What Would You See Changed in Haskell?
This report condenses the free form answers to Question 77 of the State of Haskell Survey 2025 . Due to the phrasing of the question, we have received 553 answers, with varying degrees of precision, actionability, and relevance. A team of two people (Sawa & Hécate) proceeded to quantitatively and qualitatively analyse the survey results, extracting trends and topics about what the community would like to see changed.
A qualitative analysis of survey answers entails the hard task of noticing trends and moods, as well as determining the relevance of answers. We needed to understand if an answer is an example of the XY Problem , and if so establish what the deeper problem is. Some answers were simply too vague to be classified either as a trending mood, or an actionable suggestion. Finally, one important thing to do when performing such an analysis is to have empathy for the people who spent their time responding to the survey and not outright dismiss their answers.
The following topics are the ones that have collected the most answers from the survey participants. They are presented in no particular order.
Switching APIs to using Text instead of String
A laser-focused suggestion that came up is the replacement of String by Text as the canonical type for textual data. In particular, one answer explains the reasoning:
The text library must live from base , because none of the APIs of base can use Text , and when people look at base for examples of well-written Haskell code, they end up using String , which is not the right choice.