4 ms·
Apply HN: It's 2:30am; the baby's crying
Let's keep it short:
There are apps for everything and everyone, but these rarely do exactly what you want.
Some are more feature-rich (making them too heavy for the one task you need), or too targeted, missing some feature(s) you want.
The idea is to provide a cloud-based platform for personalized apps that do a small number of exactly the thing(s) a user might want.
But who’s gonna develop these exotic apps?
Let’s make it dead easy for developers to publish small, but smart apps: the platform provides processing, storage, scheduling and integration capabilities, allowing apps to respond to events (e.g., app might run by sending SMS or when a tweet is posted).
The intent is to strip any overhead hindering developers from publishing pretty much any app they think of or commissioned to develop.
Use cases (and I mean actual apps currently used in the pilot stage):
1. SMS-based TODO app (enhances feature phone capabilities)
2. Customized morning SMS weather notifications
* customizing info, units and formatting
3. Tracking and notification of rare but important website updates
4. Automatic code backup from GitHub to BitBucket
* a royal pain otherwise
5. Apply HN leaderboard and highly-voted newcomers daily email
6. Near real-time tracking of Wimbledon match results vs. betting odds
I’d appreciate any feedback and elaborate on how the prototype works, share app code examples and answer any questions.
* Shameless plug, for someone else: Maciej Ceglowski (of Pinboard, currently apply HN leader) gave a brilliant talk titled ‘Barely succeed! It's easier!’. Check it out at goo.gl/j2SxTY
- Piskvorrr 11y agoIt's 3:30, the baby's still crying, and I'm all confused. Maybe it's the exhaustion, maybe it's the wall of text. What platform? Are you building it? Or are you building the apps? Or are you just saying that it ought to exist? The only link you're providing is some talk on YT.
- mindset 11y agoYes, I'm building the platform, not the apps. Apps will be build by developers independently, or after being commissioned by non-coding users. When the service is in decent shape to post link, I'll happily do so here, or in the show HN section
- Piskvorrr 11y agoAha, now it makes sense. Looking forward to that!
- phillc73 11y agoJust a couple of comments on some of your use cases. * I don't use Bitbucket specifically, but what's difficult about mirroring code between two different git instances? I do it between GitHub and GitLab.[0] Local git config just needs to be configured to push to both destinations. Or perhaps I haven't understood the specific "code backup" use case? * The Wimbledon match results vs betting odds could be interesting, but I would think the betting odds should very quickly reflect the match results. At least this would be my experience on Betfair. What's the opportunity here? Is it arbitrage between bookmakers? [0] https://github.com/phillc73/backblazer https://github.com/phillc73/backblazer and https://gitlab.com/phillc73/backblazer https://gitlab.com/phillc73/backblazer
- m-i-l 11y agoI would think the betting odds should very quickly reflect the match results. In actual fact, the betting odds are sometimes updated before the broadcast results, thanks to bookmakers employing courtsiders[0] to transmit scores direct. [0] https://en.wikipedia.org/wiki/Courtsiding https://en.wikipedia.org/wiki/Courtsiding
- mindset 11y agoYes, much like the stock market responds to news, mostly faster than msnbc can discuss these. The issue this app solves is the ephemeral nature of the betting data. A simple scheduling feature allowed me to have historical data easily. Otherwise, go deal with a server, scheduled jobs, etc. Not impossible, just discouraging for many people that can develop, but don't want to deal with the overhead for every little application.
- mindset 11y agoFrom my attempts, it's not trivial with bitbucket. Beyond that, the functionality for this app also includes slack notification on new releases,and a nice templated email notification on what's changed. That was the request from the guys wanting this app.The point is that an app can do a anything a user wants, tailored to that user. The betting odds app was my own pet app, mostly for testing purposes. I listed it to demonstrate something like this can be done. Otherwise, a server, database and code all have to be setup, tested, and monitored. The idea here is to let developers focus on the code only, and easily publish apps for anyone to use.
- bryanrasmussen 11y agoI guess I'm confused. Is the crying baby a problem that will be solved by one of these small apps? If so, how? Because the tools that are provided do not seem like any tools that would be especially useful to develop a crying baby handling app? also - a link needed.
- mindset 11y agoNah, the title has little to do with the app, more with my failed attempt at brevity. But an app that could stop a baby from crying would be awesome.