I'm looking for the best multilingual 'tag' approach.
Currently I'm exploring the Tagger addon from DevDemon.
I was thinking about splitting up the tags in different 'language'-groups,
but that doesn't seem like a good solution. Because, when adding/editing an entry...
you would have to add each tag in all the different languages, wich will result in a huge (inconvenient) list.
Has anyone done this before & what was your approach?
Pros & cons?
I've been longing for a multilingual tagging solution myself. You're right, even with groups, I'm not sure Tagger will quite get you there. Groupings are not visible in the entry screen, for example. Separate language groups would help you display language(group) specific tag clouds for example, but you'd have to add the tags then group them in the module's settings panel. Not ideal.
The one way I have done it for multilingual, though far from perfect, is to create a channel for tags, and use Playa to relate the tags to your entries, using categories to provide simple organization of the tags (starting with language, for example). Not as friendly as Tagger would be if it had a simple language "switch" available, but it does centralize the tags and allows you to use the Playa filter options to find the tags to assign. But this has its limitations too - the obvious one being that, unlike with tags, you cannot add new tags from within the entry screen - you'd have to add the tags first to the tags channel.
It's not ideal, but our approach has been to use primarily 'English' tags, unless a particular form language tag shows up a lot in search results. We use Tags by Solspace and it's great, but what we found is just as @Think Graphical mentioned, "the tags get cluttered very quickly, with all the multiple languages".
Our system, maybe different from others, so obviously, evaluate against what it is you're trying to accomplish. Our client basically produces multilingual publications on various issues concerning consumers— from credit to banking to identity theft, etc.
Frequently, the person searching for the publication in a particular language, doesn't actually speak that language or maybe can't read well in the language. For example, a social worker helping a first generation immigrant from Korea or the grandchild of a Hmong immigrant. The grandchild speaks English and Hmong, but may not read the language at all.
We found, when testing, that english language tags got more use than the foreign language tags, but search was a different matter. Since we were trying to limit the number of tags per entry (reduce clutter, improve/keep SEO ratings), removing tags duplicated in multiple languages was the first thing to go.
For us, search is fully multilingual compatible, and that's what we encourage for looking for multilingual words and phrases.
