Hacker News Reader

Comment by DavidSJ | Hacker News Reader

My longstanding explanation for this has been as follows:

* Improvements to open-source projects come primarily from developers scratching a personal itch.

* User interfaces are first and foremost a way for the application to communicate to the user what is possible and how to achieve it.

* If there's a problem with a user interface, it therefore most adversely impacts a user who does not yet know what is possible or how to achieve it.

* To fix a problem with the user interface, you must learn enough about the application such that you now know what is possible and how to achieve it. Hence the problem no longer affects you as much and is no longer a personal itch. So the problem doesn't get fixed.

View on HN

I agree with the above, and also would add the following:

* There is usually no clear way for a UX designer to contribute. Note with this example, under 'Contributing - Tasks I could work on' it doesn't mention UX, and the only way to become a committer is to "show us some of your patches, or solve some issues" - none of which involve UX.

* There is a slight obsession in the open source community to use only using open source tools, while the majority of modern UX practice is developed on proprietary tools (e.g. Figma).

* Design by committee is the most frustrating and slow process, and great UX design comes from strong decisions by great designers rather than community votes.

* There is an underlying assumption that if you contribute UX designs and put them in the Wiki they will be ignored, so there is no point. It's not the same as code where there is a one step process (i.e. commit), the whole team needs to embrace the UX and then work to commit it to code.

> slight obsession in the open source community to use only using open source tools

While true in some aspects, it's not true in others. Most of todays "modern" open source ecosystem live within GitHub which is not open source, using services for development and infrastructure that is also not open source. While many developers local tools tend to be open source, the rest of the infrastructure we use tend to not be open source. Even many of the biggest package registries for open source are not open source themselves.

I'm not saying 100% of the open source community leverages these services, but a lot sure do.

I agree with your top most point the most though, there is no clear path for designers to actually contribute. And if there is, it's a pipeline of "come up with requires -> design -> code", whereas contributors are not as excited about the last point if someone else did the steps before. Hence we see more code contributions than design ones.

> While true in some aspects, it's not true in others.

Open standards might have been a better term. The example in this project would be the use of "Pencil Project" for UX design rather than Figma / Sketch / Adobe XD (which to some extent may be for licencing and cost reasons too).

Funny enough, Pencil Project in itself is a testament to how badly open source software needs better/more designers who know good UX/UI. Try putting a modern designers coming from Figma/Sketch in front of Pencil Project and watch them puke colorless rainbows while trying to figure out how to do the most basic of operations.

Figma, because of the cost (or lack off) and availability, seems to be the only tool realistically available for doing open source design today. Unfortunately neither the format nor the editor is open source itself, but sometimes the means justify the end I guess.

Figma is able to import and export svg files pretty well (for my purposes) - I wonder if that's sufficient for a good number of workflows/projects?

I doubt the click-through presentation stuff can be exported in any way, but basic design work could be persisted and collaborated on by multiple people with different tools if that approach was used.

Yeah, SVG import/export for smaller elements works fine but once you get into larger files, design systems or designing entire applications in Figma, import/export of SVGs becomes untenable. Otherwise you just have the .fig file format, which is also proprietary as well.

Currently for my open source project, we all just work directly in Figma then export the design to a website via the tool we're building. Works fine for now, but not doing anything too complex. Project is https://github.com/instantwebsite in case you're curious.

> * There is usually no clear way for a UX designer to contribute. Note with this example, under 'Contributing - Tasks I could work on' it doesn't mention UX, and the only way to become a committer is to "show us some of your patches, or solve some issues" - none of which involve UX.

This affects not only UX. I'm sure I'm not the only one who has spare infrastructure and ops skills that I'd love to contribute to open-source projects - hosting CI/CD, repo mirrors, staging environment, website, chat server, SMTP gateway, for example. It'd be nice to play a small part in decentralization, for one thing.

I'm sure I'm far from the only one. I'm also sure there are open source projects who would benefit from it.

There's no process for us to match and it seems just culturally abnormal. I'd love to hear ideas.

I like this list :)

I'd add to it that developers are the ones birthing these projects, and they (or we, I should say) don't like to be bottle necked by UX.

Thank you, and I agree with your point.

My comment is intended more as a step-by-step argument rather than a list.

Reply on news.ycombinator.com