overview of problems, chances and practices of integrating UX and Open Source Software development (ideally based on research)

[jdittrich]

Story: As an opensourcedesigner, I want to know which problems and perks designers can expect in FOSS projects and what the strategies are of dealing with it.

Problem: We currently lack such an overview
Possible Solution: Create an overview based on @victoria-bondarchuk Research paper repo: https://github.com/victoria-bondarchuk/List-of-Academic-Research-on-Usability-in-FOSS

[jdittrich]

related:
#69 – cross project communication

[bnvk]

@jdittrich can you elaborate on what sort of overview you are ideally envisioning?

[jdittrich]

Some notes:

• There are incentives to introduce new features to open source software. Basically, there is few punishment for adding some functionality and probably quite some incentives to do so. (Usability in open source software development: Opinions and practice, Andreasen)
• There are no incentives of removing features, on the contrary
  • Power users will be annoyed (someone will have liked the feature)
  • You will get rid of functionality someone else programmed in their free time, they won’t be happy
  • Decision making will be based on consensus (Usability in open source software development: Opinions and practice, Andreasen); there are few usability experts, many non-experts and many opinions about not removing or substantially changing a feature.
• Creating code is seen as good/doing something, discussions, alignment etc. is seen as not doing something/not helpful (Enabling User Centered Design Processes in Open Source Communities,
• Developers are not typical end users (The Usability of Open Source Software, Nicols, Twidale | Enabling User Centered Design Processes in Open Source Communities Mads Bødker)
• Hard to modularize open source software (The Usability of Open Source Software, Nicols, Twidale)

[jdittrich]

@bnvk: Ideally something like an academic review (state-of-the-research) paper, naturally less academic, possibly including personal comments.

[belenbarrospena]

I am planning to take some time off in May to write a PhD proposal on user-centred design in FOSS. As part of that, I will need to write a literature review, and I'll be happy to share it once written.

You will need to wait for a few weeks though :)

[ninavizz]

Academia, over-dependence on lengthy/traditional research, and whitepapers (that few in design read), imho, have so far contributed more noise than signal, to the problems. :(

Simply: Design is a process that has users at the center, and isn't about colors, graphics, or logos. When users can succeed, design succeeds—but most "users" are at the center of commercial experiences, and as a (mostly commercial) community we help one another evolve through conference talks, awards, articles, and word-of-mouth that eventually establishes best practices.

FOSS I've observed, embraces very different values than those of the commercial world—where "contributor" roles are fixed, user success = revenue, and power/hierarchies are celebrated, versus eschewed. It's an epic design challenge in and of itself, to examine how methods that help us succeed in a world with 'x' values, can carry over and help us succeed in a world with 'y' values.

What would be a far more valuable artifact imho would be a playbook, guiding FOSS teams and designers in how to work together, agnostic of specific/chosen tools. Provide both with options, with project-stages/phases, with methods/best-practices, etc., and influence empathy for one another's goals—that in the end, should all point to serving users.

Tangentially, a practice I've observed in the DIY-embracing world of FOSS, is for developers to play the role that corporate projects call Product Management—ideating workflows and features, communicating them in Gimp mockups, and just kinda adding them to the frontend in a roughed-out form... later expecting a designer can smooth-out rough edges, if everyone likes it. That's not at all a user-centric approach, and for many reasons is at the core of FOSS' challenging relationship w/ UX.

We all need to learn how to co-exist as one gooey experiential banana split, not as either the rainbow sprinkles, the crushed cherry topping, the drippy hot fudge, or the scoop of vanilla ice cream. Mmm... iiiice crrreeeeaaam!!!

[ninavizz]

This is a great example of a playbook kinda thing, created by 18F—the US gov's super kickass new(ish) digital services innovation group. It's for designers, by designers—and for designers in government agencies. Super simple, mostly visual, clearly demonstrative #dontmakemethink

https://methods.18f.gov/

[grahamperrin]

The User Organisation: Structure and Governance in an Open Source Project (Jose Christian, 2015) – PhD thesis.

Abstract:

User communities are a valuable source of innovation, where user-specific ideas and prototypes are developed and shared freely. Firms have begun to acknowledge the extent of their value at each stage of the innovation process, often resulting in commercially successful products. The current understanding of user communities, however, relies heavily on the study of software developers working on established open source projects. In recent years, with the emergence of web-based information and communication technology, open source projects and hacker communities have become much more complex, supporting a wider range of contributions from a diverse group of actors. While these technological advances have led to changes in the collaborative innovation, they have not been reflected in the user innovation literature.

Using the CyanogenMod project as a case study, this study explores how a user community manages multiple forms of contributions for the creation of multiple products and services. This study finds that the multiple forms of contributions create a number of communities of practice within the project, each with its own governance mechanism and leadership structure. In addition, the findings show that the activities of the different communities within the project are managed centrally by a separate leadership team who are responsible for setting and implementing long-term plans for the direction of the project as a whole. In light of these findings, this study proposes that the governance and social structures observed in the CyanogenMod project resemble a User Organisation rather than a user community, where the control over the stages of innovation can give users control over the direction and outcomes of the project.

(Jose and I work for the same research centre.)

[belenbarrospena]

@ninavizz I agree with pretty much everything you say, including the idea of creating a repository of methods that have worked for us in the context of FOSS projects.

However, I have to take issue with your statement about academic research contributing "more noise than signal". We stand in the shoulders of giants: in our day to day jobs we borrow heavily from multiple academic disciplines, and many of the designers who led the introduction of user-centered design in industry settings were academics themselves, or in some way involved with academia. I think we should be appreciative of their work, even if some of their output does not directly contribute to our applied design activities.

There is no harm in creating a summary of the relevant literature, which by all means will provide interesting context.

@grahamperrin thanks for sharing that PhD. Sounds really interesting.

[bnvk]

I think this conversation should be moved to the Discourse platform, no? I'm closing it, but re-open if I'm mistaken.

[grahamperrin]

@jdittrich please: after the GitHub users in (at least) this topic are present at the forum, would you like this (one) issue to be migrated? Re: https://meta.discourse.org/t//46671/42 the migration service was unavailable for a few days, it's now operational.

(I'll test first with an issue that I created.)

[ei8fdb]

@grahamperrin If it is possible to migrate this discuss to discourse, then yes please do.

Everyone here, except (I think) @ninavizz is on discourse. @ninavizz please do join our discourse forum. Its for these kind of discussions.

It's dead easy and you can log in with your GH username - https://discourse.opensourcedesign.net/ and click signup.

[ninavizz]

Re: Discourse @ei8fdb: Cool, got it—thank you. Had not previously seen that, despite poking around to see how to participate.

On my prior comment: Absolutely, everything we do today was heavily informed by academia, and going forward academia will continue to inform what we do. No disrespect was at all intended to academic contributors, and I apologize for doing a poor job reflecting that in how I chose my words. Really, truly @belenbarrospena @grahamperrin. @jdittrich cited (in the comment above mine) that ideally a paper could be an output—and my objection was to the output suggestion, not the research endeavor/findings themselves. Re-reading the OG "Issue," I also realize that @jdittrich wrote it for existing FOSS contributors—not prospective ones.

My point was in reference to existing outputs of academic work currently available, and outputs being prepared and published as complete white-paper studies, today. Traditional academic outputs are regrettably too far out of reach for mid-career designers—and many of us do want to get involved in FOSS, after student loans get paid off and we're hungry for new challenges.

Most FOSS contributions are done on the side of day-job stuff... and the service experience that's being created for UX designers wanting to contribute, needs to be considered as an equally high priority, parallel to new findings from additional research. I realize an obvious thought, is "Well you people should pay attention to academic journals, then!." We don't, though—and isn't the raison d'erte of our profession, to put information where users will find it? In this instance, designers yet to be engaged in FOSS are users I'm not seeing considered—and given the volume of unusable FOSS products in need of more UX love, that's imho urgent. White papers as a published format, was what I cited my signal/noise disparagement with regards to—not the research itself. I also don't fault the academics or researchers for that, as whitepapers are research-credible formats for a reason. There's others of us in the pipeline of access, who need to step-up.

@grahamperrin The service experience your thesis can offer working professional designers (not researchers—big difference in audience!) is the signal/noise frustration point I made a fumbly jab at speaking to. I'm delighted to see you're doing the work of a PhD Thesis on the topic (actually, more than delighted!)—I'm just feeling rather desperate, to see new works made more accessible to those of us not in academia, and many years removed from our own past academic endeavors. Too many have the interest, but refuse to bother because FOSS is too-opaque of a walled-garden universe to outsiders both long removed from academia, and many-years deep into corporate cultured careers. Users pay the ultimate price for too few UX folks being involved in FOSS, and as grateful as I am for all the meaty research, I want to see it put to work.

[grahamperrin]

… doing the work of a PhD Thesis …

It's not my work, but thanks :-) … three days ago I had a chat with the author about additional things that were learnt through the research. I might see him again next Tuesday (monthly meetings of the Centre).

---

If you'd like to know more about my interests and experience, revisit https://discourse.opensourcedesign.net/t/-/41/22 in the "say hi" topic … maybe a few weeks from now. I'm not rushing to share information in that post because for me, whilst the forum takes shape, "it's nice to learn about other people" – more through their actions, across various channels, than through the words in that topic alone.

For now there's my basic profile (the avatar and so on), but please note that what's linked from the profile card is not representative of my interests in open source and in design.

[grahamperrin]

… migrate this discuss to discourse, …

Over to you, @jdittrich – or any other administrator. At your leisure. Thanks.

[jdittrich]

I did recreate the topic and some answers on discourse: https://discourse.opensourcedesign.net/t/-/150