[tiborsaas]

I don't accept that this is binary choice. The free software movement has failed to attract designers and UX minded people. As dTal said in a sibling comment:

"There's a cultural issue in free software wherein good UI is rejected as being for sissies and not worth the effort"

So there's the answer. Once the movement (whatever that is) becomes more open to accept that good software doesn't stop at being free then it might have a change to take on the world.

[salawat]

The problem there is, what happens when you get more than one chef in the kitchen?

If you want to see what I mean, checkout the Desktop Environment scene. There's so many different views on what is good UX, that if you have 1000 different designers trying to make a UI to the same program, you end up with 1000 different interpretations of what this program should "look like".

Free Software doesn't end up looking the way it does because anyone wants it to look bad; it's the way it is because that's all anyone in particular needs it to be.

I don't need a GUI for diff and patch. Some people prefer one, and that's fine. Those people can go ahead and get any of the many implementations of a visual diff/patch program. That is their choice, just as it is mine to just use the command line utility as an example.

UI/UX as it is taught in the industry comes hand-in-hand with curation, which implies a level of control over the end-user's choices in user experience, dependency graph, etc. Software freedom accepts that the ugly choice is still a choice. And maintenance-wise, that choice (the uglier, low dependency one) is a heck of a sensible default from the distribution maintainer's point of view. Same functional capability, and a heck of a lot less cruft to work with.

Besides which, in my experience, most UX folks I meet are cripplingly dependent on a single technology stack I personally have no desire to see everything reimplemented in, which is JavaScript/HTML/CSS.

I'd suffer through learning something like TCL/TK before resorting to adopting that nightmare of a toolchain. I'm what you might call a curmudgeonly old man in taste however, so that comes with the territory.

There is also the resource intensive vs. pretty trade-off to be made as well. Do I want that sexy Aero/Peek, pseudo 3D, everything has an animation look sucking up my potentially valuable and short supply CPU cycles? Or is minimalist X with a minimal window manager without widgets or gadgets or whatever they call all those extra doodads these days good enough?

Software tradeoffs are unfortunately an issue of always on engineering all the time. Something a lot of people don't necessarily find compatible with their tastes. It's certainly a good thing more people are talking about it though. Will have to keep my eye out for new distro's if the more UX savvy take a bite into making something.

[xvector]

The point is that no one will outside of an inconsequentially small group will adopt user-facing free software if it isn't easy to use and if it doesn't have a slick and intuitive UI.

Programmers perhaps do not care for this. But everyone else does. A good, competitive UI is non-optional.

> [Free Software looks bad] because that's all anyone in particular needs it to be.

This is the fundamental disconnect! It's all programmers need it to be. Regular, normal people? They also need it to be fluid, pretty, and intuitive.

Yeah, that's a need, not a want, as much as we might wish it otherwise. We might wish that people had the strength of will/whatever to stick with comparatively worse UIs in exchange for freedom, but we know that's not true, and that won't happen.

> UI/UX as it is taught in the industry comes hand-in-hand with curation [...] Software freedom accepts that the ugly choice is still a choice.

These are not mutually exclusive. UI/UX in free software can come by default as a highly refined and curated experience, with an option to switch things out if you want.

---

I agree that there is a fundamental tradeoff in time and resources between implementing "real features" and UI/UX. Even big corporations with billions of dollars to burn struggle with this. But what we need to realize is that UI/UX is a real feature.

If our priority is getting more people to adopt OSS, then it needs to be done. If our priority is getting shit done in the short term, well - I guess you can hold off on it.

[tiborsaas]

The way I see it is that free software is not designed to have these kinds of processes that makes commercial grade software good and widespread.

Of course, UI/UX is hard problem so does writing hardware drivers or optimise kernel code. Even if you use a command line tool, like GIT it has a good UX to it.

    git: 'deff' is not a git command. See 'git --help'.

    The most similar command is
 diff

I really like this and it shows that free software developers care about UX as long as it's in their domain and expertise they can execute.

> I don't need a GUI for diff and patch. Some people prefer one, and that's fine.

You have a good attitude toward this, but then there's people like RMS who flat out wants to ban JavaScript in the browser because "proprietary software". That's not how you succeed with a mission like this. (but I also don't think the state of open source software is too bad either)

> most UX folks I meet are cripplingly dependent on a single technology stack

Then don't ask them to implement the UI part. It's also fine to ask for a prototype in JS/CSS that you can reimplement with the best tools you see fit.

> minimalist X with a minimal window manager without widgets or gadgets or whatever they call all those extra doodads these days good enough?

A good minimalist design is really hard to pull off, so it's good enough, but as hard as making something really shiny.

Tradeoffs are really there so aligning them with your goals is essential.

[salawat]

I'm also a big fan of Git's UX in that regard. Solid point.

W.R.T. Stallman, I sort of agree with him. JavaScript as a portal for code getting on my machine that is ruthlessly obfuscated so I cannot efficiently determine what it is doing is a danger that I personally have significant qualms with wanting to support. I don't mind dynamic webpages, and the functionality it enables, but I absolutely abhor some use cases it enables; namely browser fingerprinting, runtime code encryption/decryption of payloads within the code, and other deceptive practices like running a cryptocoin miner or exfiltrating information I'm not okay with a la exploits like Spectre and Meltdown.

If people were decent and considerate with their JavaScript I'd be more amenable to it. However, the blatant abuse of the technology I've seen perpetrated by the industry just turns me off of it altogether.

>Then don't ask them to implement the UI part. It's also fine to ask for a prototype in JS/CSS that you can reimplement with the best tools you see fit.

Fair.

>A good minimalist design is really hard to pull off, so it's good enough, but as hard as making something really shiny.

Also fair. Do you happen to be a UX person? You sound like a blast to work with.

[tiborsaas]

It's all valid points, but I personally don't care, I rarely see a page that obnoxious because of its JS use. Most sites I find repulsive is their use of ads. Now if there were no JS in browsers, these site would still plant heaps of ads only with CSS, it's not the code execution to blame. If the publisher controls the layout then it's game over for freedom :)

> Do you happen to be a UX person?

Not a label I'd put in my title, but I like UX, it's an integral part of building good products so I spent fair bit of time exercising it. I'm a full stack developer and who likes to build experiences users love.