Hi there.
I’ve just picked up on this thread...
Quote from: OpenGeek at Mar 12, 2010, 07:57 AM
Hopefully we’ll have a project that pays us money to build a proper localization solution in the manager that supports these Context-isolated languages, let’s you quickly get translation done, but still allows you the flexibility of developing custom structures and features in each language as needed. I’ve personally never built a multi-lingual site where the site structure or even content layouts were the same across the localized sections.
YAMS does let you create different layouts per language quite easily by the way. The templates can have language dependent content like everything else.
OpenGeek, while I agree that there will be situations where people will effectively want a completely different website per language - that is, a different website structure, layout and style - and this can be handled in Revo, a lot of people don’t want the overhead of managing multiple websites. They are happy to have with a website where each language has the same document structure, but possibly a different layout and styling per language. The feedback I have got from people who have tried YAMS is that they really love having all the language variants in one place via the tabbed interface rather than having to flit between different document trees. The burden on the end user who eventually has to own the site is far less. While there may be some people for whom scalability is a high priority, for most it isn’t. I’ve got no evidence that there are scalability problems with YAMS, but I also don’t know of anyone that has used it to generate a massive site. So, what I am saying is that there does appear to be a large demand for a multilingual solution like YAMS and that demand will spill over into Revo.
I definitely do agree with OpenGeek on one thing. The implementation of YAMS is a ’hack’ in the sense that necessitates overriding a lot of core functionality to get it to work. It also involves having to customise many of the standard snippets in order to get them to produce multilingual content. This isn’t actually as bad as it sounds. Normally it’s just the templates that the snippet uses that need to be customised rather than the core code of the snippet itself, but it’s still an unpleasant feature. A good multilingual solution would ideally be more seamless than that.
I’m still quite ignorant of Revo and what it can do. I haven’t really found the time to play with it properly yet, but from what I have read so far it does seem like while it would be possible to manage different subdomains from within a single install quite easily, there is nothing inbuilt that will allow them to share the same document structure so that at least that management overhead could be reduced... Am I right? My initial investigations lead me to the conclusion that if I wanted to implement YAMS functionality in Revo, then I would end up having to do the same ’hacking’ as I did in Evo to get it to work with multilingual aliases and the like - which I find very disappointing. I was hoping that there would be features built in to Revo that would make it easier for users to develop multilingual sites in a similar way to YAMS. My conclusion is that, as Charliez notes, in terms of developing multilingual websites, the hurdles that would need to be overcome in Revo in terms of setting up a structure that allows multiple language variants of content to be easily managed are very similar to those for Evo.
MODX Staff
