Slashdot Mirror
Ask Slashdot: Should Open-Source Developer Teams Hire Professional UI/UX Designers?
OpenSourceAllTheWay writes: There are many fantastic open-source tools out there for everything from scanning documents to making interactive music to creating 3D assets for games. Many of these tools have an Achilles heel though -- while the code quality is great and the tool is fully functional, the user interface (UI) and user experience (UX) are typically significantly inferior to what you get in competing commercial tools. In an nutshell, with open source, the code is great, the tool is free, there is no DRM/activation/telemetry bullshit involved in using the tool, but you very often get a weak UI/UX with the tool that -- unfortunately -- ultimately makes the tool far less of a joy to use daily than should be the case. A prime example would be the FOSS 3D tool Blender, which is great technically, but ultimately flops on its face because of a poorly designed UI that is a decade behind commercial 3D software. So here is the question: should open-source developer teams for larger FOSS projects include a professional UI/UX designer who does the UI for the project? There are many FOSS tools that would greatly benefit from a UI re-designed by a professional UI/UX designer.
The answer is "yes". Open-Source developer teams should hire professional UI/UX designers.
Next question, please.
Just cruising through this digital world at 33 1/3 rpm...
Yes, that's the better question to ask. I think it's simply a matter of exposure. Designers just aren't as aware of the potential of open source to show off their creativity. Some sort of outreach may be in order...
Unnecessary. See work for hire. If you paid to have it created, then, unless there is a contractual term that says otherwise, you own the copyright. So, all you have to do is, make sure you don't give away more than you should to your employee.
Lots of people are paid to work on open source projects
When people are paid to work on open source, they are usually given a specific checklist of things to work on, and are paid by people that have already figured out the UI.
Also, plenty of "UI designers" aren't actually very good at designing UIs. You aren't going to get a world class designer from a Craigslist ad.
Here's a better suggestion: Read some books on UI design. A very good one is Don't Make Me Think.
The Design of Everyday Things is not about software, but is still a good book that every engineer should read.
You should always do "Hallway Usability Tests", as well as the other 12 things on this list.
Is it because people right here in this discusison keep conflating UI/UX design with some sort of frilly artistic bullshit?
It's design of a user interface. Not it's artistic merit. Not what it looks like. That is widget design, which comes after UI/UX design.
The User Interface is the buttons, hierarchy of menu, graphical cues, etc. It can all be written as a guideline, which it actually for the most part has been designed and just needs consistent implementation. Everybody has their boutique obsessions, the CUA just needs a little updating.
Good advice. Now for a trip down memory lane Google, "KDE, GNOME, Spatial Browser, debate, UI"*. You'll get an eyeful of why FOSS isn't amicable to UX/UI. I doubt little has changed.
*Expand a little to account for, "controls for everything" and "absolutely needed".
BTW UX vs UI: different things.
https://www.ready4s.com/blog/d...
Shai Schticks:"You don't make peace with friends, you make peace with enemies"
I have a damn good answer for this. Because UX/UI cannot be handled by unit tests, and does not have a definite "better" or "worse" condition at face value. I cannot count the number of times I've proposed pull requests for UX issues in open-source projects, only to have them either massively argue about the need, or flat out reject the PR. One of the more recent ones, a router OS which has basic IPv6 support was displaying IPv6 delegated subnets wrong. As an end-user, this confused me and went against my knowledge of being a network admin. As a software engineer, I went in and check the UI code which displayed the info. It was a trivial bug to fix, but instead of fixing the bug, the project lead decided to take the entire feature out as a way to "fix" it. So, instead of display correct information or even wrong information, just display NO information at all to the end-user!
The two key areas where developers need to learn to open up more to contributions is for UX/UI, and for documentation. Recently I discovered that there is a dedicated group of technical writers in Seattle who are the equiv open-source software engineers, they give some of their spare time to help document open-source projects. They've been really nice, host workshops, and teach their trade openly. *THIS* is what we need in open-source, more disciplines besides just software engineers.
"The User Interface is the buttons, hierarchy of menu, graphical cues, etc. It can all be written as a guideline,"
Umm, no. That is like saying checkstyle can design your system for you. Styleguides are good, both for UIs and code, but they cannot replace design, and that goes both for up-front and incremental design.
Anything but the most trivial user interface needs a coherent vision of how the interaction is meant to flow in order to end up any good.
Many have none, and end up terrible to use as a result. Some get by by copying something that *was* designed competently, but even then they often get the little (or big) things wrong because they have not understood the reasons behind the model design.
Reliably creating and maintaining a good interaction model requires some knowledge of how people use systems, and at least a minimum of user testing. None of this is impossible for an interested developer to learn, but it does require some investment.
sudo ergo sum