3 ms·
There are very few web apps where users need - or want - to do any kind of intelligent, complex text editing. Most web apps are virtual paper forms with pre-fi
by TheOtherHobbes 8y ago
There are very few web apps where users need - or want - to do any kind of intelligent, complex text editing.
Most web apps are virtual paper forms with pre-filled menu options and - usually - a bread crumb trail so users can change their choices and/or their minds.
Animations are largely irrelevant to the user experience. The best way to speed up this kind of web app is to iterate over and over to find the sticking points empirically, and then nuke them from orbit.
That means minimising the number of menus, making sure they have useful defaults, and spending extra time on problematic elements like date and time selectors.
The user should type as little as possible and click as little as possible to get the job done. Beyond that, "productivity" does not apply as a concept. Users do not need composability or grep-like text manipulation options.
It's also handy if a page looks good, because this makes an impression on many users and is one factor that can persuade them to come back.
So no - I do not think there's anything of value in the OP, except perhaps as another example of "Developers want to turn everything into a emacs and a build system, because that makes them happy, even though it's 100% inappropriate for non-developers."
- chj 8y agoI think the value of the OP is the proposal that we can have web-like "terminal" that accepts text commands. It's very unlikely that clicking is the best way for human to communicate with machines.
- TeMPOraL 8y agoI disagree on several aspects. I do agree that "most web apps are virtual paper forms with pre-filled menu options (...)", but that doesn't mean they don't need productivity improvement. First of all, many people use those "forms" for work, not just casual browsing. It's one thing when you're putting up some old book you found on eBay, it's another when you have to manage hundreds of such auctions each day. Suddenly, the difference between old and new web design trends is measured in 10x increase of time spent on task. I know because I sometimes help my SO do her work tasks faster, and usually get 10x-100x speedup with clever application of web scraping, regular expressions and batch processing in command line. Another thing - people don't "fill in forms" in isolation. Each webapp being a productivity-hostile silo causes pain whenever a task involves more than one of them. Often, users do in fact need "composability or grep-like text manipulation options", they just don't get them, so they suffer (or delegate to a friendly programmer, who can force a webpage to have some semblance of interoperability, whether its authors want it or not). > So no - I do not think there's anything of value in the OP, except perhaps as another example of "Developers want to turn everything into a emacs and a build system, because that makes them happy, even though it's 100% inappropriate for non-developers." It's not that - though I do believe that Emacs presents a much better UI paradigm than the one accepted in the mainstream (due to flexibility, interoperability and consistency). It's just that present day software - web apps in particular - are ridiculously hostile to productivity. The perfect solution is not doing everything from CLI, but it's also not shiny, animated web UIs exposing as little functionality as possible while still being able to sell the product. Someone upthread mentioned the concept of threshold and ceiling. Emacs and CLI tools are high-threshold, high-ceiling. Webapps and mobile apps are low-threshold, low-ceiling. Perfection would be low-threshold, high-ceiling - that is, something accessible to beginners, but designed so that they can level up and become power users, if they need/want to.