Small-Net

by

Small-Net is a social-movement born in the late 2010s and early 2020s focused on using & creating technology that serve and to put the needs of individuals first, before everything else — and in particular:

  • the needs of the individual with respect to his or her active relationship with his or her friends and family, and
  • the needs of the individual with respect to joining and actively participating in one or more (usually smaller) communities.

"Small-Net" is also sometimes written (without the hyphen) as "smallnet", and (with "small" written as "smol") as "smolnet".

Related to small-net is "small-web", which is also sometimes written (without the hyphen) as "smallweb", and (with "small" written as "smol") as "smolweb". "Small-Web", in some contexts, can be be considered a sub-set of small-net; and in other contexts can be considered more or less synonymous with small-net.

Motivation

The small-net social-movement (as well as some other social-movements) was born out of a feeling that the modern (2010s and 2020s era) Internet, Web, and social-media are broken.

Reasons for feeling that the modern (2010s and 2020s era) Internet, Web, and social-media is broken include the ubiquity of individuals:

  • getting spied on,
  • getting tracked,
  • having their privacy taken away from them,
  • getting turned into a product, and sold,
  • getting their online identity taken away from them,
  • getting censored,
  • getting manipulated,
  • getting their source of income taken from them because it suits someone else's interests, and
  • getting controlled.

In addition to this, other reasons for feeling that the modern (2010s and 2020s era) Internet, Web, and social-media is broken include the ubiquity of and obtrusiveness of:

  • advertisements,
  • GDPR cookie consent pop-ups,
  • join-newsletter pop-ups, and
  • allow-notification pop-ups.

Small-Net is partially an attempt to fix these things by using and even creating alternatives to the modern (2010s and 2020s era) Internet, Web, and social-media.

Vintage Protocols

Small-Net enthusiasts are sometimes attracted to vintage computer networks protocols. The vintage computer network protocols that tend to get a lot of attention from small-net enthusiasts include:

  • the finger protocol,
  • the gopher protocol, and
  • the telnet protocol.

With the gopher computer network protocol probably being the most popular of the three among small-net enthusiasts.

Some other vintage computer network protocols that also sometimes get attention from small-net enthusiasts include:

  • the active users protocol (IETF RFC-866),
  • the character generator protocol (CHARGEN) (IETF RFC-864),
  • the daytime protocol (IETF RFC-867),
  • the discard protocol (IETF RFC-863),
  • the echo protocol (IETF RFC-862),
  • the network news transfer protocol (NNTP) (IETF RFC-977, RFC-1036, RFC-2980, RFC-3977, RFC-4642),
  • the post office protocol (POP),
  • the quote of the day protocol (IETF RFC-865),
  • the simple mail transfer protocol (SMTP),
  • the time protocol (IETF RFC-868), and
  • the UUCP protocol.

Vintage Data Formats

Some vintage file data formats have also attracted the attention of small-net enthusiasts; namely:

  • the ANSI text format,
  • the ASCII text format,
  • the Code Page 437 (CP437) text format,
  • the Mattel Aquarius text format,
  • the PETSCII text format,
  • the Teletext text format,
  • the TRS-80 text format,
  • the Videotex text format,
  • the XSL style format,
  • the XSLT style format,
  • the XHTML document format,
  • the XML data format,
  • the ZX80 text format, and
  • the ZX81 text format.

"Modernization"

Although many small-net enthusiasts are attracted to vintage protocols, some have found those vintage protocols difficult to work with and lacking features they feel as important.

A reaction to this is that there have been attempts to create "modernized" versions of these vintage protocols, while trying to avoid the mistakes of the modern (2010s and 2020s era) Internet, Web, and social-media.

"Modernization" can include:

  • including support for Unicode UTF-8,
  • including support for URLs and URIs,
  • including support for hypertext,
  • including support for rich-text, and
  • including support for media.

Retro Protocols

Some retro protocols — that are attempts to "modernize" specific vintage protocols — that have come out of the small-net communities include:

  • the gemini protocol,
  • the mercury protocol,
  • the nex protocol, and
  • the spartan protocol.

With —

  • the gemini protocol being an attempt to "moderize" the gopher protocol, while not enabling the same mistakes of the Web,
  • the mercury protocol more or less being the gemini protocol without TLS encryption,
  • the spartan protocol being an alternative to the gemini protocol that does less "modernizations", and
  • the nex protocol being an attempt to be even simpler than and even fixing some aspects of the gemini protocol.

Retro Data Formats

Some other retro file data formats — that are attempts to "modernize" specific vintage file data formats — that either have come out of the small-net communities or have drawn the attention of people in the small-net communities include:

  • the gemtext format,
  • the json feed format,
  • the markdown format,
  • the nex directory format, and
  • the twtxt format.

Other Protocols and Formats

Although not generally considered part of small-net, these protocols and file data formats are also often of interest to small-net enthusiasts:

  • acct-URI,
  • hash-URI,
  • magnet-URI,
  • the ActivityPub protocol,
  • the ActivityStream protocol,
  • the at-protocol,
  • the BitTorrent protocol,
  • the kademlia protocol,
  • the gnutella protocol,
  • the node to node copy protocol (NNCP),
  • the nostr protocol,
  • the secure scuttlebutt protocol, and
  • the WebFinger protocol.

Small-Net Goals

There are a loose collection of goals that seem to be commonly desired by small-net enthusiasts. These goals in part explain why certain protocols and file data formats are of interest.

Not all small-net enthusiasts desire all of these goals. And some of these goals are even in conflict with each other.

But one does not have to satisfy all of these goals in the same protocol.

And also, noting all these different goals is important to understanding the small-net social-movement, even if some are in conflict.

These goals seem to be (in alphabetical order based on category, and not in order of popularity):

Alternative

  • be an alterative platform for cross-platform application development,
  • be an alterative platform for static files (rather than applications),
  • be an alterative to BBS,
  • be an alterative to the e-mail protocols,
  • be an alterative to the finger protocol,
  • be an alterative to the gopher protocol,
  • be an alterative to big social-media,
  • be an alterative to the telnet protocol,
  • be an alterative to the modern Web,

Archivism

  • be accommodating to servers with intermittent up-time,
  • (also) be available via non-Internet means in a seamless way,
  • be aware of archives,
  • be aware of mirrors,
  • be easy to archive,
  • be easy to mirror,
  • be off-line first,
  • be resistant to censorship,
  • be resistant to Internet domain-name expiration,
  • be resistant to Internet domain-name seizure,
  • be resistant to link-rot,
  • be resistant to server seizure,
  • be resistant to server termination,
  • be resistant to software-rot,
  • be servable from the most common computers — mobile phones, tablets, and laptops,
  • encourage archiving,
  • encourage mirroring,
  • include support for a single complete download of "everything",
  • include support for downloading just the "latest" updates,
  • make archiving automatic,
  • make mirroring automatic,
  • make it so a 'response' can be a full resource for a single request (ex: don't have to go download extra images, pages, CSS, JS, etc),
  • support a single download of many files,
  • support a an archival download of many files,

Better

  • be faster,
  • be more private,
  • be more secure,
  • support 'directories' (of files),
  • support machine-legible semantics,
  • support 'partials',
  • support parameterized 'partials',
  • support 'slots',
  • support the separation of content from navigation/UI,

Fix

  • be a "fixed" version of the e-mail protocols,
  • be a "fixed" version of the gopher protocol,
  • be a "fixed" version of the gemini protocol,
  • fix identity,
  • be a "fixed" version of the NNTP protocol,
  • fix payments,
  • fix the Internet,
  • be a "fixed" version of social-media,
  • be a "fixed" version of the Web,
  • be a "fixed" version of Usenet,

Legibility

  • the protocol should be programmer-legible,
  • the protocol should be simple enough for a competent software developer of 3 to 10 years of experience to be able to create a (simple) working client as a weekend project,
  • the protocol should be simple enough for a competent software developer of 3 to 10 years of experience to be able to create a (simple) working server as a weekend project,
  • the protocol should be simple enough for a competent software developer of 3 to 10 years of experience to successively add advanced features to their client as a series a weekend projects,
  • the protocol should be simple enough for a competent software developer of 3 to 10 years of experience to successively add advanced features to their server as a series a weekend projects,
  • the default file data format to be easy enough to parse for a software developer of 3 to 10 years of experience,
  • the default file data format should be human-legible — i.e., legible by humans who aren't programmers,
  • the default file data format to be writable by the average computer-literate (non-programmer) person using just a text-editor (for some definition of "text),

Lifeboat

  • be a viable lifeboat for evacuees from big social-media,
  • be a viable lifeboat for evacuees from the Web,
  • be a viable lifeboat for micro-blogging,

Modernizations

  • be a "modernized" version of BBS protocol (for some definition of "modernized"),
  • be a "modernized" version of the finger protocol (for some definition of "modernized"),
  • be a "modernized" version of the gopher protocol (for some definition of "modernized"),
  • be a "modernized" version of the telnet protocol (for some definition of "modernized"),
  • include support for content-addressing,
  • include support for hypertext,
  • include support for media (audio, images, video, etc),
  • include support for rich-text,
  • include support for Unicode UTF-8,
  • include support for URLs and URIs,

Nostalgia

  • be nostalgic to those who were on-line (on the Internet, BBS, or elsewhere) in the 2000s
  • be nostalgic to those who were on-line (on the Internet, BBS, or elsewhere) in the 1990s
  • be nostalgic to those who were on-line (on the Internet, BBS, or elsewhere) in the 1980s
  • be nostalgic to those who were on-line (on the Internet, BBS, or elsewhere) in the 1970s

Revive

  • revive ASCII-art,
  • revive ANSI-art,
  • revive blogging,
  • revive the blogosphere,
  • revive blogrolls,
  • revive link directories,
  • revive long-form writing,
  • revive personal home-pages,
  • revive personal wikis,
  • revive self-contained sites,
  • revive something like the (pre-JavaScript pre-CSS) Web of the 1990s,
  • revive 'surfing' — going from page to page or site to site via one page or site linking to another page or site,
  • revive vlogging,
  • revive web-rings,

Perma-Computing

  • be resistant to software-rot,
  • support programming,
  • support setting pixels,
  • support Unix/Linux style piping,

Privacy

  • support anonymous usage,
  • support pseudononymous usage,
  • make spying (that feels ubiquitous on the Web) difficult if not impossible,
  • make tracking (that feels ubiquitous on the Web) difficult if not impossible,
  • prevent spying on what 'pages' you view or read,
  • prevent spying on what 'sites' you access content from (ex: being able to tell what Internet domain-names you resolve),
  • prevent web-bugs,

Take-Over Resistance

  • be decentralized,
  • be difficult to embrace-extend-extinguish (EEE),
  • be distributed,
  • don't make Internet domain-names mandatory,
  • encourage multiple client software implementations,
  • encourage multiple server software implementations,
  • encourage a large number of client software implementations,
  • encourage a large number of server software implementations,
  • make creating client software a good learning exercise for students,
  • make creating server software a good learning exercise for students,
  • make creating your own client software easy to do for a a competent software developer of 3 to 10 years of experience,
  • make creating your own server software easy to do for a a competent software developer of 3 to 10 years of experience,
  • try to make it so multiple client software exists and new client software keeps on getting created, so that a malicious party doesn't have a single point of attack (i.e., a single client) that if they compromise to compromise users' privacy,

Trust

  • avoid certificate authorities,
  • prevent or make 3rd-party content-modification detectable if not very difficult,
  • prevent or make man-in-the-middle (MITM) attacks very difficult,
  • prevent someone else from adding to your trusted keys,
  • prevent someone else deciding what your trusted keys are,
  • prevent someone else from replacing one or more of your trusted keys,

Unobtrusiveness

  • prevent or make unwelcomed advertisements very difficult,
  • prevent pop-ups,

User-Experience

  • get rid of the need for passwords,
  • be able to (also) be viewed from a computer-terminal-emulator,
  • be able to (also) be viewed and used without (necessarily) needing a mouse,
  • have the default file data format be easy enough that an average person could learn to write it using just a text-editor (for some definition of “text”).

Debates

There are several debates within the small-net communities:

Encryption

  • should encryption of communications be mandatory
  • should encryption of communications be optional
  • should encryption of communications be prohibited

Media

  • should it be text only (for some definition of "text")
  • should audio be supported
  • should embedded audio be supported
  • should images be supported
  • should embedded images be supported
  • should video be supported
  • should embedded videos be supported

Old Computers

  • should the default "text" format of a small-net protocol only support the features that the vintage VT05 computer-terminal had — i.e., no blinking, no bolding, no underlining, no reverse text, but also no lower-case letters
  • should the default "text" format of a small-net protocol only support the features of the ASCII data format
  • should the default "text" format of a small-net protocol only support the features that are renderable on most of the popular modern computer-terminal-emulators
  • should we embrace the "best" of old computers without being restricted by (all) their limitations

Rich Formatting

  • is 'bolding' a type of punctuation or styling
  • is 'italicizing' a type of punctuation or styling
  • is 'underlining' a type of punctuation or styling

The Web

  • should a small-net protocol be intentionally incompatible with the Web — except maybe over bridge-proxies
  • should a small-net protocol be at least minimally compatible with the Web — to make it so people have a "gentle" path to using any small-net
  • should a small-net HTML make use of Web-components