3 ms·

We should dial it back to the last known good state and build from there: bash and vim, praised be thy names. The supposition that modern UIs are based on is t

We should dial it back to the last known good state and build from there: bash and vim, praised be thy names. The supposition that modern UIs are based on is that they should be drop dead simple for a brand new user to pick up and use without training, but at the expense of flexibility and utility. You can't overwhelm them with options, even under menus, and you need to present large buttons for them to immediately identify and mash. Basically, Fischer-Price toys. However, you're only briefly a new user. You may be using a tool every day for months or years, so the obvious design philosophy should be focusing on efficiency (i.e. keyboard control, snappy response, advanced features) for the long term user.

fpig 6y ago
This is totally true (other than the bash/vim part, there certainly is value in having a graphical UI). My conclusion is that these inefficient designs aimed purely at someone who is seeing the app for the first time are a result of two things: 1. Designers / UX "experts" generally have no clue what they're doing. I don't know why, but every single designer I've met has only designed trivial software. I've never met anyone who's designed complex systems, like MS Word, 3DS Max, etc, stuff that people use for hours every day to perform complex work. At the same time, half the developers I know have worked on very complex software at some point in their careers, so it's not like complex software does not get built at all. Does anyone know a single UI designer who has designed an app of that level of complexity? It's really strange that they literally don't seem to exist. So they've never worked on an app that users use extensively, daily, for years, to perform complex tasks, and their whole mindset is about "how will someone seeing it for the first time use it". So when you do bring UI experts on such an app, they have no clue what they're doing, and users complain about their design decisions. And people who get hired to design things are people who've designed e-commerce web sites and mobile apps in the past, because those seem to be the only people that exist. 2. If it's an app that will be demoed in person or demos will be provided as a part of marketing, anyone involved with sales in any way will push hard for this because it demos well. They're not demoing to the guy who's been using the app for 4 hours a day for 5 years, they're demoing to the guy who is going to be using the app, or not even to that guy, but to his boss who isn't even going to be using it. I regularly get requests like "can we make this automatic so the user can do it with 1 click" when it's an action the user will very rarely perform so it doesn't matter whether it takes 4 or 2 seconds, and it provides more flexibility to the user to do it in 2 clicks instead of the app making an assumption that joins 2 actions into one. The reason? They will do it in the demo and they can say "look you can do this with 1 click" even though from the perspective of actually using the application it's not helpful. I am almost ashamed that I worked on an app where you upload stuff by having an "Upload" button on the main page, that you press, and then it guides you through the process of selecting where to upload and what to upload, and then it uploads. Think about how inefficient and cumbersome this is. I guess the point is that if you've never seen the app before, and want to upload, you see "UPLOAD" staring at you, so you press it and it guides you through the rest, there is basically no way to get confused. In fact every action in this app works through a "wizard" of some kind, there is no concept of freely navigating and contextually performing actions on objects as you navigate. If you want to perform an action, you start by pressing a button with the name of that action, then get guided through the rest over multiple pages. It's the literal opposite of how I would build an app if I was building it for myself.
chrismcb 6y ago
Vim is a far cry from being drop dead simple.
- systemvoltage 6y ago
  Yeah, vim is the exact opposite of user-friendly-app [1] but it pays off in the long term with efficiency dividends. [1] http://www.terminally-incoherent.com/blog/wp-content/uploads/2006/08/curves.jpg http://www.terminally-incoherent.com/blog/wp-content/uploads...
- hodgesrm 6y ago
  Right. The editor where you learn new commands by making typos. :)
HeavyStorm 6y ago
> The supposition that modern UIs are based on is that they should be drop dead simple for a brand new user to pick up and use without training, but at the expense of flexibility and utility. Spend almost a decade building enterprise software and heard EXACTLY that a stressful number of times. On one hand, I think that maybe my customers (mostly Fortune 500) couldn't hire anyone with at least 90 IQ, because, shit, it had to be simple! And also, probably, those companies provided the worse customer service ever, since they wanted to have people operating their software with zero training. Now, besides making the interface dumb, less productive, we had the problem that people were much more concerned with making things easier to use (and pretty) and not productive. (and don't you think that they'd ever talk to the actual user...)

fpig 6y ago

This is totally true (other than the bash/vim part, there certainly is value in having a graphical UI). My conclusion is that these inefficient designs aimed purely at someone who is seeing the app for the first time are a result of two things: 1. Designers / UX "experts" generally have no clue what they're doing. I don't know why, but every single designer I've met has only designed trivial software. I've never met anyone who's designed complex systems, like MS Word, 3DS Max, etc, stuff that people use for hours every day to perform complex work. At the same time, half the developers I know have worked on very complex software at some point in their careers, so it's not like complex software does not get built at all. Does anyone know a single UI designer who has designed an app of that level of complexity? It's really strange that they literally don't seem to exist. So they've never worked on an app that users use extensively, daily, for years, to perform complex tasks, and their whole mindset is about "how will someone seeing it for the first time use it". So when you do bring UI experts on such an app, they have no clue what they're doing, and users complain about their design decisions. And people who get hired to design things are people who've designed e-commerce web sites and mobile apps in the past, because those seem to be the only people that exist. 2. If it's an app that will be demoed in person or demos will be provided as a part of marketing, anyone involved with sales in any way will push hard for this because it demos well. They're not demoing to the guy who's been using the app for 4 hours a day for 5 years, they're demoing to the guy who is going to be using the app, or not even to that guy, but to his boss who isn't even going to be using it. I regularly get requests like "can we make this automatic so the user can do it with 1 click" when it's an action the user will very rarely perform so it doesn't matter whether it takes 4 or 2 seconds, and it provides more flexibility to the user to do it in 2 clicks instead of the app making an assumption that joins 2 actions into one. The reason? They will do it in the demo and they can say "look you can do this with 1 click" even though from the perspective of actually using the application it's not helpful. I am almost ashamed that I worked on an app where you upload stuff by having an "Upload" button on the main page, that you press, and then it guides you through the process of selecting where to upload and what to upload, and then it uploads. Think about how inefficient and cumbersome this is. I guess the point is that if you've never seen the app before, and want to upload, you see "UPLOAD" staring at you, so you press it and it guides you through the rest, there is basically no way to get confused. In fact every action in this app works through a "wizard" of some kind, there is no concept of freely navigating and contextually performing actions on objects as you navigate. If you want to perform an action, you start by pressing a button with the name of that action, then get guided through the rest over multiple pages. It's the literal opposite of how I would build an app if I was building it for myself.

chrismcb 6y ago

Vim is a far cry from being drop dead simple.

systemvoltage 6y ago
Yeah, vim is the exact opposite of user-friendly-app [1] but it pays off in the long term with efficiency dividends. [1] http://www.terminally-incoherent.com/blog/wp-content/uploads/2006/08/curves.jpg http://www.terminally-incoherent.com/blog/wp-content/uploads...
hodgesrm 6y ago
Right. The editor where you learn new commands by making typos. :)

HeavyStorm 6y ago

> The supposition that modern UIs are based on is that they should be drop dead simple for a brand new user to pick up and use without training, but at the expense of flexibility and utility. Spend almost a decade building enterprise software and heard EXACTLY that a stressful number of times. On one hand, I think that maybe my customers (mostly Fortune 500) couldn't hire anyone with at least 90 IQ, because, shit, it had to be simple! And also, probably, those companies provided the worse customer service ever, since they wanted to have people operating their software with zero training. Now, besides making the interface dumb, less productive, we had the problem that people were much more concerned with making things easier to use (and pretty) and not productive. (and don't you think that they'd ever talk to the actual user...)