Simon Willison's Weblog: ux

Quoting Jared Palmer

2025-02-07T06:39:38+00:00

Confession: we've been hiding parts of v0's responses from users since September. Since the launch of DeepSeek's web experience and its positive reception, we realize now that was a mistake. From now on, we're also showing v0's full output in every response. This is a much better UX because it feels faster and it teaches end users how to prompt more effectively.

— Jared Palmer, VP of AI at Vercel

Tags: ux, ai, prompt-engineering, generative-ai, llms, vercel, deepseek, ai-in-china, prompt-to-app

Ambsheets: Spreadsheets for exploring scenarios

2025-02-05T02:50:22+00:00

Ambsheets: Spreadsheets for exploring scenarios

Delightful UI experiment by Alex Warth and Geoffrey Litt at Ink & Switch, exploring the idea of a spreadsheet with cells that can handle multiple values at once, which they call "amb" (for "ambiguous") values. A single sheet can then be used to model multiple scenarios.

Here the cell for "Car" contains {500, 1200} and the cell for "Apartment" contains {2800, 3700, 5500}, resulting in a "Total" cell with six different values. Hovering over a calculated highlights its source values and a side panel shows a table of calculated results against those different combinations.

Always interesting to see neat ideas like this presented on top of UIs that haven't had a significant upgrade in a very long time.

Via Hacker News

Tags: spreadsheets, ui, ux, geoffrey-litt, ink-and-switch

Mapping the landscape of gen-AI product user experience

2024-07-20T04:40:42+00:00

Mapping the landscape of gen-AI product user experience

Matt Webb attempts to map out the different user experience approaches to building on top of generative AI. I like the way he categorizes these potential experiences:

Tools. Users control AI to generate something.

Copilots. The AI works alongside the user in an app in multiple ways.

Agents. The AI has some autonomy over how it approaches a task.

Chat. The user talks to the AI as a peer in real-time.

Tags: matt-webb, ux, ai, generative-ai, llms, ai-agents

ChatGPT in "4o" mode is not running the new features yet

2024-05-15T18:25:07+00:00

Monday's OpenAI announcement of their new GPT-4o model included some intriguing new features:

Creepily good improvements to the ability to both understand and produce voice (Sam Altman simply tweeted "her"), and to be interrupted mid-sentence
New image output capabilities that appear to leave existing models like DALL-E 3 in the dust - take a look at the examples, they seem to have solved consistent character representation AND reliable text output!

They also made the new 4o model available to paying ChatGPT Plus users, on the web and in their apps.

But, crucially, those big new features were not part of that release.

Update 10th December 2024: ChatGPT Advanced Voice Mode has now been available in the mobile apps (and desktop app) for a few months, but advanced image output mode still isn't available yet.

Here's the relevant section from the announcement post:

We recognize that GPT-4o’s audio modalities present a variety of novel risks. Today we are publicly releasing text and image inputs and text outputs. Over the upcoming weeks and months, we’ll be working on the technical infrastructure, usability via post-training, and safety necessary to release the other modalities.

This is catching out a lot of people. The ChatGPT iPhone app already has image output, and it already has a voice mode. These worked with the previous GPT-4 mode and they still work with the new GPT-4o mode... but they are not using the new model's capabilities.

Lots of people are discovering the voice mode for the first time - it's the headphone icon in the bottom right of the interface.

They try it and it's impressive (it was impressive before) but it's nothing like as good as the voice mode in Monday's demos.

Honestly, it's not at all surprising that people are confused. They're seeing the "4o" option and, understandably, are assuming that this is the set of features that were announced earlier this week.

Most people don't distinguish models from features

Think about what you need to know in order to understand what's going on here:

GPT-4o is a brand new multi-modal Large Language Model. It can handle text, image and audio input and produce text, image and audio output.

But... the version of GPT-4o that has been made available so far - both via the API and via the OpenAI apps - is only able to handle text and image input and produce text output. The other features are not yet available outside of OpenAI (and a select group of partners).

And yet in the apps it can still handle audio input and output and generate images. That's because the app version of the model is wrapped with additional tools.

The audio input is handled by a separate model called Whisper, which converts speech to text. That text is then fed into the LLM, which generates a text response.

The response is passed to OpenAI's boringly-named tts-1 (or maybe tts-1-hd) model (described here), which converts that text to speech.

While nowhere near as good as the audio in Monday's demo, tts-1 is still a really impressive model. I've been using it via my ospeak CLI tool since it was released back in November.

As for images? Those are generated using DALL-E 3, through a process where ChatGPT directly prompts that model. I wrote about how that works back in October.

So what's going on with ChatGPT's GPT-4o mode is completely obvious, provided you already understand:

GPT-4 v.s. GPT-4o
Whisper
tts-1
DALL-E 3
Why OpenAI would demonstrate these features and then release a version of the model that doesn't include them

I'm reminded of the kerfluffle back in March when the Google Gemini image creator was found to generate images of Black Nazis. I saw a whole bunch of people refer to that in conversations about the Google Gemini Pro 1.5 LLM, released at the same time, despite the quality of that model being entirely unrelated to Google's policy decisions about how one of the interfaces to that model should make use of the image creator tool.

What can we learn from this?

If you're fully immersed in this world, it's easy to lose track of how incredibly complicated these systems have become. The amount you have to know in order to even understand what that "4o" mode in the ChatGPT app does is very easy to underestimate.

Fundamentally these are challenges in user experience design. You can't just write documentation about them, because no-one reads documentation.

A good starting here is to acknowledge the problem. LLM systems are extremely difficult to understand and use. We need to design the tools we build on top of them accordingly.

Update: a UI workaround

On May 16th around 1PM PT OpenAI released a new iPhone app update which adds the following warning message the first time you try to access that headphones icon:

New Voice Mode coming soon

We plan to launch a new Voice Mode with new GPT-4o capabilities in an alpha within ChatGPT Plus in the coming weeks. We'll let you know when you have access.

Tags: text-to-speech, usability, ux, ai, openai, generative-ai, chatgpt, llms

How do you accidentally run for President of Iceland?

2024-04-29T15:31:13+00:00

How do you accidentally run for President of Iceland?

Anna Andersen writes about a spectacular user interface design case-study from this year's Icelandic presidential election.

Running for President requires 1,500 endorsements. This year, those endorsements can be filed online through a government website.

The page for collecting endorsements originally had two sections - one for registering to collect endorsements, and another to submit your endorsement. The login link for the first came higher on the page, and at least 11 people ended up accidentally running for President!

Via Baldur Bjarnason

Tags: usability, ux

Quoting Erika Hall

2024-04-24T00:31:27+00:00

A bad survey won’t tell you it’s bad. It’s actually really hard to find out that a bad survey is bad — or to tell whether you have written a good or bad set of questions. Bad code will have bugs. A bad interface design will fail a usability test. It’s possible to tell whether you are having a bad user interview right away. Feedback from a bad survey can only come in the form of a second source of information contradicting your analysis of the survey results.

Most seductively, surveys yield responses that are easy to count and counting things feels so certain and objective and truthful.

Even if you are counting lies.

— Erika Hall

Tags: surveys, usability, ux

Quoting David Pierce

2024-04-12T12:39:48+00:00

The language issues are indicative of the bigger problem facing the AI Pin, ChatGPT, and frankly, every other AI product out there: you can’t see how it works, so it’s impossible to figure out how to use it. [...] our phones are constant feedback machines — colored buttons telling us what to tap, instant activity every time we touch or pinch or scroll. You can see your options and what happens when you pick one. With AI, you don’t get any of that. Using the AI Pin feels like wishing on a star: you just close your eyes and hope for the best. Most of the time, nothing happens.

— David Pierce

Tags: usability, ux, ai

The Eight Golden Rules of Interface Design

2024-01-09T21:37:20+00:00

The Eight Golden Rules of Interface Design

By HCI researcher Ben Shneiderman. I particularly like number 4, “Design dialogs to yield closure”, which encourages feedback at the completion of a group of actions that “gives users the satisfaction of accomplishment, a sense of relief.”

Via Hacker News

Tags: usability, ux

The 6 Types of Conversations with Generative AI

2023-11-23T17:37:04+00:00

The 6 Types of Conversations with Generative AI

I’ve hoping to see more user research on how users interact with LLMs for a while. Here’s a study from Nielsen Norman Group, who conducted a 2-week diary study involving 18 participants, then interviewed 14 of them.

They identified six categories of conversation, and made some resulting design recommendations.

A key observation is that “search style” queries (just a few keywords) often indicate users who are new to LLMs, and should be identified as a sign that the user needs more inline education on how to best harness the tool.

Suggested follow-up prompts are valuable for most of the types of conversation identified.

Tags: usability, ux, userresearch, ai, generative-ai, llms

Why Chatbots Are Not the Future

2023-05-15T20:54:29+00:00

Why Chatbots Are Not the Future

Amelia Wattenberger makes a convincing argument for why chatbots are a terrible interface for LLMs. “Good tools make it clear how they should be used. And more importantly, how they should not be used.”

Tags: design, ux, ai, generative-ai, llms, amelia-wattenberger

Do you need the feature in Dropbox mobile app that allows using the chosen files in offline mode? Why?

2013-10-29T10:53:00+00:00

My answer to Do you need the feature in Dropbox mobile app that allows using the chosen files in offline mode? Why? on Quora

I use this all the time. It's especially useful for travelling (when you're abroad you often don't have inexpensive cellular data or access to WiFi). I use it for:

Storing PDFs of my flight details and hotel reservations (using OS X's print-to-PDF feature)
Sometimes storing downloaded maps of the local area
Storing details of my travel insurance
Scanned copies of my passport

I usually combine it with Instapaper - I save Wikipedia pages about the places I am visiting offline on my phone with Instapaper so I can refer to them when I'm actually there.

Tags: mobile, quora, dropbox, ux

How can I produce an animated prototype out of designs for an iOS app?

2013-10-13T10:31:00+00:00

My answer to How can I produce an animated prototype out of designs for an iOS app? on Quora

Keynote is a surprisingly good tool for this kind of things, especially since they added path based animations to it a few years ago.

I know a designer who uses PowerPoint to create interactive wireframes.

Tags: iphone, mobile, prototyping, quora, ios, ux

How long should I budget for an experienced designer to design a responsive ecommerce store?

2013-08-13T15:00:00+00:00

My answer to How long should I budget for an experienced designer to design a responsive ecommerce store? on Quora

There's no single answer to this - it depends on the scope of the project. A one-page store selling 3 items is quicker to design than a thousand page store with dozens of category homepages etc.

I suggest putting together a brief and getting estimates from a number of different designers - that should give you a pretty good idea of how long it should take.

Tags: design, startups, ui, web-development, quora, ecommerce, ux, responsivedesign

What are the best web design and web development conferences/ meetups in Central & Eastern Europe (2013)?

2013-05-01T14:51:00+00:00

My answer to What are the best web design and web development conferences/ meetups in Central & Eastern Europe (2013)? on Quora

DjangoCon Europe is in Warsaw this year, on 15th-19th May 2013 - if past experience is anything to go by, that should be an excellent conference (though obviously with a strong bias towards Django).

We have a list of Web Design conferences and events in Central and Eastern Europe on Lanyrd - you might find our full list of Conferences in Central and Eastern Europe useful as well.

Tags: conferences, design, ui, web-development, quora, ux

What would be the best design conference to attend in 2013?

2013-02-07T11:10:00+00:00

My answer to What would be the best design conference to attend in 2013? on Quora

That's a pretty tricky question to answer... there are a lot of excellent UX conferences around (we're listing 69 upcoming UX events on Lanyrd at the moment, and more get added frequently). A few things you should consider:

Where are you based, and how far are you willing to travel?
Are you interested primarily in learning things or in meeting relevant people?
What's your budget?
Are you looking for a single day event or something longer?

In terms of UX there are more academic events (ACM CHI), more commercial events (the An Event Apart series, UX London), or events like IA Summit that bridge the two worlds.

There are also an increasing number of smaller one-two day events dotted all over the map - events like UX Hong Kong 2013, Convey UX in Seattle, UX Munich, UXMad 2013 in Madison, UX Lx in Lisbon...

Sorry this isn't a direct answer to the question, but the truth is that there's a whole bunch of factors in picking the right event for your time and budget.

Tags: conferences, design, ui, quora, ux

What are some good ways to insult an Information Architect?

2012-11-29T15:37:00+00:00

My answer to What are some good ways to insult an Information Architect? on Quora

Ask them if they've started calling themselves a UX designer yet, or if they've jumped straight to Service Design :)

Tags: information-architecture, quora, ux

Are there any good examples of 'large amounts of information being displayed simply and intuitively?'

2012-09-25T18:27:00+00:00

My answer to Are there any good examples of 'large amounts of information being displayed simply and intuitively?' on Quora

Much of Edward Tufte's career has been dedicated to documenting (and inventing) ways of handling these problems - his books are an absolute joy to browse through: http://www.edwardtufte.com/tufte/

I also really enjoy XKCD's attempts at this kind of thing:

Tags: quora, ux

How do you make an existing web application more mobile-friendly without rebuilding it?

2012-09-05T12:22:00+00:00

My answer to How do you make an existing web application more mobile-friendly without rebuilding it? on Quora

Learn about responsive web design. Provided your site is built reasonably well using CSS for layout there is a TON of stuff you can do with CSS media queries to make your site work better on small screen devices. For example, using media queries to detect small screen (mobile) devices you can...

- switch from two columns to a single column layout
- hide irrelevant content entirely
- use shorter labels on your site navigation
- serve up larger, more mobile friendly click regions
- switch to a font family and size that's easier to read on mobile

Here's the original article that coined the term: http://www.alistapart.com/articl... - there are plenty of excellent resources, tutorials and books around these days too if you want to learn more.

Tags: mobile, mobileweb, ui, quora, ux

Which sites do a good job of guiding first time users?

2012-02-23T11:33:00+00:00

My answer to Which sites do a good job of guiding first time users? on Quora

The Twitter sign up process is fascinating, and constantly evolves. I create a new Twitter account every six months or so just to see what their latest iteration looks like.

It's interesting because Twitter is a particularly hard thing to explain to a new user - there are a lot of concepts to get your head around, and the natural reaction is "why is this useful to me". The site just doesn't make sense if you don't start off using it with a good set of relevant followed accounts.

Here's an interesting example flow: if you're signed out of twitter and you visit someone's profile page there, you'll see a strong "sign up to get updates from person-name" call to action. If you start signing up there, Twitter will suggest accounts related to that person as additional things to follow.

Tags: ui, quora, ux

What activities, games or examples have you used to persuade developers that they are different from 'real' users?

2012-02-22T13:52:00+00:00

My answer to What activities, games or examples have you used to persuade developers that they are different from 'real' users? on Quora

I doubt there's anything as effective as getting them to watch a well-run usability test - either a video, a fancy one-way glass setup or just having them quietly observe a zero-budget testing session in a coffee shop.

Tags: ui, usability, quora, ux, userresearch

Why does Facebook Comments Box only support two levels of hierarchy?

2012-02-22T10:12:00+00:00

My answer to Why does Facebook Comments Box only support two levels of hierarchy? on Quora

I've heard in the past that regular (i.e. non-geek) users often have enormous trouble understanding hierarchical comments - they don't understand tree based file systems either.

There's another problem with tree based comments, which is that it's very hard to identify which comments have been most recently added.

Tags: facebook, quora, ux

What is the "best" programming language to learn if you want to mockup your own ideas but don't have a technical background?

2011-02-09T14:18:00+00:00

My answer to What is the "best" programming language to learn if you want to mockup your own ideas but don't have a technical background? on Quora

I knew a very talented UX designer at Yahoo! who did all of his interactive mockups in PowerPoint - including widgets that you click to transition to another "page" in the interface. I've heard of people doing the same thing in Keynote, and OmniGraffle Pro also has tools for creating interactive mockups.

Other than that, you can't go wrong with learning HTML and CSS - it'll pay off many times over, and you can learn just enough jQuery to mock up some interactions.

Tags: information-architecture, quora, ux

What are some good user interface/user experience workshops/conferences coming up this March(p.s. distance isn't a factor, I'm willingly to travel)?

2011-02-08T09:58:00+00:00

My answer to What are some good user interface/user experience workshops/conferences coming up this March(p.s. distance isn't a factor, I'm willingly to travel)? on Quora

We have a list on http://lanyrd.com/topics/user-ex... - the best option in March looks to be Adaptive Path's MX conference: http://lanyrd.com/2011/mx/ / http://mxconference.com/

Tags: conferences, ui, quora, ux

What are the best free resources to begin learning UX design?

2011-01-09T13:56:00+00:00

My answer to What are the best free resources to begin learning UX design? on Quora

We're collecting videos and slides from conference sessions covering user experience on Lanyrd - here's 10 videos and 14 slide decks:

http://lanyrd.com/topics/user-ex...
http://lanyrd.com/topics/user-ex...

Tags: design, ui, quora, ux

What UX/UI conferences in the SF Bay area are worth attending?

2010-11-20T17:18:00+00:00

My answer to What UX/UI conferences in the SF Bay area are worth attending? on Quora

BayCHI is excellent from what I've heard: http://www.baychi.org/

We have a list of upcoming UX events in California here, with an iCal and RSS feed that you can subscribe to: http://lanyrd.com/topics/user-ex...

Tags: conferences, san-francisco, ui, quora, ux, sanfranciscobayarea

To what extent is it still valid to assume that your web app users are stupid?

2010-10-13T12:58:00+00:00

My answer to To what extent is it still valid to assume that your web app users are stupid? on Quora

They're not stupid, but they're probably WAY less web literate than you might expect - unlike you, they haven't spent their entire career learning how the web works. See the famous "What is a browser?" video the Google Chrome team released:

Tags: ui, usability, quora, ux

Why doesn't Facebook use nicer URLs?

2010-10-13T12:50:00+00:00

My answer to Why doesn't Facebook use nicer URLs? on Quora

Just noticed this link: http://www.facebook.com/notes/fa... - so it looks like things are beginning to improve.

Tags: facebook, urls, quora, ux

Dark Patterns: Forced Continuity example, Audible.com

2010-10-12T10:55:00+00:00

Dark Patterns: Forced Continuity example, Audible.com

Dark Patterns are user interfaces that are designed to trick people. I just submitted Audible.com for their habit of signing up users for a $7.49 “gold membership” without making it clear on the checkout screens that this is a recurring monthly charge, not a one-off payment.

Tags: usability, ux, recovered, audible, darkpatterns