The Integration components/products are switching to PatternFly 4 (PF4) in a synchronized way.

1. Milestone 1: Masthead consistency (shipped in 2.5)
2. Milestone 2: Masthead, Nav, About; Login (openshift) (targeted for 2.6)

At this moment admin portal is just using styles to replicate PF3 and PF4 look & feel. That's not scalable and we want to truly move to PF4, using PF4 components.

Conversation topics:

IE11

IE11 support will be deprecated in the 2.5 onprem release, meaning theoretically we will stop supporting it in 2.6 (summer 2019). But ideally work on PF4 should be merged into master and deployed on saas long before that date, meaning that we can not just give up on IE11 support for the time being.

• long living pull requests that don't support IE11 (less work, more merging trouble in the end)
• short lived pull requests that do support IE11 (more work, less merging trouble)

Progressive enhancement.

It's highly desirable to not depend on javascript for each page load as this will dramatically slow down our test suite and make the app less accessible.

• inject js after rendering pure html/css
• server side rendering
• ignore and proceed to depend 100% on js