Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Trackiem – Custom trackers to track anything over any period of time (trackiem.com)
29 points by avinoth 15 days ago | hide | past | web | favorite | 14 comments



Bit of a backstory for how Trackiem came to be.

I've always been obsessed about tracking individual pieces of information in my life and finding patterns for certain activities. Like, activities that lead up to a headache., correlations between number of coffees I drink and number of hours I sleep and so on..

I have tried using specific tracker tools before but they always inevitably becomes not useful as they are targeted towards specific niche or goal (like habit building, budget planning, etc)., whereas I wanted something generic that offers just the right tools and lets me build my trackers to the way I want. Hence, I built Trackiem.

I've built it to be as much un-opinionated as possible, by just offering the components that makes up a tracker and let users customize it the way they want.

This is only the initial version which lets you create trackers with basic "types" and "frequencies" with reminders. I've planned exciting things down the line for though, such as customized dashboard & reminders, more nuanced tracker types - this will allow you to build your own finance tracker, time tracker and so on.

Excited for the launch and any feedback is much appreciated!


As someone who built a habit tracker, I like it. It looks elegant and easy to use and you seem clear on what you are optimizing for.

A couple of thoughts.

I think it's important to be clear whether you building an app that is optimized for tracking for insight or for change. That is what explains why I took a different approach to designing a habit tracker than you are taking. My primary was for change. Yours sounds like for insight.

The reason to be clear is that it can help you filter out feature requests.

There's only one other example of what you're building that I've seen and thought was well done. http://reporter-app.com/

If you are planning to charge, I would want to know that before I sign up for a free version. "Free until 2021, then $5/month."


Thanks for the suggestion and the associated link.

You're right, the primary motivation was to do tracking for insights but down the line I would like if this can be agnostic even from that., like one feature I was planning to do is "Overlay charts" which lets you overlay data from multiple trackers to figure out patterns. On a different line, I've visioned if it could sort of have a "Streak" builder with triggers after certain "value" is achieved, which then allows it to have "goal" based tracking abilities.

Will do a write-up on the "manifesto" of the product, and will ensure the intent is clear from the landing page.

As for the pricing, I really haven't decided on one and when to charge. But rest assured, I've always planned to grandfather users who signed up before any payment plan is launched. Basically, anything you use for free will forever be free.

Thanks again.


Seems neat, but some screenshots or a demo account would be nice so I can see the product before I give away my email.


Makes sense. I've composed screenshots of important pages here - https://www.dropbox.com/sh/3h6xg78lfc402k5/AAAwAYw3Yohsvp84c...

I understand if you meant to have it in the landing page, which I will get on it, I'm planning to have a demo account setup by end of this week.


agree with that. Also, detailling the pricing plan when leaving beta period would be nice


I've attached some screenshots on the parent's reply, here it is - https://www.dropbox.com/sh/3h6xg78lfc402k5/AAAwAYw3Yohsvp84c...

As for the pricing, I really haven't decided on one and when to charge. But rest assured, I've always planned to grandfather users who signed up before any payment plan is launched. Basically, anything you use for free when you signup will forever be free.


Seems like something I have been looking for. I am currently using a combination of Daylio and a random counting app I found on iOS to track myself on some habits, but this looks like a better solution.

Any plans for native mobile apps, or web only?


> Any plans for native mobile apps, or web only?

Definitely! I put off building this for sometime mainly because I wanted to do it as App first. But since web version had a faster iteration, i got this one out to start with. I've already did some work on the app front and hopefully will get it out by this year end.


What stack did you go with for this?


Its Rails+Postgres on the backend & Vuejs (TS) on the frontend. Hosting is done through dokku on a Hetzner VPS cloud.


Curious why you made the yes/no separate from the frequency tracker when under the hood I'm guessing you're storing a boolean anyway? You're likely also tracking a type already, so you could use 0/1 data and the type to change your graphs, then people could change the tracker type in the future if necessary.


Under the hood the frequency is also handled as numbers (0/1), there is a way for users to change the type if desired from Boolean to X Times, but complications stem when they want to do the other way around.

Some of the main pain points it might introduced are into visualizing the historical data. When you've been saying yes & No, and suddenly you want to track count, you'd see bit of inconsistency on how the data is presented. This proves hinderance to further work on finding the patterns. Hence why..

The way it is structured now, tracker is a one immutable component of type & frequency and when one of it changes, its a whole new tracker. This would simplify the experience for the user down the line. Rest assured, if this seems to be bothering users much, I'm willing to see if it could be done the other way as well.


Seems useful.

TL;DR on my questions:

- API

- Leverage tooling for infrastructure monitoring

- Clients (Python, or others)

- CLI

- Hooks (launch scripts on events)

- Events abstraction to connect different systems (cloudevents.io)

- Groups

Have you thought of treating it like events in production and leveraging monitoring tools for humans? For example, we do work with events to monitor infrastructure and trigger actions, but that is what many of us do for a living, and not usable by most people.

You can have a "Human Reliability Engineering". If you bring the tooling to monitor infrastructure to regular folks, that could be something. Because then you can leverage tools with common abstractions and language (metrics, events, etc.) You can then leverage Grafana, Prometheus, Elasticsearch, and all the tools that do that very well for infrastructure.

Speaking of abstractions, have you checked cloudevents.io[1]? You can then open yourself to various protocols and destinations that understand these events, and people could build applications on top of your service.

Also, have you thought of opening the API and adding a client or CLI so people can control it without opening the mobile app?

For example, as a user, I could generate API keys from your site and then use these keys in a client (Go, Python, whatnot) to pretty much do anything I could do with the application, get data, push data.

Also, have you thought about adding hooks so I can write a program that says: whenever I add an event, run scripts in certain directories. Similar to Git hooks.

The script is executed with the event as an argument, and does whatever I want it to do. It can be as silly as the following:

I add an event of drinking a cup of coffee. There's a script that gets that event, and decrements a counter somewhere of coffee capsules, and if the coffee capsules are under a certain threshold, trigger an alert or something.

If there are groups, I can have a hook that alerts several people in a group that we're out of coffee.

This may be useful for people recovering from an addiction, for example. A group trying to quit a certain substance. Every time a member of the group consumes said substance, it triggers an event.

They could monitor themselves, but they may also monitor each other and see that member A is doing fine, but member B is starting to drink more and more this week. What's going on? Let's check up on them, talk with them, support them as they may be going through a rough one.

Someone more experimented could see _patterns_ of events. Substance consumption, missing workouts, increased spending, etc. Is there disengagement?

Once you generate that data in a standardized way that can be leveraged by mature tools, people can build things on top of it and you can get many ideas and use cases.

[1]: https://cloudevents.io




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

Search: