I guess that there are "content creators" who are not interested by video or click-bait as well as those "content consumers" who are looking for geniously interesting content written in a concise and clear way. Substack seems a good site for this but in general it seems to me that this is sth that is missing in today's internet.
a couple of decades ago, I was still a lowly applied mathematics (and aspiring programmer) student when i proposed to a professor to change the compiler options from O2 to O3 to gain performance (reducing run time by a significant percentage like 50% or more...) only to be disregarded by saying "no, compiler optimizations break numerical accuracy".
I am really curious to know how I am wrong in that.
As a long time X user and now on bluesky i can tell that there is a huge difference in the content which makes it clear that the platform was pushing accounts with specific political views (or muting others)
shallow knowledge on my side, but it seems that the AT proto requires some fancy relay node that, as of now, are only hosted by big corps, so no one has a a fully independent bsky instance
This seems to be left out of the conversation when comparing Bluesky with the wider fediverse. The AT protocol relies on a bespoke setup and is complicated to implement. ActivityPub is comparatively simple, and can be implemented on pretty every thing with just a few endpoints.
There is not a bespoke setup that you need to implement atproto. In fact, there are already a variety of applications making use of it (some to a higher degree than others). There are community implementations of app views, relays, PLC directories, and PDSes already in the wild, and - although I admittedly have a biased ear on the conversation - developers tend to appreciate the _lack_ of complication when implementing things.
you can run a resequencing relay good enough to feed an atproto appview pretty cheaply - it just needs to subscribe to a few thousand websocket endpoints to get a live tail of the whole network (incoming traffic is well under 50Mbps at peak in my experience)
running an archival mirroring relay is storage-intensive (on the order of tens of TB iirc?) but only serves as an optimization (you can backfill full atproto repositories straight from the relay instead of needing to reach out to the relevant data server)
You cannot run a bluesky instance on your own. The AT protocol isn't really a protocol in the same way that matrix or mastodon is. It's just an API, you know like all platforms used to have?
Please point me to instructions on how to make my own bluesky instance that federates with the existing bluesky, and which does not rely on bluesky's authentication.
Can you prove to us that it is possible to interact with someone on Bluesky without being registered on Bluesky?
Threads use e ActivityPub but we know it's bullshit and that they will shut down the network as soon as possible. Bluesky is the same shit. AT protocol is just a honey pot.
I did not use Bluesky a lot, and just set it up to try self-hosting.
From what I know :
- Right now you can self-host your PDS (Personal Data Server) that hosts your user account basically, and all of your user content (posts, images, etc...)
- The easiest way to register is to use bluesky's frontend, and specifying your PDS address, I didn't try to self-host it too but I would assume if you self-host the bluesky web app, or just do the api calls to your PDS by hand or in another application, registration would work too.
- From what I understood, when you post, it is completely stored on your PDS, and then there are kinds of "mega nodes", (I think they are called Firehoses ?) That aggregate from a bunch of PDS, (and the official bsky.social PDS) and present it to end users. I don't think you can self-host theses just yet.
So, in the end, I think you can absolutely interact on Bluesky without being registered, by installing your own PDS and only making calls to create an account on it and post on it, without having an account elsewhere. However, the "end user" part is still closed off for now, since the Blue Sky Frontend uses *official* aggregators that could theoretically refuse your posts.
Won't your content be deprioritized if you do that though? Or won't it be further down the road? I fail to see any reason for Bluesky to be honest once the user base is big enough and people are silo'ed into it.
You are right, these are what matter and thanks for the suggestions.
I am not particularly on a budget but I don't either require wifi7, I am not willing to pay 3* more than I should.
I am lazy and searching for something that is supported by merlin and I will look into the used market as well.
I'm not convinced that I would choose Merlin, because it only supports a very limited amount of devices. However, I had Asus routers in the past and they were not half bad.
If you don't need USB and could switch to OpenWRT, you could go for a Netgear WAX 202 - I got a used one for 50 bucks, pretty stable so far. Or maybe a TP-Link AX 1800 for 60 bucks new.
Again, I very much agree with you, very sound advice. The reason I stick with merlin/Asus is that I do not want to spend too much time, I want to buy sth, configure/deploy/forget.
if you want top shelf routers, it will have qualcomm chips (cpu/soc/radio). If it is a budget one, it will have mediatek.
Also, we had a flood of demand and new versions coming out. which literary added nothing. Some will wrongly say "mimo", "beamforming", etc... but those were all present since "wifi5" (ac) and have just been "standardized" in wifi6 (ax), wifi6e (ax+6ghz), wifi7 (be)... the real improvement after wifi5 was simply higher frequencies (pointless since your internet pipe is still crap) and more power (by relaxing power limits). really nothing else because even the features that were standardized might not be present or be badly implemented anyway, like half wifi6e routers not even having a 6ghz radio.
i love this blog, i hold Chris Wellons to a very high estime BUT i utterly disapprove the usage of macros so much, especially to wrap cstd types, functions, etc.
I'm currently on a C++ (mostly C with C++ compiler) trip. It does make some things easier and some things harder. It makes it easier to work with C++ developers :-). I sometimes use the more involved C++ features but often regret it after because of complications.
But one thing that makes it worth it is the removal of the struct tag space. I have a strong dislike for the struct tag boilerplate in C, but the alternative -- typedef boilerplate -- in C is unbearable to the point that I have a macro to define structs in C that does this automatically.
#define STRUCT(name) typedef struct name name; struct name
STRUCT(Foo) {
int x;
int y;
};
But macros often come with disadvantages. In this case it's that many IDEs have trouble finding the struct definitions from a usage site.
a manual transmission car does not require more focus and thought as very soon changing gears becomes a reflex, ie it doesn't require the driver's prefrontal cortex. The same can not be said for repetitive work in programming (eg as in programming in assembly or C)
IMO, this opinion deserves some thought even though it is not as bad as it sounds
reply