Thanks! So far, we’ve managed to get a few companies to use free software, but it’s still a relatively niche sort of thing.
Er, a few, actually. Some companies are pretty worried about their customers burning things and getting angry about it. With massively-overpowered lights making 4000+ lumens in a very small form factor, a few items have earned reputations as “nut roasters” which have literally burned holes in people’s pants while the pants were being worn. It takes only a few seconds.
So they’re interested in shipping with safer settings by default, while still allowing people to unlock the full power after they’ve bothered to read the manual.
Not every light is like that, though… so the defaults are configurable for each model, and some models will enable full power by default even in simple mode. It depends on the light’s power-to-mass ratio, target market, and the desires of the company who makes it.
Additionally, the user can choose to enable full power in simple mode… but it requires reading the manual to figure out how.
As a general guideline, the simple mode’s default limit is a little higher than what each model can thermally sustain during use. Sometimes that means full power, sometimes not.
Aside from companies, the main people whose requests are being addressed by the Simple UI are people who wrote very angry reviews about how Anduril is a useless, overcomplicated toy, and they want it to just do normal things like a normal tool so they can get some work done without having their tool spaz out.
The way people expressed the idea may have been rude, but the underlying point was solid. There is a design philosophy that a creation is not finished when there is nothing left to add, but rather, when there is nothing unnecessary left to remove.
So the Simple UI is an attempt at that design philosophy. It takes the full-featured original UI and then removes things until only the essentials remain.
I hope we can get to that point eventually. But so far, very few manufacturers seem to do any safety testing… and none have attempted to modify the code. Mostly they just don’t want to get returned items, or reviews complaining about the product being too dangerous or complicated.
It sounds crazy, but one of the top reasons for returns and refunds is: people have difficulty figuring out that they need to remove a legally-mandated sticker from the end of the battery. And they return the item instead of reading the manual. So it needs to ship, by default, in a state which is compatible with that type of user. Anything complex or confusing gets locked behind a knowledge gate, so new users won’t bump into it by accident and panic.
Anyway, every company I’ve asked about this has said they like the idea and look forward to it… because it’ll hopefully make their lives easier. But we still have to work out the details of where the default limits are for each supported light.
I’m honestly not sure what would be better. I’ve been using 3C lockout for over a week, and the extra speed is nice… but I kinda prefer 4C so far — especially now that there are direct shortcuts between ramp and lockout.
About Anduril dying, I’m not too worried. It’s reaching a point of diminishing returns, where further optimizations don’t provide much benefit.
So I’d like to focus future development on the FSM library, and on other UIs. People have had some really interesting ideas for other types of interfaces, and I think those ideas deserve more attention.
Also, on the ecosystem in general. Like, making it so more lights are easy to reflash, making it so people can choose whatever UI they want at time of purchase, and generally getting companies onboard with the concept of free software.
Exactly. Those ideas generally need to be developed into their own complete applications, not just tacked onto something else as an afterthought. I can cram some things into Anduril in a relatively coherent manner, but some ideas are different enough that they don’t fit… and they make a good starting point for a completely different interface.
Like, let’s say I’m making a salad. A lot of ingredients can go into a salad and it’ll still be good. But if someone wants to remove the lettuce add replace it with bread, melted cheese, and tomato sauce… I think what they want is actually a pizza, not a salad.
Both are good. But they’re better as separate items, instead of blended together.
As for what Anduril is, it’s mostly three things:
- A solid core designed for a single e-switch and a single light source
- A bunch of extra stuff I personally like
- A bunch of extra stuff suggested or contributed by the community, when it can be made to fit in a coherent manner
The solid core is what I found when I did an informal survey of various flashlight communities to see what people liked best for interfaces with one e-switch. Around the world, people mostly settled on a single idea. It needed just a few essential functions:
- Off 1H -> moon or low
- Off 1C -> med or mem
- Off 2C -> high or turbo
- On 1C -> off
- On 1H -> up / down
- On 2C -> to / from high or turbo
Other details varied, but the part above was almost always exactly the same no matter where I looked. So that is the core of Anduril, and anything which doesn’t fit the above description is … not Anduril.
For example, changing “Off 1H” to be a momentary function… is a big enough change that it probably should be its own UI. One implementation of how that works in practice is tterev3’s wonderful MELD firmware.
Also, ideas about multiple buttons or multiple sets of LEDs are great… and worth developing a brand new UI for. Trying to retrofit Anduril over that would be really awkward and would likely produce a poor result.
But my time is limited, so I’m trying to finish this project before I start on any new projects.