Ask HN: Do you have something you continually work on for years? | Hacker News Reader

Ask HN: Do you have something you continually work on for years?

5 pointsge9636 comments

This is a personal project/passion you develop.

I'm the kind of person that starts/abandons a lot of projects. Or I'll just do enough to get it done but then I'll move on and do something else. Most of the projects I've built are just weekend projects.

So I'm wondering what is something you work on for years and years and it keeps growing/would be hard to build from scratch again.

It would be nice if it was a tool you use like a notepad or some finance thing, personal app, etc...

View on HN

I've been working on lots of little tools for years, sure. They are all extensions of theories or long term projects though, so they are like theory-practice glue scripts in many cases.

A basic first step here with projects (tool or no) IMO is to keep a log.

Over time you add different logs for different projects. It's important to have lots to choose from. One of the best ways I've seen people crash and burn their long term projects is to make sure there are only like three of them. Especially if they are passion projects.

Then I have a system that kind of shuffles the various log topics and I pick some that I feel like working on, usually at least 2-3 a day.

So there are things like stories, TTRPGs, lists (this is a great place to start IMO, maintaining a list over time can teach a lot and is very forgiving), bigger hobby projects, and so on.

IMO #2 one of the most helpful things you can write in a log is "this project isn't going anywhere and I don't know what to do about it."

Just some thoughts, good luck.

I do keep devlogs. Nice to refer back to.

> a system that kind of shuffles the various log topics

that's an interesting thought

I think more so, the project kinda just dies, like it's over... built it. So I'm hoping one day I find some that will be "life long" where I can pour time into them and hopefully it's worth it.

In that case it seems like a very good idea to track your interests.

As an example, I started a list of my interests and it just passed 1K this year. It is extremely helpful in showing me what my more stable, longer-term interests are. That is, generally the level of energy I can devote to the interest is high. Projects are more likely to move forward.

In these cases I know I'm going to be more likely to come back to or continue a project. But with more transient interests, that can be a really dumb idea. By definition the same energy with which you started the project is now gone. This is almost certain to be difficult, if not depressing work.

Over time one can begin to track interests and access to project energy like it's a kind of horoscope, in some ways. It's pretty fascinating.

I do have various forms of personal thought collection eg. chrome extension, apps, web apps... I am lacking the "in your face/reminder/summary" part.

How are you handling that/what is your interface like?

Also 1K you mean individual topics or entries

edit: I also save all of my open chrome tabs before closing/starting new

Thought collection...you got me thinking. Some years ago I finally decided to really own this question, after decades of using others' tools, from Dokuwiki to Keep to Notecase Pro. I thought it'd be hard to maintain my own but that hasn't been the case at all.

So I guess my first suggestion here is to casually start your own, opinionated system if it seems useful. Because today I feel like I can solve just about any problem with mine.

My system is based around some standards, like I use Markdown and folder taxonomy, with a few types of syncing for redundancy. I write my own tools to parse for various tags and things. Like I have one snippet that will compile a daily schedule for me based on tags it finds in my markdown files. For interfacing I use a combination of snippets (calling external tools usually) and system keyboard shortcuts. There are a number of different search shortcuts using different apps like FSearch, there are shortcuts for opening yesterday's/today's/tomorrow's journal page, a snippet for recommending books and movies, etc.

1K (items in the list; the file itself is 53KB) is hard to explain. At first it was very troubling to be excited about the list growing... Is this just a numbers game, or what? But pretty soon I found that for daily use, specifics are very important. So for example "The Cold War" is in there but it's worthless because I shouldn't have to think so vaguely when it's suggested to me as something to dive into. It's not helpful.

Better would be "Ranking favorite Cold War movies in my TV and Movies Log" or "Reading about BRIXMIS". Which are both in there.

Tabs are an interesting topic to me as well, though admittedly not in the interests file yet ;-)

https://www.friendlyskies.net/maybe/why-closing-tabs-is-so-h...

Thanks for the ideas on the notes.

Yeah... I like Apple's note app, also like MS one note/Medium editor where it has the bolding and ability to drag/drop image. Not hard to do on the fly but yeah.

I like that though like functions eg. "Snippet recommend books"

At the moment I'm just dumping things into a basic MySQL stored on a Pi ha. Need to centralize it, probably make it remote/cloud. I also use static encryption but not the local network one.

I'm gonna look at my stuff, think about making new wrappers... Tried cross platform before electron did a pretty good job. Mobile not so since I'm not a mobile developer. Would like a widget on Android. I'm using color notes at the moment not "secure" since not my own app.

yw. So it seems you have a helpful kernel of a discussion in your mind, regarding tooling for knowledge. I know that kind of thinking really helped me. gl with everything.

I wrote a tool to help students organized and sustain academic projects. It took me 5 years and 13,000 lines. I worked hours on it almost every day. The reason I could sustain such an effort was I once had to withdraw from Stanford because of ADHD, and I wanted to help others avoid my problems. [I though I might make a few bucks too.] I think it might help some of you sustain your projects. If I mention it's name, I'll get in trouble from the moderators. Look on the Mac App Store under Streamline Academic Writing.

If your comment is relevant to the topic, you can share links. This isn’t Reddit.

That's the ideal. This is my third HN account. My previous two accounts disappeared after I make relevant references to my app. The use of power corrupts.

I've been working on an uptime monitoring SaaS (that evolved into more of an "Incident Management" service with status pages) for about two hours each weekday since February 2021.

I think what keeps me going at this point is mainly habit and still learning new things each week, whether it's product management, marketing, accounting, or how to survive 19k people flooding your demo status page...

(the link is https://onlineornot.com, for the curious)

This is awesome, great to see more products cropping up in the space too.

For context I started https://rootly.com/ if you have any feedback :) Happy to talk live and compare notes!

I read about your service here and have been using it for a while. Your self promotion is working :)

I’ve started a lot of ambitious DIY projects that I then abandoned (AR glasses, beaglebone based cellphone, etc).

My (still unfinished) PhD thesis is my longest project so far and its teaching me how to keep grinding. I feel like having a deep motivation and some structure helps for me (although a very tenuous one as I started during lockdown so my PhD time was mostly remote work so far).

Being autonomous motivates me to do things, but having people you have to explain your stuff to helps to keep working when its hard.

How far did you get with the hardware projects? It is a bit of work to design something/get it made and also approved.

I had pcbs made. For the ar glasses I had a working v0.0.01 (working on a small microprocessor and low battery, with bulky cables but it could run a production ready screen nicely)

Edit: I’m talking about the screen because the hardest part was probably to source it and then interface with it (my screen was 720p and like 1-2inches large which is of course not your standard part).

Also I have to add my background in electronics is being the average arduino tinkerer, having an ambitious project in domains I know better helps finishing stuff too

Dang yeah I personally am an Arduino/RPi/ESP/Teensy tinkerer as well but have yet to actually make own my PCB design (gerber file? or something).

That's a goal in the future, that would pretty neat use Kicad or something.

So what happened to the AR project? Just scrapped it?

I'm backing a PCVR company and they're making small display stuff too it's pretty neat.

Streamie (streamieapp.com), which started its life as CameraMan (macOS only), was my DIY solution for getting my security cameras on my desktop computer. Later, when tvOS (Apple TV) finally added VideoToolbox support, I had a much better platform to target for such things.

I started dabbling with this project in 2014. It’s a full time endeavor these days. I become _very_ uncomfortable when I think about how I’d react if I lost all copies of the source code. I don’t think I’d try to start over.

That's a cool project. I looked at your pricing page.

How is business? Don't have to say. This seems like a project/area where it would be hard to compete against big names although I don't know of any funny enough. It is nice if you get use out of it yourself.

My co-founder and I were working at Instacart and started building https://rootly.com/ (incident management on Slack) before turning it into a full-time job, building a team, and now working with some fantastic companies like Canva, Grammarly, Squarespace, OpenSea, etc!

Congrats

I have been working on decentralization for 3 years and almost have it universally solved. Decentralization is not what most people on HN appear to think it is.

> I'm the kind of person that starts/abandons a lot of projects.

Is this fun for you? If so, why change it?

It is fun as far as easy wins, I just want to find something long lasting/not just crap I put together and is unusable to others.

What’s most important? To make something long lasting, something that’s not crap, or something that’s usable to others? Because there’s a lot of long lasting crap that’s really usable/useful to others. And there’s a lot of great ephemeral stuff that’s really usable/useful to others.

Importance is subjective. I would like to be financially free eventually. But also will continue to build things/keep learning. Everything I build is public but no one uses it which I guess why do I care.

There's also the "hard" barrier. A lot of stuff on YT is like easy to consume/hand waves over the difficulty but that's what gets the mass attention.

Useful is a self-feedback mechanism since people asking you to fix things/seeing an impact that helps drive it/make it long lasting.

We need you here:

https://news.ycombinator.com/item?id=32430592

Have been and will continue to, living code gen and data modeling

https://github.com/hofstadter-io/hof

Is there a video of it working? I see low code/block diagram.

I was thinking about that, would be cool to stitch things together in a miro style, maybe it works... not sure if these are what notebooks are eg. Jupyter.

I don't have anything recent on YouTube, but I do stream development on https://twitch.tv/dr_verm which often includes running it, also early preview of new features. I'm getting ready to start the new "create" command, which will be like react create-app, but for any combo of tech / lang, and also getting updates from upstream. This is something like a living bootstrap for anything, fixing something that always frustrated me... one time bootstrapping.

https://docs.hofstadter.io/first-example/ takes you through the process of building up and using a code generator

I will check some vids out. I just glanced through the readme so I can't say I 100% get it right now.

Cool and thanks!

After the next release, I'm planning another big documentation push. I'll be sure to make a video since you aren't the first or last to ask :]

I'm looking at your folder structure (in a video).

Just thinking the generators could be a way to learn syntax between different languages.

Also have not heard of CUE before interesting work.

Wonder if twitch has playlists, looks like you're 4 streams in so far. Hope the mic/sound situation improves over time. Not sure if it's a filter seems to cut off/start late as you talk. -- that seems to be just on the first video nvm

CUE (https://cuelang.org) is not my work, it's Marcel van Lohuizen's (https://twitter.com/mpvl_)

I maintain https://cuetorials.com and am betting the farm on CUE with Hofstadter

That's an interesting idea that multilingual generators might be useful for learning! I'll have to riff on that idea at some point

Twitch only keeps 2 weeks of videos, unfortunately

I'm planning edited videos for YouTube, with playlists, but also embedded in the docs per section. I even have some CUE that helps automate playlist organization

Yeah it would be cool to see it from the beginning if you have the footage still.

Anyway good luck, I plan on passively watching the twitch recordings.

Reply on news.ycombinator.com