you know those american hoarding shows where people are like "hell no i'm not throwing out the cardboard box my toaster came in"
running a VPS is the exact opposite of that
"yeah i can host those files on S3, nobody will notice the latency"
"you seriously expect me to store a PNG when WEBP is around 30% more efficient? fuck outta here"
"wait, i'm storing HOW many megabytes of man pages? those have got to go"
at this point i'm so desperate that i'd gladly pay vultr $10/month for the privilege of plugging an external hard drive into the server
yes i know it's a virtual machine but you can do USB passthrough come on i don't care if it's a SCSI 5400RPM piece of shit just give me SOMETHING AAAAAAAAA
i get it. block storage is hard because you need to physically reserve that entire chunk of storage for the customer. i understand.
but surely, surely, if wasabi can give you a TERABYTE of object storage for six bucks a month, i should be able to get a little more room?? come ON just plug a fucking usb stick into it or something
pet decided to go with a hetzner server because it's cheaper and holy shit the delay over ssh and sftp is unbearable
vultr offers block storage BUT only if your server is in NY or NJ because it's "in beta"
you can't provide a service for four goddamn years and say it's still in beta you DUNCES
i swear to god it is IMPOSSIBLE to get an australian server that provides a decent amount of storage for less than US$60/mo
right now i'm paying US$40/mo for a quad core CPU, eight gigabytes of memory, and fucking 160GB of storage
umart is currently selling a 240GB SSD for AU$60. my phone's microSD card has more storage than my goddamn webserver and i am UPSET
i don't like to get overly optimistic about stuff like this like poor old berners-lee but man is it exciting to live in a time when decentralised, distributed networks and protocols like activitypub and scuttlebutt and IPFS and so on are becoming popular
are we witnessing the rebirth of the idea of an open web? i think we are, and the question is how far will it go - will we stop where we are now, or will we keep going, adding new and exciting ways to use the web? i don't think we'll ever reach the "every person has their own website" level, but i have to say, it's amazing to see this happening and it isn't showing any signs of slowing down yet
the semantic web has been called everything to an overly optimistic and highly impractical pipe dream to one of the last vestiges of the concept of an "open web". it's fascinating to read and learn about.
i highly recommend giving this article a read if you want to know more: https://twobithistory.org/2018/05/27/semantic-web.html
the author also has a follow-up article about FOAF in particular: https://twobithistory.org/2020/01/05/foaf.html
the semantic web
another major problem is that there's no true "universal ID" that you can use to link to anything. when you say that you work for company X, how do you know whether it's company X, florida, or an entirely unrelated company X in new zealand? centralised services like facebook can solve this easily - give every person and business a unique internal ID - but you can't really do that on the semantic web. if you're using someone's ICQ number as a unique way to identify them in your FOAF markup, and the other person deletes their ICQ number because they (gasp!) aren't using it anymore, that link will break.
the semantic web
just like there are issues with centralised services (some that i've laid out here, many that i haven't), there are also a lot of issues with the decentralised way the semantic web works.
if someone's website goes down, and nobody has a local copy of the semantic markup that described them, they effectively disappear from the semantic web. in the same way that accounts from dead instances slowly fill up your follower count on fedi, your FOAF markup would contain more and more links to people who only exist in the form of an email address on someone else's website. this is great for the right to be forgotten, but not so great for genuine issues, such as if someone's unable to pay for their web hosting anymore. there are solutions to this, but it's a pretty major problem.
the semantic web
of course, none of this really ever took off. FOAF is for all intents and purposes a failure, very few services use semantic web markup, and those that do sometimes have their own specific formats - for example, you can include twitter specific meta tags on your website to tell twitter what image you want to show as a "card" for your site
the cynical argument is that this proves that the web - or even people as a whole - gravitate towards centralisation, but i don't think that's the whole truth. i think it's a combination of big companies like facebook and myspace who "got there first", the relative complexity of writing your own semweb markup compared to typing your name into a facebook registration form, and perhaps the biggest factor: the fact that, unlike in the late 90's, most people don't have or want their own personal webpage. you don't need to pay server hosting fees or domain registration to create a facebook account.
the semantic web
friend of a friend (FOAF) is a machine readable language for describing your network of friends. much like on facebook, where you can add people to your friends list, you could use FOAF markup on your website to provide a list of people you're friends with, linking to them with a unique ID (email address, phone number, PGP fingerprint...) to show that you're friends, or at least, that you're friends with them
in 2007, berners-lee published an essay where he envisioned the so-called "Giant Global Graph" (GGG) of the semantic web becoming just as important as the world wide web - that "www" and "ggg" would become equally as important. in what is perhaps my favourite quote by someone overenthusiastic about technology, he said "I express my network in a FOAF file, and that is a start of the revolution."
the semantic web
in 1999 (yes, that long ago), tim berners-lee said:
"I have a dream for the Web [in which computers] become capable of analyzing all the data on the Web – the content, links, and transactions between people and computers. A "Semantic Web", which makes this possible, has yet to emerge, but when it does, the day-to-day mechanisms of trade, bureaucracy and our daily lives will be handled by machines talking to machines. The "intelligent agents" people have touted for ages will finally materialize."
around a dozen years later - a long time in terms of technology and especially the internet - we have these "intelligent assistants", but they don't work through the semantic web - instead, they use central repositories of information like facebook and wikipedia and youtube, and if you don't appear on there, your voice assistant won't know you exist
when i update (for example) my address on lynnesbian.space, that update carries across to my social networking account, my online shopping, my contact card in your phone, and so on - the idea of updating several dozen web profiles if you move to another country is no longer necessary
it's a well known fact that a LOT of websites handle things like names (thou shalt have a distinct first and last name) and gender (thou shalt check thine binary box) pretty poorly, but in the world of the semantic web, this wouldn't be a problem - if the standard you're using allows anything to be typed into the gender field, websites advertising compliance must support it
the semantic web is a really interesting idea: you include machine readable data on your personal website, and that data can be parsed and understood by "intelligent agents" (think siri, google assistant) as well as other websites that provide functionality like social media and shopping and appointments and so on
let's say i join a new social networking platform. when i sign up, instead of having to fill out a long form with dozens of questions about my name, date of birth, location, and so on, i just tell the website to pull my data from lynnesbian.space, which contains all of that information already
i can't find a link to it now, so idk if it's true, but i remember hearing a story about a sorting system that was used for (i think) a student radio website that would first turn a name into the "library version" by replacing e.g. "the cronchers" with "cronchers, the", and then sorted them alphabetically
this was all well and good until they tried to add a song by a band called "the, the" - the system got stuck trying to convert the name so that it didn't start with "the", forever swapping the order of the two "the"s in the name to no avail
leftist polyam transbian linux nerd. cute bune (kin). vry gay
please CW drugs and selfies and sex stuff
i have a blog: https://bune.city
@email@example.com's anti-chud pro-skub instance for funtimes