Hi, from the UX discussion that we had in AMS I got a feeling that we simply don’t know who our users are. Would anyone by interested in helping with Ethereum user survey?
3 Likes
Here is a summary of the session that I wrote from the notes on the whiteboard…
UX session notes - OG council, Amsterdam 2022-04-24
The session discussed UX in general. There is a “whiteboard note” that tried to capture some of the discussion, and I am attempting to decipher that here.
Questions / Topics:
What do we mean?
What we mean is trivially described as “making things simple”. Key questions we noted include:
- Do users achieve what they want?
- Do users achieve what we want them to achieve?
- Are people spending time doing tasks that could be automated away?
- Are people comfortable working with the system? More than with other alternatives?
What are the problems?
We identified a few barriers to usability:
- Fear…
- …of the consequences of a mistake. Web3 systems are unforgiving compared to e.g. bank accounts or credit cards
- …of being embarrassed
- …of the consequences of a mistake. Web3 systems are unforgiving compared to e.g. bank accounts or credit cards
- Language and accessibility issues
- Many systems are command line based, and often error messages are generic rather than informative, or are excessively “cryptic”.
- The other major interface pattern is using Web apps. Here there is available knowledge and guidance on how to improve: WCAG, W3C’s i18n work, …
- Many systems are command line based, and often error messages are generic rather than informative, or are excessively “cryptic”.
- Technical issues:
- Performance is unpredictable, but also there is rarely feedback on how long things take, how they are progressing…
- Errors are obscure (see above: accessibility and language issues)
- Performance is unpredictable, but also there is rarely feedback on how long things take, how they are progressing…
Better UX examples
- Uniswap
- Grid+
- Pryzm
- POAP
- Dappnode
- Wallets (but with the caveat that they aren’t generally great UX)
- “point and click” interfaces
- These can be built with Web technology for which there is significant work avaialable on how to provide better user experience
- but there is a concern that these can hide problems, or make security properties hard to discover.
Worse UX examples
- Ethereum clients
- Essentially Eth1 / Execution clients
- Command Line is not how people usually operate software in this century
- Essentially Eth1 / Execution clients
- Being surprised at fees for multiple transactions
- Trying to run a node, and interact with it from other applications
Why UX matters
- There are alternative platforms. If they are easier to work with, they will attract capital, projects, mindshare, and developers that would otherwise consider the Ethereum ecosystem as a natural home.
- Provides confidence in the ecosystem, which in turn encourages investment and the creation of a larger community of developers in a “virtuous circle” - a feedback loop that encourages growth of the ecosystem.
- It’s necessary to achieve the often-claimed goals of enabling decentralisation and improving democratisation
- It can help people understand security better, and so make fewer mistakes.
So, what next?
- We encourage people to publish the results of research into usability in our space.
- It would be nice to see simplified interfaces that provide all the information a user needs to complete an action (i.e. they won’t be surprised halfway through), and does not require them to manually do things that are unnecessary.
- It would be nice to have more “easy entry simulations” - e.g. going through the process of a transaction, or of adding a contract to a network, without incurring the costs of it happening “for real”
…
2 Likes
Thank you for summarizing it. I think that anyone interested in Ethereum’s UX will find these notes interesting and helpful.
Regarding the user survey (or maybe better “community survey”), I’m happy to announce that a team has gathered and will soon start working on making it happen. Of course, anyone interested in helping can still join us.
1 Like
Excellent! How do people join? What do you have as starting material, plan, …? (I’m looking for stuff written that we can share - beyond even this forum, e.g. in translation)
Right now people join by giving me their e-mail address 
But we will for sure setup some kind of IM channel/forum/mailing list. We have some draft documents: rules (how to ask the questions, what not to ask, privacy, etc.), question ideas and a list of organizations that could have done UX research in the past (we want to contact them). I imagine that when these drafts are finalized and become internal set of guidelines for the team, we make them all public.