Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
An app can be a home-cooked meal (2020) (robinsloan.com)
186 points by jsomers on Sept 13, 2022 | hide | past | favorite | 51 comments


Wow, this article hits home. Recently my wife and I were admiring the full moon, and she wondered aloud how many moons we'd been together and said, "maybe you should use your programming stuff to find out."

A month later I surprised her with a home cooked Android app called RacoonMoon. It features a random picture of us, and along with some very cheesy love messages and inside jokes, says how many moons we've been together. It barely works, and she loves it. We are the only two users.

I wrote the app in Python and ported it to Android using Kivy, which was not an easy or straightforward process. It seemed to be the best way given that I didn't want to take the time to learn Java or Kotlin, but if I were to do it over maybe I'd bite the bullet.


Awesome. Can you give a lightning talk on that? Me and lots of others would be interested.


Sure, sounds fun. This entire thread is very inspiring.


Just adding my +1 for a talk/blogpost on this :)


That's a wonderful gift.


Godot would be great for this kind of thing. Super easy android development


I have several 'home-cooked' apps. One is our shopping list. I have a Pi with a touchscreen in the kitchen running it, and it's easy for us both to add to from there. I also built in functionality so that if we're using a recipe from the web, we can share it to the screen, which is neat.

We can also access the shopping list on our phones while shopping. I built a feature that sorts the products by aisle (specific to our local supermarket), making it easier to do the shop easily/quickly.

I'm sure there are commercial (not even paid neccessarily) apps that do all this, but this one is used only by us and it's built to perfectly interact with our lives and circumstances:

* Things are hard-coded that would make no sense in a publicly available app, or would be very difficult to build customisation for. Hardcoding things is easy, building UI is hard. We're so used to building things in the commercial space for scale we forget how easy it is to build small things that will never need to scale beyond a userbase of 1-2.

* It has remained almost entirely unchanged in UI and features for ~10 years now. Can you say that of any other piece of software you use on a daily basis?


> I'm sure there are commercial (not even paid necessarily) apps that do all this, but this one is used only by us and it's built to perfectly interact with our lives

There is another factor you may not be aware of: shopping apps are juicy targets for acquisition by marketing/data mining company. I mean, there is no clearer indication of what you want to buy than what you put in your shopping list, so it's perfect source of data to mine.

Out of milk is such an example: it was a good app with 5 million users, then it got sold. Users became the subjects of data mining, or their account got deleted (if in Europe). Fun!

https://www.prnewswire.com/news-releases/inmarket-acquires-o...

https://twitter.com/JamesLiamCook/status/1407370148349198341


wow that app is taking a nosedive in the reviews https://play.google.com/store/apps/details?id=com.capigami.o...

I've been curious if the play stores etc of the world would mind supporting some kind of banner or other notification for when apps change hands between owners. Something to tip them off to the fact that the app is now under new management, much like how restaurants hang banners when new management takes over a spot. Now that I think about it I wonder why restaurants even feel it warranted to notify the public, must have some positive image optimism behind it I guess or is it something regulated perhaps?


> it's built to perfectly interact with our lives and circumstances:

this is it right here! I've been contemplating getting an old Android tablet just to run some home-cooked app. Would be nice if iOS devices could support side loading of apps, but I'm cool having an Android device just for running these focused apps and that alone.


> In a better world, I would have built this in a day, using some kind of modern, flexible HyperCard for iOS.

>

> In our actual world, I built it in about a week, and roughly half of that time was spent wrestling with different flavors of code-signing and identity provisioning and I don’t even know what.

Are you listening, Apple?


Apple doesn't care.

To paraphrase Steve Jobs, "You're doing it wrong". Just build a web app.

If you are paying an annual fee just for the privilege of being permitted to write software for Apple's platform, Apple can safely bet that you'll put up with the rigmarole of app deployment as well.


The "modern flexible HyperCard" would be something like Swift Playgrounds today. SwiftUI might have all sorts of problems for building large apps or getting very specific with UI design, but it absolutely excels for building "informal apps".

Only problem is that there's no iOS version. :/


The UI front still has a long way to go but I found that Shortcuts in iOS has some pretty powerful abilities if you know what you’re doing.

My partner and I have a shortcut for “what floor is the car on” since the car is parked in a neighboring garage without assigned spots. We have a shared note and the shortcut appends a line with the date and floor on it. A second shortcut will read the last line of the note and parse the floor and present it.

Shortcut hacks we found:

- parse text from a shared note. Remember scaling is a non-issue.

- there is a “ssh to a server” shortcut option to trigger python scripts or background tasks (or home assistant, etc)

- edit on mac app

- iOS major updates break things :(


A common person isn't able to do what the author did, meaning they'll have to reach for the app store. Conveniently, Apple gets a cut of sales on the app store. The conflicts of interest are clear.


The author is “a common person” though? They’re a fiction writer who doesn’t and hasn’t worked professionally as a software developer. that was the point. They’re “home chefs” not professional.


Aren't free apps a conflict of interest then?

I can imagine that a good platform for making "home-cooked apps" would be beneficial both to Apple and to buyers of their products.

I believe Shortcuts was a move in the right direction.


> Aren't free apps a conflict of interest then?

They are indeed, I think. When was the last time you saw a really great free app? That was at the beginning of the App Store ecosystem, I think they've moved away from it.


Arent all the best/popular apps free these days? I have no data to back this up, but in my experience people have gotten less willing to pay money for an app than when the app stores launched (presumably because everyone is used to "free" apps that you pay with your data to use).


In my family home growing up, me and my sister had our bedrooms upstairs while the dining/living rooms were downstairs. When it was dinner time, my mother would yell out "Dinners ready!" up the staircase - often to deaf ears, as us young teens were pre-occupied with loud music, gaming or talking on the phone.

After coming down one-to-many times to a cold dinner, I developed my first "app" with MIT App Inventor - Dinners Ready, an app with three buttons: "kid A", "kid B", and a larger "Both". Pressing a button would send a simple text message saying dinner is ready.

My mum loved it. Suddenly she didn't have to worry about fruitlessly yelling to her distracted kids. I wouldn't have to rely on my grumbling stomach to come downstairs to an empty dining table with my food going cold. We used it for years, and she still brings it up from time to time. To this day it remains my most well-reviewed app with 100% happy users.


Really resonates with my recent experiences.

I want off the software equivalent of the fast food menu - crap designed to bring in the most revenue for the least expense, user health and preferences be damned.

I've already started replacing a lot of the "as a service" software I used with self-hosted equivalents.

The next one I really want to hit is "family chat" since Apple is still making a mess of it for folks with families split between iOS and Android. I'd been looking at the self-hosted options, and none of them really hit the spot - I might just bite the bullet and make my own.


> family chat

Please build this! I have a large family with a mix of android and ios users, ages 18 to 80. We end up using long running text message threads as an ad-hoc social network, with people posting pictures, links, updates, comments, etc., and things get disorganized because discussions are not threaded.

I would love "reddit for families" -- for lack of a better description. Threaded discussion with easy "posting" of pictures, links, and videos, with a simple interface similar to imessage or google messages, which everyone already knows how to use.


Just to play devils advocate, what if you just made a private subreddit for your family?

While I also struggle with the android/apple divide in my family I feel like getting them all to agree to use an existing social platform would be exceptionally difficult, but it would be orders of magnitude easier than getting them to use a custom or relatively unknown app for comms unless it did something really special or novel.


So at least in my case -

I have a couple of problems with outsourcing this kind of thing like that.

Namely - a private subreddit would work but comes with a LOT of downsides.

- Ads (this is pretty much a dealbreaker on it's own...)

- Signup

- Changing UI/UX

- Lack of control over uploads/content (who backs it up, when, how?)

- Lack of control over the general experience

The nice thing about custom software is that as long as the user count is low, and it's family that you trust, you can ditch a LOT of the pain points of online software.

ex: who needs signup/login when I can just embed a shared key?


I'm fully in agreement with all of your downsides (although ads aren't necessarily a dealbreaker for me).

The pain point for me is that my family/friends circle isn't willing to try and use new software especially stuff that might be custom or have some lack of features.

It took me a year or more to get my dad to send me pictures and videos over FB Messenger instead of SMS (he has an iphone and I have an android). Maybe I just need some new persuasion tactics but the idea of putting energy and time into building a custom app for my family and then dealing with pushback and a reluctance to use it just sounds exhausting to me.


> ... what if you just made a private subreddit for your family?

Yes, I have toyed with this idea, it is after all almost exactly what I'm looking for. But ultimately rejected it for the reasons @horsawlarway stated. I guess I just don't trust Reddit.


My family has been using GroupMe this way for many years.


Thank you for the recommendation! Looks interesting and I will definitely check it out.


Perhaps not quite what you're looking for, but our family recently just switched to our own Zulip instance, and we've been loving it.


I wonder how he handles TestFlight's 90-day limit before a build expires. I'd probably just setup a GitHub Action to run every 1-2 months (to give some buffer if something goes wrong) to push a "new" build.

Sidenote, Lambda/DynamoDB/S3 (and maybe cloudfront) is a really nice little stack to work with for one-off projects like this. I've used it a couple times now for small seasonal/burst-y projects and it's always been a joy to use. It doesn't always scale (as in larger codebase) great unless you put in the work but for a handful of "endpoints" with storage (S3) and data (DynamoDB) bucket it's very nice and dirt cheap.


Instead of complicating it, you can just plug your iPhone into your mac. That version lasts a year.


But that doesn't work for the use case described in the article, right? You wouldn't want everyone in your family to have to install form your machine.


Why not? It’s 4 people. If they’re geographically spread out, then it could be an issue, but many families even spread out likely see each other once a year.


For every update, you would have to connect them to your machine.


I very much enjoy this post. My first introduction to Robin Sloan was via the novel Mr. Penumbra’s 24-Hour Bookstore. It featured Ruby programming as a central part of the plot. Ruby! Programming languages, and coding, in a novel? What a wonderful thing to find. I wish him great success.


I've tried various note taking apps and never found one that worked how I liked. I now use a Dropbox folder full of plain text files and a handful of Python scripts. I've never published them, they're full of hard-coded values that only I use, and they lack 90% of features in most note-taking systems. The simplicity works for me, and that's great.


Hah, I once worked at a startup that tried to pivot to basically this almost immediately after this article was originally posted and the CEO read it, needless to say it didn't work out.

But it is fun to sometimes program something small like this for yourself (and friends/family), without having to think about the many things you would think about in a business environment. If you can let go of those mental constraints, that is. I recently made a little mobile remote control app for my e-ink picture frame and the code is terrible. Everyone who sees it is impressed and it works without any issues 90% of the time, took me about a day and I get happy every time I use it.


(The article is not really about that app, but anyway...)

This is also pretty close to what BeReal is, which has seen huge success as a startup. I wonder what the difference was that made it work.


Home cooked apps FTW! It’s honestly such a joy to program stuff for you and the people you love and remove the expectation of it doing well in some arbitrary market


Absolutely! I love seeing my friends use an app I made for all of us.


Related:

An app can be a home-cooked meal - https://news.ycombinator.com/item?id=22332629 - Feb 2020 (130 comments)


I really enjoyed the article. I've several personal apps, and as with home cooked meals, I've a favorite 'home-cooked app' that I iterate on more than others.

Famnom evolved from spreadsheets, to python scripts, to a web-only django app hosted locally, to a web + ios + android client apps. I find it useful daily, and it has a few other regular users (besides my family), though arguably my family are my toughest users :).


I have a similar story. I had an itch about improving my communications and scratched it. My biggest issue was the App Store itself and I documented my start at https://medium.com/@mtc.dev/my-first-android-app-story-331c9....


Would be interested to know how long such an app could exist on the App Store. Apple is not fond of apps that are rarely updated.


I'm hoping PWAs continue to get better so that app stores aren't the only way of having an app-like experience. App stores just aren't designed for this kind of use case, and hurdles like "We're dropping your app bc you only have 4 users"* take the joy out of home-coding.

* fictional quote but it wouldn't shock me if it became real


One of my "home-cooked" apps has been in the App Store for 12 years now!

It's never really gotten much traction, and I mostly maintain it for me and my friends.

I love to make small incremental updates to it. Really fun experience overall!


I just wanted to say that this article fits with my personal experience so well, so refreshing. I love messing with code and different technologies, but mostly I really only want to please myself or family members with my creations. To be relieved of the need to feel professional in my endeavour is enlightening.


I was surprised that it was Robin Sloan the author who did this - building an app of that sophistication is no mean feat for someone who’s not a professional.

His books are excellent. I highly recommend both Penumbra’s 24 Hour Book Store and Sourdough.


I find it interesting that I seem to have a different takeaway from most of the other commenters here.

I assume HN is dominated by professional programmers? Id assume the reactions would be different in a demoscene or raspberry pi forum?

For me this is what I like about open source software. It's a home cooked meal that the chef has shared. It's not trying to please every one, it's written by the author for the author.

Both activities have creative elements, and a "what happens if" element. And I can certainly see the sharing motivation for both.


I made a micro-journaling web app that has three total users (my mother, my friend, and me), and only one weekly active user (me): https://joyjar.zachgarwood.com

It looks kinda goofy and wouldn't hold up to a large userbase, but it was fun to write, and it's nice to have something that does exactly and only what I want it to do.


I like the learn to cook analogy. It's perfect.

Also a good retort to the whole "you don't need to go to a famous culinary school to ger hired as a chef".




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: