joshsharp

Good luck with your launch :)

What actually is it though? What do you do with it? I can't tell.

Still working on https://exist.io, a platform for quantifying and tracking data about your life. We've been around for about a decade now and seen many competitors come and go — this niche is definitely not at the peak of the hype cycle any more. On the other hand, Apple Watches have normalised tracking body stats like activity (and sleep to a degree), and yet we think there's still so much more that can be done with that data. We're still (always) working on improving our insights and correlations to find better ways to use the data to optimise your health, mood, productivity, and so on. But not with "AI" :)

And you're using https://changemap.co for your public roadmap, another incredible tool in its own right ;)

If you like this, you may enjoy this old blog of Street View images as art: https://9-eyes.com/

If you want to do something similar without hosting it yourself, you can try my thing https://exist.io. It has a full API in addition to a bunch of integrations.

Yep, there's an API and import/export.

It just goes read-only. You can still access and export bookmarks.

Now wouldn't it be great if it was viable to make the same decision about distributing your iOS app.

Gluon is also a language built on Rust: https://github.com/gluon-lang/gluon

Cool to see that this is written in Crystal.

We do not need to honour cookies when making requests to the API as we use token authentication. Even if we wanted to keep users logged in within the web view, which we don't because the web view is just for static help pages, we wouldn't want to send those to the API.

Sure, the cookies expire (in two weeks I think, in this case) but that doesn't invalidate the issue that the wkwebview silently shares cookies with nsurlsession, nor does it negate the confusing journey of tracking down the bug. Sheesh.

They shouldn't be logging in via the webview within the app anyway, so that's moot.

We are using cookies, as the post says, as a fallback authentication so devs can browse the API from a browser if they're logged in to the site.

Using the Authorization header.

For context: Exist is a personal analytics platform where we import data from sources like fitness trackers, Apple Health, productivity trackers, email and calendars, social media etc. and find insights and correlations in the data. We've been doing this for about 3 years now, and we've just added the ability to track your own tags, probably our biggest feature release so far. These are just binary yes/no tags you can add to a day for context, like "coffee", "sex", "meditation", or whatever you like. We can then use those alongside your other data to find correlations like "you have a worse day when you track 'nootropic'" (which was the case for me), or "your weight is higher when you track 'eating out'".

It's not really what you're after, but we're some folks trying to compete with Pinboard. I'm sure Maciej once tweeted something about it being a business that could be successful for a couple of indie devs, not a big VC-backed startup, and that's what we are... Although the successful part is still a ways off :)

Anyway, our thing is https://larder.io. Our angle is that we'll sync your github starred repos as bookmarks along with everything else you save.

You'd think so, but a bunch of users who keep happily using Larder without paying for it say otherwise! It seems like plenty of people don't hit that limit. I probably wouldn't either, actually. It's a tricky balance, because it has to be a high enough limit to still be useful (apparent tick) but low enough to encourage upgrading (uh,not enough data yet). It is a free trial and not a free tier, after all — the idea is that if you find you're using it regularly and hitting the limit, then perhaps you might be interested in paying for it.

Sorry this isn't clear. The trial restriction only kicks in after the first day, so that when you first sign up you can go nuts testing it out. I'll add that clarification to the account or upgrade page somewhere.

It's a general bookmarking tool that is focused on developer use-cases.

As I mentioned in another thread, in the future we'll be working on other developer-oriented features like an API, webhooks, maybe Stack Overflow integration or a CLI client, depending on what users want. GitHub sync is just the first big feature :)

No, we're going for straight bookmarking — more Delicious than Evernote.

It's not required, it's an option if you want to sync starred repos.

You can select multiple by hitting "manage" and then using the checkboxes that appear. Sorry that isn't clear.

Search does partial matches for url, description, title etc., but tags are the one thing that do require a full match. If you type '#fu' into the search, for example, you'll get a '#fuck-gfw' suggestion. (I have no idea what GFW is by the way, so I cannot approve or condone this sentiment)

I'll add it to the suggestions in the roadmap.

You're correct, we don't use GitHub for social sign-in. You connect a GitHub account in the next step of the signup process.

Well the reason we're actually charging for Larder rather than pursuing some kind of "growth" strategy is to do our best to avoid going out of business due to lack of funding. We're building a lifestyle business rather than a startup, to be honest, and it's highly unlikely we'll ever get acquired either. We just want to make a living building things and selling them (unusual, I know). So while I can't make any guarantees, what's most likely is that if Larder ever does shut down or go away, it'll be through a decision we make personally, and so we'll provide plenty of time to export your data (which is already possible from the account page, whenever you'd like it) before taking it offline.

I'm not aware of any other bookmarking services specifically tailored towards devs, so I can't speak to that. In terms of what we do differently to other general bookmarking services, our big one right now is the GitHub integration — we sync your starred repos as bookmarks, tag them by language automatically, and also sync related metadata like last release and commit date. I find it really handy to have my bookmarks and stars in one spot, because I save a lot of dev-related resources and I can never remember, did I bookmark an article about that library or did I star its repo? Now I can just look in the one spot for everything. Also, the metadata on last commit etc is very handy for keeping an eye on how active libraries are without having to watch them and wade through the full feed of their issues, etc.

In the future we'll be working on other developer-oriented features like an API, webhooks, maybe Stack Overflow integration or a CLI client, depending on what users want.

We're still quite early on in building it, but the missing things you'd expect like an API and mobile clients are in the works. If you sign up and find you'd like something else that's missing, feel free to vote or suggest things on our public Trello roadmap [1]. We want to keep adding more features that aid developer productivity, so if you have related ideas we're keen to hear them.

[1]: https://trello.com/b/0gw1nPAH/larder-roadmap

One thing I found surprising on reading this is that they essentially spent $10k and two months ($5k listed as saved * 2 months) to figure out they weren't using their logging infrastructure any more. Wish I could be that gung-ho with resources!

Hi Bruce! Good to see Open Dylan getting some attention on HN :)

I think it's really interesting to see the divide between visualisations and "actionable insights" in the discussion here. I'm definitely more about the latter, which has been my focus when building Exist[1] — which you could say is an uglier AprilZero/Gyroscope. AprilZero looks like a magical movie interface (and I feel like Anand has said he was inspired by this sort of thing before). In contrast it's definitely obvious I'm not a designer! So while Exist aggregates a lot of the same sort of services, it's much more about trying to make it work for you — we track rolling averages for all your data points, find simple insights each day (best/worst steps for x days) and automatically correlate everything against everything else to find interesting relations. Correlation != causation, of course, but seeing things like "Your weight is higher after you check into Little Greek Souvlaki" (a genuine result of mine) feels sort of like magic to me, even though with some thought I probably could've figured that out on my own.

Where we still have a way to go is tracking things at a smaller than day-level granularity — I really want to find relations between things that happen around the same time, like checking in to a specific place vs productivity levels, but that level of smarts is still a fair way off I think. Plus the "actionable" part of "actionable insights" is a whole can of worms in terms of computers telling people what to do.

All the same, exciting times.

[1] https://exist.io