Button Button


I'd like to see things happen to make it easier to collaborate, because it would allow specialization on the parts of problems that are more fun personally

Rockstars and toxicity

I know people who enjoy programming more than I do, but many want to be 'rockstars' whether they have design skills or not. Collaborating with a designer has some sort of implied dominance that "takes the fun out of it" for them

I only know 2 ways to get things done in this environment as a designer and neither does anything to de-escalate the toxicity:

  • Present ideas in a way that promotes those with ego issues taking credit for implementations

  • Do it myself

Both of these are straight out of my grandmothers' advice for "dealing with men" so, yeah. That's a thing...


Seriously, I'm not interested in designing to establish dominance or social capital. There are things that don't exist yet and I wish they did. The infrastructure we need to have games and media in the fediverse without it turning into a capitalist shit show is 2-3 steps from the current infrastructure and also 2-3 steps from infrastructure proposals like Spritely. We know the infrastructure is flawed, yet the proposals seem so far away

That's the fedi as an example. The same is true for producing games and media, themselves, as it is for the infrastructure. The choices for creators right now are you either specialize in what you love and come to hate it because of rent seekers or you have to spend a great deal of time outside of your interests in order to complete your work. Frankly, we can do better. This isn't a rant; it's a challenge

We have the skills on the network to do the things that need to be done. The obvious example is coding. If you want to build a thing and have the coding skills, but aren't sure where to begin... that's a design issue that I could help with. Tell me what you want to build, what tools you want to use, and what your needs are as a developer. I'm in a space as a designer where I could do free design work if building the thing is on my very long to-do list, or I can work for as little as 1/3 market rate if it's not something that I find ethically disagreeable. I'm seriously motivated to provide support here, and I only want to charge what is necessary for those working on projects for their own fulfillment to maintain their sense of ownership of those projects

Ultimately, we need better ways to find collaborators. It's fairly obvious how certain patterns of commerce will play out, so some of the spam prevention infrastructure in the proposal stages needs to be built out in the process of creating channels for economic match making

I hate sports analogies, but...

Let's talk about how to move the ball forward

I'm thinking specifically about some strategies in American Football from the 1970s-1980s called "West Coast Offense" where receivers would run patterns that compensated for deficiencies in the passing game of quarterbacks by putting themselves in position for high percentage passes to be completed, even in cases of extreme pressure. Especially in the earlier versions, the playbook would call for the passer to pick a spot in the field likely to be free from interference and dump the ball in that spot with confidence that there would be a receiver in position to catch it

The fedi isn't a football team; we're mostly a herd of cats

We can, however, follow the development trends and try to predict what needs to be done next

For example, those designing and building prototypes for OCaps have mentioned the idea of "stamps" (more like a visa stamp than a postage stamp) as a proof of work for reaching public inboxes. There's also the Groups Actor and the underutilized tags property in ActivityPub. Machine learning is out of favor in the fediverse, because it's been falsely billed as a foundational technology, but it may have a legitimate application in situations where individuals and small communities have control. Training ML is labor intensive, which is why it's not more widely deployed for small groups. ReCaptcha was built as a way to crowd source OCR training. Building meaningful connections in networks of consent probably will involve Groups as an additional structure to instances for scoping interactions in public spaces and building out tags as another way to discover others who have similar ideas to one's own in specific ways. If group moderation and curating tag trees are too labor intensive to be practical, can we use the proof of work requirement behind the idea of stamps enabled by OCaps to reduce the manual work of training the ML classifiers? I obviously believe so

More work for me

So I guess that I'll dump some of my higher priority designs in a blog series. If you want to work on something you see and would like more details, tag me