Hacker Newsnew | past | comments | ask | show | jobs | submit | mholt's commentslogin

On the contrary, I think it's even better to use real data instead of fake data. Using fake data looks tacky and unprofessional.

So you'd prefer to look a screenshot where half of the content is blurred out?

Yes, absolutely moreso than a demo that is faked.

Just fixed that. Thanks. You can download the latest CI artifact for the patch (the link is on the Install page in the docs).

Thanks a lot, I wasn't really sure if reporting here or on github would be faster, I'll try to report on github in the future!

Yep, I encountered the same problem. Initially it was solely API-based, but as they got locked down and paid-only and removed entirely, I went the bulk download route.

Thanks man! Good to see you here. Nice suggestion, I have a list on a text doc here, I'll add this one :)

Got a lot of feedback here so far, plenty to keep me busy. And there will be an import API to let scripts/programs push data to Timelinize, even if it doesn't support those data sources built-in.

Thanks for the feedback; especially about the name. I'm glad you value the project's goals!

Yeah, big fan of Perkeep. I think it's well-engineered. (Better than Timelinize probably! Just missing some important features for me.)

Thank you for the great feedback!

> You might suggest to users the following use case. “If you want to create a Timelinize data store, but don’t feel that your own local systems are secure enough to safely hold a basket with a copy of every egg in your life, you might consider the following use case, which some of our customers implement. Once our twice a year, update the data store, but store it on an external disk. When the update is done, take the disk offline and keep it in a drawer or safe.”

Sure, I like that.

> I always wondered how cool it would be if I could tell some Spotify-like system, “I’m 20 miles away from the lake, we are going to stay in the cabins a week, just like we did 10 years ago. Play me the exact same songs now that played at every turn back then.”

It's inevitable for an LLM and other tooling like personal assistants to be integrated to this thing, so yeah that sounds like a great use case.

> For a name, how about: ChronEngine That name seems pretty free and clear from previous use, if you like it grab ChronEngine.com before some squatter does and thank me with a phone call, I would enjoy a quick chat with you. ... Oh and make a list of anybody that says they have done an elaborate something similar with Excel (like me and at least one other person in this thread) and maybe have a discussion with them some time, we/they might have some useful insights.

Not bad actually; and sure, we can chat either way. Feel free to book a time at https://matt.chat.

> Your web page might benefit from a grid that lists all the input sources you accept, with hotlinks next to the names that give a pop-up quick summary of what that source is about, and maybe some color coding that shows something like “green = rock solid lately”, “yellow = some users reporting problems”, “red = they just updated their format, it’s broken, we are working on it”. You will/are facing challenges similar to Trillian, a chat client from the early 2000s that would try to maintain ongoing connection compatibility with multiple other chat clients such as AIM/ICQ/MSN. Also, the grid could have a “suggested source sets” filter that helped people find what 5 (for example) input sources they might select for their use style.

I LOVED Trillian, thanks or the nostalgia. Oh man, they're still alive: https://trillian.im/ (I love the icon...)


> Interesting. But this seems to be limited to the supported data-sources?

Data sources are extensible, they basically just need to implement 2 methods: https://pkg.go.dev/github.com/timelinize/timelinize@v0.0.24/...

> No generic format to build your own sources?

I'm designing an import API that will let any script or program push data to the timeline.

> Also, why the focus on a timeline?

Seemed to make the most sense.

> Not all personal data or tasks are time-related.

Probably true. Do you have examples of what you're thinking of?

> If you want to have collaboration on this, maybe think about splitting it into backend and frontends. With the backend being the storage, and it's import-mechanism, and the frontend being open for anyone to put their own view onto the data, and maybe work with them.

That's how it is right now; it's a client-server architecture. It has a CLI and a JSON API so you can build your own front-ends.

> Maintaining personal data has demand here, but so far there is no real unified project catering to a broad range of services and different types of data. So there is a space to fill and a community to build IMHO.

I agree, this is a huge void that needs to be filled!


> Probably true. Do you have examples of what you're thinking of?

Bookmarks, contacts, notes for example, they are usually not time-based but by context, category or whichever organization their usage demands at the moment. They can be time-based in a journal, but usually I do not remember websites or people by the time I encountered them, so I would not search them this way.

The other question is how personal is personal in this project? Are my IMDB-ratings valid data for this? My Ebook-collection? My Steam-account? Those are usually things I would not manage in a timeline or a map, but with different interfaces and features.


Bookmarks are bookmarked at a specific time.

Contacts are already handled as first-class concepts in Timelinize, known as "entities". Entities don't have timestamps, but they have a name and attributes that can be anything, and the relationship between the entity and its attributes can be bounded by time.

Notes are also created and updated/changed at specific times.

> Are my IMDB-ratings valid data for this? My Ebook-collection? My Steam-account? Those are usually things I would not manage in a timeline or a map, but with different interfaces and features.

Could be, if you wanted to timeline those! (Like, when you rated a certain movie. Or when you got or read a certain book. Or when you got a new Steam game.)


Thank you!

And yeah, I feel ya. This lets you delete things too.


Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

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

Search: