beautiful soundscapes
50 karma · joined March 16, 2019
I do think people nowadays over-index on iteration/shipping speed over quality. It's an escape. And it shows, when you "ship".
If queries can be dynamically written on the fly, we wouldn't need automation/agent to interact with the UI (as is the case with chatGPT operator or anthropic computer use). Instead, an intelligence layer for the frontend could dynamically query based on a graphQL schema and render with some basic html elements.
Thoughts?
Practical answer: we want to do cool things like "hover over keypad and say a name" and then you can talk to them. Only possible when we build the whole stack.
Thoughts?
Here are concepts of the protocol that we poorly explain on the landing page: 1. You tune into you're top-of-mind folks. This set of people is different for everyone and it changes project to project. But it's generally only 5-8 people. If your device is ON and you are not in another conversation, these people (and any combination of them) can get to you instantly. 2. Other people have to send you voice messages. 3. When you press the play button, you only go through all of your top-of-mind peoples' messages. Then there is a filter called "inbox" for "others". 4. If you press the flash button on the keyboard, you are OFF. Everyone sees a "-" for you. You don't hear anything.
So in aggregate, although it doesn't seem so at first glance, I built flowy so that there is less noise. We following "less is more" principle a lot more than it seems. It's just the sound of your people: no notifications, no fake AI voice assistant sound. This being said, I realize that we do not do a good job of articulating how everything works, so we will make a full demo video at the end of the year.
They pay thousands per employee on saas tools + remote office reimbursements + give iphones and cell plans to their employees. What flowy offers is (hopefully) priceless to them; that is, if the promise is fulfilled. They want to try it for a few teams for a couple months, and go from there.
IMHO, we underrate how much we invest in commz. Human to human commz is worth more than any computing tool. What is an organization without people collaborating? A band where everyone's playing solo. This is part of our philosophy :)
They pay thousands per employee on saas tools + remote office reimbursements + give iphones and cell plans to their employees. What flowy offers is (hopefully) priceless to them; that is, if the promise is fulfilled. They want to try it for a few teams for a couple months, and go from there.
I also want to make sure that those who use the keypad are 99% at their home desk because we can nail that experience better than building a mobile variant right now. I'm planning a mobile version (keyfob/earpiece) for the future that would cover a wider audience's needs.
However, perhaps eventually there will be a battery variant; from an engineering perspective, it adds complexity to make sure that it lasts and handles different environments. I thought about adding battery and giving customers a wireless charging pad so that it can just sit on it. But I told myself, for v1, let's just make it stationary just like mechanical keyboards; reliable, simple, minimal.
All jokes aside, I appreciate the simple and actionable feedback.
Knowledge work, for example, found email and hung onto this. All of a sudden, we lost simplicity and are in this bubble of noisy "software interfaces" being the standard.
Perhaps it's just that newer generations don't know how things used to work, and the old people are retired or can't do much about it?
I have different approaches on screen + commz combo that uphold flowy's product principles of keeping it lightweight, and augmenting human conversation. Your idea around this is really interesting; I agree that this is a "killer" feature if done right.
Our bet: flowy is what loom users have been reaching for, but haven't been given.
Still, I have seen that these folks still have presentations in English and with diagraming and such, they are able to articulate something that's more digestible than full on documentation. I think loom sort of videos would still be good for these people to diagram and demo things. We want to add something like this with video + screen broadcasting.
Yep speech to text won't be fluent with technical terms. I'm hoping it does 80% of the job, and then constantly improve it based on what you mentioned. Thanks for the suggestions! Perhaps starting with editable speech to text would be viable, while trying hard not to promote async communication to a point where people aren't conversing naturally as humans do but constantly stopping, editing and typing.
That makes sense, we need to try different messaging to hit it on the nail, but "walkie-talkie" seems to click as a decent initial introduction.
I totally agree. When I'm in flow state, I don't want anything to disturb me. Hence, I need a power switch, which is the power button in the widget. I'm a minimalist myself and having an off switch has been awesome for me, as compared to all of the configuration I have to do with Slack notifications to have it do what I want.
Thoughts on this?
We have live transcription and power search on the roadmap as this is a common objection. (audiobook vs. book argument)
Speech to text is pretty good these days, and it's 2023, we shouldn't be typing to talk in our opinion.
Search: "Play (or let me read) what Johnny told me about databases last month or so"
Me personally? When it comes to going back in a conversation, I prefer sitting back and listening at 2x speed about complex topics so that I can look across the room and think instead of staring at my screen reading essays.
We think the philosophy that flowy is built on is the only reason it will beat other standards that we have just lived with: We live in a distracted world. We used to shuffle paper, now we shuffle 20+ apps merely trying to say a few words to each other.
On the product side, we focus on how we can make this more like a natural human conversation. This results in reduced cognitive load and easier conversational flow.
Thoughts?