation bubbles

[LightKnight]

I have already argued in favor of a better space for translations' justification.
There has been a change recently (and I feel stupid commenting it so late), and there is a problem with it. It is the return to previous translation : when it comes to this, there is no possibility to justify our change, which is a big problem !
Could this be fixed, please ? I always take the effort to justify my changes, (which by the way not all are doing, but anyway), and I'd like to go on this way.

[Een]

I have already argued in favor of a better space for translations' justification.
There has been a change recently (and I feel stupid commenting it so late), and there is a problem with it. It is the return to previous translation : when it comes to this, there is no possibility to justify our change, which is a big problem !
Could this be fixed, please ? I always take the effort to justify my changes, (which by the way not all are doing, but anyway), and I'd like to go on this way.

Sorry, don't see how to do that: comment is linked to the translator who made the translation, and when you revert to a previous translation, you basically "give it back" to the previous translator (which is right if that translation was best). So it's only that translator who can comment the translation...

[LightKnight]

Previously, changes that put back an old translation were considered as new ones, and so the comment going with it. So there was no problem.
It is normal, and important that you justify why going back to a previous translation, because it's a new action, it is not neutral.

[Een]

Previously, changes that put back an old translation were considered as new ones, and so the comment going with it. So there was no problem.

Actually, there was a problem, because someone could do the right translation, then someone else mess it up, then another person fix it, and the first person to do things right was not credited for it.

[LightKnight]

Ya, I can understand it's the problem.
I wonder if the most important is the efficiency of translations and the communication that goes with it, or the justice in the rewarding of translators, but anyway.
Is there no possibility that we allow people going back to previous versions to justify in a common bubble ? By the way, some moments I was allowed to do so, for a few seconds (then the bubble disappeared), and so once I could only put a "1" that we still see, in the bubble of the previous translator.

[Een]

I wonder if the most important is the efficiency of translations and the communication that goes with it, or the justice in the rewarding of translators, but anyway.

Both, of course

Is there no possibility that we allow people going back to previous versions to justify in a common bubble ?

Not sure, could be difficult to twist the way things have been built. Does that really happen often?
Other translators feedback on this would be welcome. I can try to look into this if that's meaningful for many translators.

By the way, some moments I was allowed to do so, for a few seconds (then the bubble disappeared), and so once I could only put a "1" that we still see, in the bubble of the previous translator.

Strange, that shouldn't happen
I'll check this.

[diamant]

Previously, changes that put back an old translation were considered as new ones, and so the comment going with it. So there was no problem.

Actually, there was a problem, because someone could do the right translation, then someone else mess it up, then another person fix it, and the first person to do things right was not credited for it.

You could credit the first person to do the right translation without destroy the whole history of translations with the associated comments. When someone revert to a previous translation, just add an automatic comment with a special color (which cannot be edited) to give the references of that previous translation.

[Een]

You could credit the first person to do the right translation without destroy the whole history of translations with the associated comments.

That's worded quite negatively (history is not destroyed) and not very clearly. I'll suppose you mean having reverts show chronologically in the history.

When someone revert to a previous translation, just add an automatic comment with a special color (which cannot be edited) to give the references of that previous translation.

It may seem right abstractly, but it"s technically less workable than the current compromise, and doesn't answer the request of the op for an editable comment for reverts (since you propose an automatic non editable comment with a reference - not a currently existing functional concept - to the previous translation).

Also, reverts *should be extremely rare*. I suppose I'll have to check that.
If that's not the case, the problem is not really technical, probably more about guidelines and allowing translators to report bad practices.

[diamant]

You could credit the first person to do the right translation without destroy the whole history of translations with the associated comments.

That's worded quite negatively (history is not destroyed) and not very clearly. I'll suppose you mean having reverts show chronologically in the history.

Désolé si mon anglais manque de nuances. L'historique n'est effectivement altéré qu'en cas de retour(s) à de précédentes traductions, auquel cas les "actions d'annulation" n'apparaissent pas, ce qui rend l'historique peu intelligible en raison du "brouillage" de la chronologie des actions.

When someone revert to a previous translation, just add an automatic comment with a special color (which cannot be edited) to give the references of that previous translation.

It may seem right abstractly, but it"s technically less workable than the current compromise, and doesn't answer the request of the op for an editable comment for reverts (since you propose an automatic non editable comment with a reference - not a currently existing functional concept - to the previous translation).

Also, reverts *should be extremely rare*. I suppose I'll have to check that.
If that's not the case, the problem is not really technical, probably more about guidelines and allowing translators to report bad practices.

Ici, l'idée d'insérer un commentaire automatique NON éditable en cas de retour à une précédente traduction n'empêche en rien l'auteur de l'action d'ajouter son propre commentaire à la suite, ce dernier commentaire restant éditable à l'instar de tous les autres commentaires faits par les membres. Le commentaire automatique permet d'une part de référencer la traduction qui sera créditée, d'autre part de signaler, par une couleur différente, l'action d'annulation non créditée.
Contrairement à ce que vous pensez, ces retours à de précédentes traductions sont fréquentes, ne serait-ce que pour corriger les erreurs (sachant qu'aucune validation n'est requise avant l'enregistrement d'une nouvelle traduction).

Édition du 14/07/2017 8h48 : NON manquant

[Liallan]

I have to agree with LightKnight. (I thought I understood what diamont was saying until I used the translator on his French, and now I'm confused too.)

So right now, if #2 is more correct, and I would be #4, it doesn't list #4, but just shows #2 as green and currently correct instead of #3. Since it turns green, the system obviously knows which one is the current one. If it's not yet validated, it knows when the time comes who gets that credit.

If instead it would post and show #4, if it recognizes that it's a duplicate of #2, it could still post #4, but still make #2 the green/current one, and credit that person. It obviously does know #4 is a duplicate or it could not follow its current behavior. And it's not going by date because #3 is the newest, but #2 is the "currently correct" one.

Isn't there a reasonable way to simply make #4 post (so we can do comments), but still have #2 show as the current correct one? The only difference is whether #4 shows up or not. It seems that's a lot easier than trying to make a special way to add a comment and that sort of thing.

I don't know that I've reverted back to prior ones "a lot," but sometimes. There are times I'd like to, but without the ability to add a comment as to why, I don't like doing it. And someone could just change it back again if they don't see the reason.

Also, if someone honestly believes a prior version was best, but they're really hung up on getting credit, it might tempt them to change some other minor thing just to have theirs get credit, instead of simply reverting to the best one.

I think credit is important as well, and since everyone doing this (including those fixing English) is spending their time, I do think it's legitimate to get the "gifts." But, that said, if there isn't a good way to get both and I had to pick between having the best version or someone getting credit, I'd pick having the best version. If someone is more concerned about getting credit, well, maybe they shouldn't be doing it.