Parlez-vous Drupal? Translating a complex Drupal site into a second language.
What happens if your client has a complex Drupal website and decides it needs to handle another language? What if you need to develop and maintain the two sites separately: the old one and the new multilingual one? What if you have to work on both sites simultaneously? What if there are lots of images with text in them? What if the site uses Panels along side a large list of other contributed modules?
This session will cover how a large, capable team of Drupal project managers, developers, and content editors rallied together to successfully architect, configure, and translate a technically-challenging bilingual website within a very tight deadline.
We will cover:
- Project needs and prep:
- Create a French/English language switcher
- Adding and displaying French and English content
- Apply translations to a complex site that makes heavy use of Panels
- Client and Dev team requires training on adding and editing different types of content
- Initial (unilingual) site creation:
- Understanding exsiting structure of our complex heavily-customized Drupal 6 site.
- Porting our site to Drupal 7 and freeing us from Features.
- Building and planning for an infrastructure that supports multiple sites spawned off of a single code base and a single database starting point.
- Analysis, planning, and implementation for multilingual support:
- Site architecture analysis
- Module selection and configuration
- Pain points… aliases, custom code, panels & SEO
- Content translation, import, and hand-off:
- Track and manage the translated content import process.
- Dealing with images and heavily stylized / coded pages.
- Training the team: client content managers, developers, QA team.
- Pain points… finding strings to translate, my string search is case sensitive?, so many places content can exist!, content import during ongoing development.
- Lessons learned!
- Expect multiple rounds of content QA, even if you import translations after development is complete.
- You are never actually done translating everything, even when you are confident you have to be.
- Planning for multilingual work can only get you so far, you have to get your hands dirty to find all of the edge cases.
Comments
Gábor Hojtsy replied on Permalink
This sounds like a great talk proposal, very well thought out outline :) Maybe not even enough time to cover all the great topics, so I'd think about removing some of the less related items (at least according to my understanding), eg. the multisite and D6 -> D7 migration. That would let you focus more on the core topic.
susanmccormick replied on Permalink
That's a good point. Thanks for the input!
Kristen Pol replied on Permalink
Yes, we can make sure to streamline it to focus on i18n :)