How do we enable researchers to do analyses in usable, accessible web interfaces at scale?

The Question

How do we enable researchers to do analyses in usable, accessible web interfaces at scale?

Why It Matters

Computational research increasingly requires serious computing power, but most researchers aren’t systems administrators. The gap between “I have data” and “I have answers” shouldn’t require a computer science degree to cross. Web interfaces can democratize access to computational tools, but only if they actually work at scale.

Current Thinking

The challenge is multi-dimensional:

1. Usability - Interfaces that don’t require training to use
2. Scale - Handling real datasets, not toy examples
3. Accessibility - Working for researchers with varying technical backgrounds
4. Reliability - Long-running analyses that don’t lose work

Galaxy has proven this is possible, but the space for improvement is vast.

Open Subquestions

• How do we balance power-user features with newcomer accessibility?
• What’s the right abstraction level for exposing computational complexity?
• How do we handle the inherent tension between flexibility and simplicity?
• How do we make failure modes understandable to non-technical users?