Hacker News Reader

Comment by interactivecode | Hacker News Reader

Projects like these are saying just because we figured out how chairs work, everyone should use the same chair.

It just completely misses the point, people like it when they build something themselves it looks how they like it. Want to spend years working on something that looks exactly like everything else?

View on HN

I agree - this fear of "reinventing the wheel" needs to go away. People gravitate towards products that suit their needs, and if they don't they'll cook something up themselves. It's really not the end of the world if you're writing something that someone else somewhere wrote. Everything has its advantages and drawbacks.

don’t reinvent the wheel is just a clever way of saying don’t innovate

That’s not what this is about at all. This is simply looking around at people using benches, stools, stumps, and piles of rocks and suggesting we come up with a word to describe something constructed that you sit on... then looking at the nicknames currently being used and highlighting the most common. It has nothing to do with prescribing any particular style of chair.

Project like there are a start to cleanup the broken ecosystem.

   we figured out how chairs work, everyone should use the same chair.

Actual message they are are conveying is "We figured out how UI work, everyone who hasn't figured it out come and have a look".

   Want to spend years working on something that looks exactly like everything else?

We already spent 40+ years on UI and why it has become such a mess now? Do you have a solution ?

Has it become a mess though? I agree there are tons of frameworks out there but if you focus only on the main ones there are not more than for the trio of windows/linux/macOS. Heck, on windows only you can use half a dozen frameworks without even journeying to more esoteric ones.

Ultimately there are different frameworks precisely because different people have a different idea about how a UI framework should behave.

To many developers, what the software does is several orders of magnitude more important than how it looks.

There are certainly cases where visual design is much more critical to what you're building, but there are plenty where it's little more than an annoying distraction.

Use standard html then

A select is a select using a native widget

An input is a standard OS input widget

And so on

That does not really work. The whole point of frameworks is to reduce work.

But I can do things faster using MVC on server, ES6 + jQuery on front end, VS having to use something like Angular.

It also looks and feels better.

> But I can do things faster using MVC on server, ES6 + jQuery on front end, VS having to use something like Angular.

That boils down to which technologies you are familiar with.

> It also looks and feels better.

And this is highly subjective too. Of course looks are very independent from system to system but the feels depend on the tech stack quite a bit.

The point of frameworks is to provide a common operating pattern (familiarity). Frameworks often increase work at maintenance time due to being higher level abstractions that dictate conventions and practices that limit expressive independence.

Actually no that was the point of browsers...

Where do you think 'select' came from? Long ago we had many UI select boxes in many native frameworks...and decided that was too much work so we made tags...then browsers...most of the disconnect comes because UI controls have so many usages that ultimately you always want to bust out some code.

If OpenUI wants to enhance the default behavior of select, input, credit card, etc, cool, if they just want to build anothet framework then I'd rather use svelte or angular or vue. The large libraries already standardize, its just a matter of who 'wins' out, and there are already only a few SPA winners.

I guess the OP was answering to

> To many developers, what the software does is several orders of magnitude more important than how it looks.

Native widgets are still the easier way to produce an UI that looks good enough to work with

This link[1] explains how they work on different OS, if they work for desktop apps, why shouldn't they work for web based ones?

[1] https://developer.mozilla.org/en-US/docs/Learn/Forms/Basic_n...

to many customers, there’s a handful of different software choices that more or less do the same thing. so interface and design plays a big role in experience

If it’s really about what the software does, then the conversation would be about exposing proper APIs that let me view the data, query it, schedule operations on it, etc.

But then offering a shinier/slightly more bespoke way of doing that becomes a competitive advantage for your product, and now you get why everyone wants to build their own custom UI.

The reason why the aforelinked to Open UI project will never be the One and Only way to build UIs on the web is because there’d too much economic value in being able to do precisely what it can’t do!

Reply on news.ycombinator.com