"now that ncurses support is being included, a visual mode text editor would make a lot of sense. Every OS needs it text editor."
"crypto functionality. If this interface had built-in support for lightning payments, that would be fun!
The crypto community is dying for a decentralized platform to build upon. We have so many great DAPP developments that are polluted by using centralized tools such as AWS, because they don't have any other stable alternatives at this point. We need to build a simply but effective Urbit demo app that can provide these developers with the "aha!" moment they need to realize that Urbit can be the solution. Remember your own "aha!" moment that inspired you to be here. For me it was as simple as one single evening reading the bitcointalk.org bulletin boards six years ago. Building that kind of UI on Urbit and exposing it to the growing army of DAPP developers could be a great first step"
"urbit apps should be programmable or scriptable so that you can get data you want out of them or configure them to watch for stuff you're interested in"
"the idea of a "twitter scripting language" doesn't make sense for actual twitter but would for urbit's twitter-equivalent"
"(something like the idea of agents from the semantic web)
"also search will need to work across federated servers for some things, eg. a federated etsy would be fun but it'll only be useful if you can search for "t-shirts" across all publicly-available stores"
"Some way to get files into and out of my planet. Something like Dropbox."
"A blacklist. Everyone on Urbit has been awesome, but it would be nice to have a list of people that you never want to DM you ever."
" Some kind of posting and looking up info about the owner of planets. I can get my head around ~piddip-fatryx but I'm old fashioned and would prefer to think of '~piddip-fatryx, who is Fred from Des Moines.'"
"A way to use your planet to control what info comes into and goes out of a particular moon. The use case I am thinking of is educational apps for my children, but I suspect IoT devs will want to harden their moon-devices, too."
"Also I would be interested in seeing a collaborative 'repl/editor'"
"We should make a choice as to if we want to integrate with outside editors (and if so implement the LSP) or want to build our own."
"I love the notion of a per core immutable always on version control"
"+1 on app research consumption monitoring"
"someone once suggested adding minikanren as a query language, that sounds cool
I guess there's a broader question as to whether urbit needs a query language (like sql) and what it should look like"
Modules
"a streaming module"
"bazaar/shopkeeping module"
"a shopkeeping module available for anyone with a star to instantly set up a means of selling planets (and whatever else)"
Applications
"Get a custom t-shirt with your Sigil via Printful API"
"Blockchain light node and crypto wallet"
"A personal assistant that other people can access to interact with me: (1) you want to send me money, what's the best way to do it? (2) you want to buy me a gift, what do I like? (3) you want to schedule something with me, or maybe you just want to see me more – ask my assistant to work you into my schedule"
"LiveJournal." "A simple broadcast-mode livejournal replacement would be nice
interoperability with notes-taking similar to this would be cool: https://cl.ly/318f74. This application 'Standard Notes' allows one to install a plugin that lets you publish to a 'blog' service"
"I personally love the idea of a single 'creative surface' (like text input) which can then be delivered off to a variety of outputs/end-surfaces"
"A bookmarking service (like pinboard.in) would be great too. Not a lot of data, but a lot of social-interactions."
"It might be even useful to creating a personalized/social search engine. You're looking for X? Check bookmarks of trusted friends (and friends of friends)"
"i like this deterministic password manager idea, would be useful for generating new keys every minute or something as a 2fa style code across different urbit-related apps (Bridge <> Landscape)"
"i like the deterministic password manager 2fa in urbit, because a lot of existing solutions is already done, so a lot of work integration/UI is done; fork + add Urbit intgretaion, submit a PR, done"
"I'm a software engineer, but my main interest in Urbit is not necessarily to hack on it (although I do want to do that too) but to share content, build communities, etc. I love the idea of Agean as a whole and would like to see it become a robust tool to build such communities. IMHO it would be a pity if it would linger in an unstable or unusable state for too long. I'm really looking forward to the different tests with communities that you are going to do. And just to clarify, I'm not looking for just one more way to build online communities. No, I'm 100% behind the fundamental ideas of Nock, Arvo, and so on. But I'm interested in how they'll come together in Agean in general and Landscape in particular. What I'd love to see is a twitter replacement, but with more powerful content federation features, and an easy way to share content with the open sea that the internet is, but also keep certain things internal to smaller Urbit circles, to keep them in my shallow little backwater, so to speak. Essentially the possibility to build a single feed that combines all the verous interests from the Urbit world into a home base, and lets me reshare some of them with the world. The way "inbox" works in the browser, already looks promising!"
"One step planet creation. It would be really cool if a star owner could enter one command which would: register the planet in Azimuth, and then also spin up a VM which starts running the Urbit runtime for that planet, and then sends an email with ownership info. In other words, distribute a planet without the recipient having to figure out AWS or ideally even know what an AWS is."
"this is the future, we don't need to delete... ever."
Platforms
"A tasteful means of broadcasting various applications/functionalities/utils built on/for urbit, somewhere on the website ideally the first probably-not-apt metaphor is some sort of 'app store', but i get the sense that's not how you'd want to message that like Ur-Chess, and the MUD that was made a long time ago, things like that. Maybe as a way to show people are -actually - building interesting/useful things via Urbit in its early age."
"Yahoo Pipes, but way better"
"I have a very down to earth dream: finally fleshing out the new web server and coupling this with some more permanent API. We could have many apps springing up, but with "shifting sands" foundations it is difficult to keep up for people with no direct Tlon link."
Design Principles
"I think there's gotta be a nice balance between customizability and ease-of-use for the specific types of things people do on the internet. usually this shows up as companies offering a specific semantic interaction that -prescribes a decision on the customizability<->accessibility spectrum (i.e. LinkedIn, Shopify, Wordpress), but there might be a good way to do that while respecting the open network"
"or for planets, a module for sharing out moon invites for friends/family members would be nice, ideally something as simple as like, "links" and simple sign up"
"Someday, Urbit will cause data models to standardize across the OS and app ecosystem, allowing User Interfaces where users visually 'throw' objects between applications, the way that Linux lets you pipe commands."
"shared data models is totally the holy grail for something like urbit — the problem is adoption and defining data models that happen to strike the right level of abstraction for whatever semantic thing you're trying to do, like messaging for example: you can't (easily, efficiently) build all of the features of FB Messenger on top of something like XMPP, so they invent their own data model that won't be compatible with every other messaging interface."
"'Applications' as just config files for a pipelining cron-like process runner. I create a text file on my machine with some front-matter that labels it a tweet. It syncs to Urbit. It gets sorted in the file-system. It gets added to a queue for posting to Twitter tomorrow at 11am. I receive a notification to review before it's posted. That's just, like, my workflow man. Other people can fork it. Nobody ever needs to use Buffer again."
"I think it is generally a good idea to leverage some obvious properties of urbit: it can be (somewhat trusted), relied on as always-on and always reachable. That why I think the 2nd-factors of all kinds would make a lot of sense."
"this is probably way in the future and not very down to earth but ~nolmeg's comment about a personal api has made me think that, although with a small impact in the short term, making data sovereignty a real thing is possibly one of the most disruptive things of urbit. without something like urbit most of the decentralized services end up becoming yet another centralized one or they simply tell you to install mongodb to store offchain data. now, if your data is a digital asset is also a resource that produces value, so i can see urbit as the platform for things like openmined or the oceanprotocol for ai data. probably having these things in hoon will take some time, but using tensorflow.js or syft.js could be a cool project to integrate in the frontend.
on a more down to earth note, i love the personal dashboard/assistant idea ~hidrel, i was using camelcamelcamel.com the other day and thought: why do these people need to know what i want to buy on amazon, i could make this on urbit!"
Problems
"I run online communities and all of the tooling is absolutely insufficient. A community-built modular community building platform on Urbit would quickly take over the world."
"Problems with online community software: (1) you can't (re)design the interface around a conversational mode you want to explore – you need to choose between entirely different products, (2) with most things, you can't customize the skin / "vibe," which I think we're used to now, but looking at something like https://lain.church/ you can see how much we're missing there, (3) often core features are just implemented the wrong way – like Telegram notifications are inconsistent across platforms and a lot of people don't like how they work, (4) I suspect that if you built a community platform as a platform – not unlike Wordpress – you'd have a proliferation of very useful tools available, (5) most tools feel "noisy" and distracting"
"so i was showing a friend around the primer and video and a few blog posts and docs yesterday, and he raised some points i thought were interesting. mostly boil down to: having a unique hierarchical cryptographic identity in a market similar to real estate is great and all, but if urbit gained wide-spread adoption especially in corporate contexts, it didn't feel feasible to him that the model of having only planets->(moons->)->comets would be a data structure with sufficient levels for organizations to model their own trust hierarchies in real life even if you had enough planets for one per employee (and counting out stars and galaxies as network infra)."
"I don't like sharing my location with everyone who knows me on facebook if urbit could give me something like a "personal api" - fine grained control over who gets to see what subset of data on my urbit, updated in realtime, that would be very useful"
"Most of the users right now are NERDS(tm) and I think that will be true for some time. So we should give them tools to move off of github and assocaited tools."
Features
Modules
Applications
Platforms
Design Principles
Problems