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

Is it really decentralised? What server can I sign up to? Apart from bsky.app or am I overlooking something?


https://github.com/bluesky-social/pds You can run your own server if you like

There are also 3rd party algorithms, moderation, and interfaces you can mix and match


Also bsky doesn't suppress links to other sites the way X, TiKTok, Facebook and many other networks do. That's a more basic "federation" practice because it lets you

https://www.theverge.com/2023/10/23/23928550/posse-posting-a...


There are also a number of projects that bring comments on ATProto back to the original content

Do you think WhiteWind fits into the POSSE model because your content lives in your ATProto database?


In my mind POSSE is about loose instead of close integration.

If I understand what's going on with WhiteWind it's taking advantage of another dimension of federation which the "fediverse" doesn't address, which is identity. It seems it's a blog service that is linked to my ATProto account that anyone can use so long as they have one.

I'll make the case that the fediverse style of federation has a low level of trust for this sort of things. If I log on to a service that uses Facebook or Bluesky or whatever for identity, there's always the risk that the people who run the service can let somebody else log in. I've got some trust in mastodon.social but plenty of small server administrators in the fediverse have demonstrated that they can't be trusted.


I was thinking more that they have a custom lexicon, so you could create and view WhiteWind records without using their app at all. That your content lives on the ATProto network, and more specifically in your own database which you can migrate and/or run yourself. This would seem to align with your loose coupling line of thinking

https://github.com/whtwnd/whitewind-blog/tree/main/lexicons/...


So if I am a bsky.social user the content gets stored in their database, and if I have a PDS the content gets stored in my PDS?


Yep. If you look at the repo browser for @bnewbold.net, you can see his blogposts: https://atproto-browser.vercel.app/at/did:plc:44ybard66vv44z...


My understanding is that your content gets stored in your database, regardless of PDS host. Bluesky hosts most PDS

I believe each user gets a dedicated sqlite database, or at least a dedicated repo. I recall reading somewhere about the implementation moving to use sqlite


"My" instance of sqlite that they host?


yeah, that is my understanding, if I understand your question correctly.

Of course it is all hidden from view, there is a merkle tree involved, and you probably want to use a higher level client than SQL

This talks about why SQLite and not Postgres

https://github.com/bluesky-social/atproto/discussions/2287

Repository spec:

https://atproto.com/specs/repository

Small program to export:

https://github.com/bluesky-social/cookbook/tree/main/go-repo...


Some more interesting breadcrumbs (what does it take to implement a repo)

https://github.com/bluesky-social/atproto/discussions/2644


Cool, didn't know that.. PDS sounds interesting.


There is a community ATProto discord to get help if you need it

https://discord.gg/VweB8gn5


yup, it seems like it is more mature than last time I looked. Going to be spinning up my own when I get a chance


It’s not really decentralised. You could run your own server, but everyone is on their server so they can pull the rug anytime they want.


There are over 2700 independent PDSes.

https://github.com/mary-ext/atproto-scraping

The number seems to double every few months. It's about as decentralized as Mastodon with its handful of servers controlling most of the network and a long tail of small and single user independent instances. The only difference is recovering from bluesky's official instance (bsky.app) dying is you would have to connect to a different Bluesky instance or convert your PDS data into a website, or port it to some other social media software like Mastodon. I wouldn't be surprised if there are already projects underway to make it possible.


Mail app (part of macOS)

Folder per research topic for a bit of structure.

In the Todo folder, 1 email is 1 todo. Reply can be used to change the todo or add more information. When task is finished then email is moved by replying done to the email and rules will take care of the rest.

With Safari I can email an article to myself that is basically only the article and not the whole website.

Search in Mail app is amazing.

I have ~20k emails now in my notes folders.

Also part of many mailing lists. Discourse forums often have mailing mode. Mail is basically my second brain. I find things that are important quicker in Mail then via Google. Rust mailing list (via discourses mailing list mode) is a life saver. Beauty of it is, all emails are on my disk so it also works offline.

Total emails in my mailboxes come close to ~500k now.

Also from other places I can send notes to myself with special email addresses.


pretty cool. I don't use it like that, but can see it being useful.

I like Safari send to mail feature too.

I made small Mail app macOS extension called alto.mail. Thinking what features can I add to make it even more useful.


this is the most sophisticated workflow i've seen, but if it works that's great, if not what else do you wish for sometimes would be there but is not?


Ralph Waldo Emerson – ‘Fiction Reveals Truth That Reality Obscures.‘



I was inspired to write this post by this one lol. I don't read Hacker News and my friend just sent me the link to the post you mentioned and told me that it's similar to what I do) But actually apps have a lot of differences. I built it for myself first and it helped me a lot, so I decided to share it with the world


That was a different Android application, this one is for iOS. Unfortunately Screen time APIs that Apple provides are rather limited compared to Android, but the app still does it job by preventing you from spending too much time on social media.


I lost 30 kilograms in 3 months.

I stopped eating all grains/wheat, potatoes, corn, and rice. Also trace amounts. So 95% of the supermarket I can’t eat and skip while doing the groceries. After a while it’s easy.

Plus I walk 5 kilometers everyday. It’s just 1 hour. I listen to a podcast. In the weekend more when time and conditions allow it.

I need to update my blog. But you can follow my journey if you want. Username + .com


https://journeytolunar.com

Follow my epic journey of 384400 km (distance to the moon) by walking, hiking, biking, running, swimming, rowing, and more...

1008.45 km / 384400 km (0.2623%)

I need to update the stats. It has been a month and I am a bit further now. Also trying to automate it a bit. Building my own distance tracking iOS app.


Well. Where is your blog?


I used to have one listed in my profile but I removed it for lack of interest and nuked the VM it was running on. I maintain it locally and may put it back online some day.


Hike in every country and territory of the world. After that is done I deserve some rest…


Absolutely, it is very impressive that it has been preserved for so long. (Since we humans tend to destroy stuff…) And it is the book that is most printed and distributed around the world. Plus available in the most languages of all books. That alone makes it the best book.


It’s only turned off in the US right? Would they use the data in the rest of the world since SpO2 data is available there?


The SpO2 data isn't useful for detecting sleep apnea because the sampling frequency on a smartwatch is way too coarse to pick up on it.


oh that's too bad. The O2Ring samples every 2 or 4 seconds but a lot of cheaper devices take an average reading once a minute or so which definitely obscures sleep apnea. (normal adults take a breath every 3-5 seconds when sleeping. with obstructive sleep apnea that can be more like once a minute.)


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

Search: