adaptTo() Website Requirements

Quantities Today

• Assets total: 1,914
  • Images: 1,429
  • Downloads: 324
  • Content Fragments (Speaker): 161
• Pages total: 831
  • Number of Sites: 13
• Tags: 14

Feature Breakdown

• AEM 6.5 application
  • 9 templates
  • 23 content components
  • 12 structural components
• Main Features with Examples
  • Homepage: https://adapt.to/2021/en.html
  • Content Page: https://adapt.to/2021/en/conference.html
  • Gallery: https://adapt.to/2021/en/conference/gallery.html
  • Speaker Listing: https://adapt.to/2021/en/conference/speaker.html
  • Speaker Details: https://adapt.to/2021/en/conference/speaker.detail.suffix.html/bertrand-delacretaz.html
  • Schedule: https://adapt.to/2021/en/schedule.html
  • Talk: https://adapt.to/2021/en/schedule/repoinit-a-mini-language-for-content-repository-initialization.html
  • Newsletter registration: https://adapt.to/2021/en/conference/newsletter.html
  • Ticket Shop (IFrame Integration)
  • Talk Archive: https://adapt.to/2021/en/archive.html
• Notable Functionality
  • We host multiple versions of adaptTo() websites on the same domain, older version can be accessed as well
  • The content components are quite basic, but a mixture of content components, stage, sidebar, footer teasers, youtube video, image
  • The schedule is a smart feature building automatically based on talk pages with metadata and duration, and additional non-talk schedule elements (e.g. breaks), automatically calculating the schedule and timings. partial parallel tracks are supported as well e.g. https://adapt.to/2020/en/schedule.html#day-1, https://adapt.to/2020/en/schedule.html#day-3
  • The speakers are maintained separately in a "speaker database" (currently content fragments) with reference to speaker image, bio and some metadata. A versioning feature is implemented because the speaker metadata may change over time (esp. company employment/association), but we keep one speaker object which displays different data in different yearly editions of the conference (but all related talks).
  • The speaker listing is build automatically based on the speaker assignments in all talks in the current year. The speaker detail page lists all talks of the current year, but also links to all talks from other years.
  • The talk archive lists all talks from all years an can be filtered by tags, year, speaker. Additional a full text search is supported, which currently searches the metadata of all talks (including the talk description), but not the binary files (presentations).
  • Image gallery automatically lists all images in an asset folder, they are not uploaded/maintained individually in the gallery page
  • Different views of the schedule e.g. for live agenda monitors during the conference
  • Usercentrics Consent Management with automatic blocking of features if not consent is given (youtube, newsletter, ticketshop, google maps)

Editorial Use Cases

• We have two different target audiences for maintaining the websites: Marketing Users and Schedule/Talk Maintainers
• The marketing users mainly write texts, upload content images, create new pages etc., the usual business user CMS stuff
• The Schedule/Talk maintainers work more with well-structured data: speaker, talks, schedule planning, associations to presentations, code downloads, talk videos. But the talk details and speaker bios also include formatted text.

Architectural Aspects

• We have different "media formats" (aspect ratios, resolutions) on the website. The original images may have to be cropped, although currently they are usually pre-cropped, but it would be nice to do this in the application (as it is possible with AEM).
• The speaker data and speaker portraits is maintained independently of the yearly editions, the other data and assets are usually directly associated with a yearly edition of the websites.
• Main navigation and footer navigation is built automatically. Part of the navigation is a list of all yearly editions which is also built automatically.
• Context-aware configuration is used to maintain some configurations across all years, some per yearly edition
• There is currently no "user-generated content", although it would be nice to build on an architecture that allows this in the future. Example use cases: integrate feedback/evaluation sheet in website, talk ratings in talk details etc. - ensure that each user can do this only once with an anonymous key or other super-lightweight feature
• Talk archive looks up all talk data from all yearly editions. The tag/year/speaker filtering is done on the client-side, but the full text search is executed on the server site (currently in with oak with JCR full text search)
• It would be nice to re-use the existing URL structure which is already quite short and useful (.html could be left out, with a redirect if its still used)
• We have a set of short redirect rules currently maintained in dispatcher config like https://adapt.to/cfp, https://adapt.to/tickets pointing to the current year, and some for previous years e.g. https://adapt.to/2019/call-for-papers
• The existing website does not implement SEO best practices (except using nice URLs and semantic markup)
• The existing website has a very old FE implementation. We are planning to replacing it anyways, probably with a new Vue-based FE.