Tootfinder

"#ActivityPub is at an interesting place. Since its intial adoption by #Mastodon in 2017, the standard has become all but ubiquitous across the #Fediverse. In less than seven years, it has become the mo…

The Efforts to Extend ActivityPub
ActivityPub has fostered a social web for millions of people. To grow, some developers believe it needs improvement. These are their efforts.

there's probably other DHT-like tables besides the standard XOR metric that torrents use.
like, if you're literally doing web-of-trust stuff, small network stuff applies right?
high clustering coefficient, low path length?
so something that's good at getting you a good-enough path length of the local group is probably fine. Doesn't need to be global.
could probably still have the exponential backoff drop. or maybe faster?
distance tables:
direct friends
friend 1 (store all)
friend 3 (store... half?)
friend 7 (store... a quarter? an eighth?)
friend 15 (there's a lot here, store few)
https://web.stanford.edu/class/cs124/lec/socialnetworks21.pdf

In bing's search results, website icons are often not shown for unclear reasons, even if they're properly referenced, in the right fileformat, and at standard locations. According to some online forums, the solution is to write to Microsoft support. Like... what? (I mean... it appears it's a working solution, but... How's this even possible? Do they fix the bug one by one per web host?)

Interesting stuff...
(the original opt-out idea was nuts... it had to cause a major storm. Becoming opt-in, even if discoverability isn't all that well implemented, I don't think I'll mind.
That being said, Bluesky should simply implement ActivityPub. That's the goddamn W3C standard, for crying out loud!)

Bluesky and Mastodon users are having a fight that could shape the next generation of social media | TechCrunch
As decentralized social networks become more popular, the way different protocols interact could set the stage for the future of the web.

You know, I remember using both JPEG and GIF in the mid/late 1990s.
1998 Kyle would be incredulous if someone stated that the same image formats stick around for more than a quarter century.
https://hackaday.com/2024/04/07/jpegli…

Jpegli: Google’s Better JPEG And Possible Death Knell For WebP
Along with the rise of the modern World Wide Web came the introduction of the JPEG image compression standard in 1992, allowing for high-quality images to be shared without requiring the download o…

I think that popular "fullstack" frameworks should provide better support for modern web security features, such as:
- Content-Security-Policy
- SubResource Integrity hashes
- Trusted Types (I know this one is not standard yet)
I finally managed to implement those features in a site I'm working on... but it has been quite an adventure to do that in "userland". The right place for that kind of logic is inside the framework.
I don't know w…

I think that popular "fullstack" frameworks should provide better support for modern web security features, such as:
- Content-Security-Policy
- SubResource Integrity hashes
- Trusted Types (I know this one is not standard yet)
I finally managed to implement those features in a site I'm working on... but it has been quite an adventure to do that in "userland". The right place for that kind of logic is inside the framework.
I don't know w…

Looking at the beautiful standard for "normal-gauge traction vehicles, classification and marking", I'd like to point out one more historical curiosity.
Electric multiple units are classified as designed "for local and suburban traffic, on lines with low and high platforms" (EN) and "for suburban traffic, on lines with high platforms" (EW) [1].
Let's note that the standard dates back to a time when high-floor railcars were a norm. According to Wikipedia, EN57's "floor is 1153 mm above top of rail" [2]. For these EMUs, the difference between "high-platform" and "low-platform" units boiled down to the latter having stairs leading down. So both kinds of units could be used on high-platform lines.
Most of the modern EMUs are low-floor, and are not suitable for high platforms. Therefore, they technically do not meet the definition for "EN" marking. For example, the Newag Impuls trains ordered by SKM Trójmiasto (the urban train operator) had their train floor height specially raised to 960 mm ATR [3].
[1] #rail

Today you can find three systems of designations for rolling stock in Poland:
1. Using 1969 norm.
2. Manufacturer's markings.
3. European Vehicle Number.
Sometimes markings 1. or 2. are used depending on the train operator. For example, the 4-car Impuls train could be marked ED78 (sometimes EN78) or 31WE.
The first kind of marking includes stock type (E — electric, D — multiple unit for long-distance trains), construction number along with the number of cars and electric network type (70–79 range is for 4-car units for 3 kV DC network). Unfortunately, the norm doesn't account for many modern EMU types, so we have assignments that are "stretched" or out-of-range: 2-unit Elf is EN96 (according to the norm, that range is for different voltage), 8-unit Dart is ED161.
The second kind of marking is basically "free for all". Newag and Pesa mark their EMUs using successive numbers, with "WE" suffix (for "wieloczłon elektryczny", or a weird way of saying EMU). The 2-, 3-, 4-, 5- and 6-car Impuls trains are respectively 37WE, 36WE, 31WE, 45WE and 35WE. The same Elf trains are respectively 34WE, 21WE, 22WE, 48WE and 27WE.
EVN marks specific cars rather than the whole train. Let's take at the marking of one of ED78-001 cars:
94 51 2 140 430-1
It includes the stock type (94 — EMU with velocity < 250 km/h), country of origin (51 — PL), purpose (2 — passenger train), type of electric network or transmission (1 — electric, one system), power (4 — 1.5 MW to 2 MW), car number (0430) and a control digit. In my opinion, it's harder to remember, and you don't get proper distinction between different train makes (for example, 0429 is an Elf train, 0430 is Impuls, 0703 Elf again…).
How do Newag hybrid trains fit in all this? The "pantograph-enabled" version, 36WEh is also marked EN63H — i.e. it is an EMU with non-standard suffix. The EVN is 90 51 2 42… (90 — "different tractive stock", 4 — 2 systems, electric diesel).
The 36WEhd version "without a pantograph" is also known as SA95 — a diesel railbus outside normative number range. EVN's 95 51 2 62… (95 — diesel multiple unit, 6 — electric transmission).
#rail