git ssb

2+

dust / spaceship



Commit b9c9a521ba683f500d3db7587a88ce82a4793431

Browse Files

broke rationale out of README, added content

dust committed on 4/22/2016, 5:34:21 AM
Parent: bd86a8875815b270325815e589712004d937767c

Files changed

README.mdchanged
rationale.mdadded
README.mdView
@@ -17,184 +17,15 @@
1717 … - [spaceship-engines](./spaceship-engines.md): a document describing the basic
1818 … "API-like" functions needed to realise the schema.
1919 … - [ ] TODO: reference engine implementations in:
2020 … - [ ] scuttlebot
21- - [ ] swarmbot
22- - [ ] tor
21 …+ - [ ] hyperdrive/swarmbot
22 …+ - [ ] onionspace/tor
2323 … - [ ] hydra
24- - [ ] ipfs
24 …+ - [ ] orbit-db/ipfs
2525 … - ??
2626 …
27-## motivation
2827 …
29-### the ghost of cyberspace
30-
31-the existing internet's not really space. it's not really a place you can
32-travel.
33-
34-"the" internet, the one ruled by ICANN, a host of peering and consumer-facing ISPs who play
35-resource games with each other, certificate
36-
37-this internet that is talked about these days is a small sliver of all of the
38-other things the initial decentralised network project is and was, even at that
39-time. that sliver used to be called the "world wide web". it is no longer
40-worldwide--it is balkanised and corporatised, it is regulated and stratified,
41-and its hierarchy is incredibly firm. you know that some things will be hidden
42-and some will be visible when you v
43-
44-and that's just it--"the" internet, as it is now, is a collection of
45-financialised behemoths (not even metaphorically, this is the age of the server
46-farm) who package up code and content to serve to consumers. sometimes they're
47-"content creators", but they don't benefit the way an actual "creator"
48-might. this line was written in 2016 CE and we're still digital sharecropping.
49-
50-even if you're not a marxist (or a telekommunist ;), you would probably like to
51-own what you create, right? this is literally a world where people work for
52-scraps (or sometimes for nothing at all) doing things like complete menial
53-videogame tasks, write articles mainly designed to steal attention (often
54-instead of actually informing),
55-
56-this kind of network is not much of a space, and certainly not a safe space to
57-be. if it's any sort of space at all, it's like the railsea of china mieville's
58-novel: it approximates an ocean, but all you can do is switch rails (and
59-companies and governments are out to put a price tag and patrol on every inch of
60-it).
61-
62-with the advent of increasingly decentralised, private, independent (in the
63-sense of capacity) communications, we've been given a chance to revisit the
64-conception of 'cyberspace' as a separate performative domain.
65-
66-actually, you know what? scratch the 'authoritative' passive voice. here is what
67-happened: a bunch of really thoughtful people did an unbelievable amount of work
68-for free and gave it away honestly. that continuing work, in the form of
69-protocol designs, code, tests, server time, commentary, conversation, and jokes,
70-creates new spaces the way the internet did before existing social/economic
71-empires and edifices colonised it.
72-
73-we've taken it calling it 'cypherspace', since cryptography is one of the
74-pillars on which it rests.
75-
76-cypherspace is a place whose entryways and underpinnings lie outside of
77-monolithic server farms, name authorities, global singletons, or any other
78-hierarchy dwarfing any of the people who visit or inhabit it. its limits are
79-defined solely by the choices of those who work within it, and the consent of
80-those around them.
81-
82-while all of this is great news, decentralisation, and the independence that
83-lies beyond it, can't be a purely network layer movement--it needs to happen at
84-the application layer also. it needs to put people at its center.
85-
86-
87-in order to do this, we need to re-evaluate some old architectural assumptions.
88-
89-
90-### old assumptions
91-
92-what the "server" has to offer in the client-server model is meaningless in a
93-decentralised space. basically, a server only gets you a few things:
94-
95-- uptime
96-- processing power
97-- packaged data (in other words, an API)
98-- a location (in some field or geography of reputation, throughput, legal
99- regulation)
100-
101-decentralised systems typically provide these to every person within them. the
102-actual requirements behind uptime (availability, eventual consistency) are
103-provided by asynchronous and aggressively egalitarian propagation protocols,
104-like gossip or torrent. processing power is provided by inexpensive
105-general-purpose hardware (notably, driven by open standards). typically, a
106-person who has guaranteed access to a computer has more computing power than
107-they can use.
108-
109-the place of APIs, data packaging, and location in a decentralised space is the
110-exactly what the spaceship design aims to address.
111-
112-in a spaceship, the components of the imbalanced client-server model that are
113-kept out of your control are transformed into components that sit inside your
114-spaceship.
115-
116-- engines instead of APIs
117-- galaxies instead of application stacks, server clusters, and other large-scale
118- structure
119-- spacecraft instead of browsers
120-
121-#### engines v. APIs
122-
123-recall an API is basically a public commitment by a service provider to do a
124-certain set of things if you give it the right request. usually this is "package
125-data", but since the results can range from "update record" to "audit this
126-human's state of mind" to "kill somebody", the whole "data" thing has gotten
127-loose from the purely informative conception. (and anyway, physicists recognise
128-this as a purely conventional distinction.)
129-
130-but there's no provider in a decentralised space! there's just information in a
131-space, and you can either see it, or you can't. (and with an async request, you
132-can just plan to do whatever is needed in the eventuality.) moreover, what you
133-do with the information that's been shared with is honestly up to you.
134-
135-in that case, what you need is some programming that moves you through data (or
136-moves data thru you ;). that's what an engine is. you may say "get me another
137-page of this", or you may say "warp 8 to the happening jams"; it's all about the
138-mechanisms you want to build in.
139-
140-saying that, you can expect a fair number of common types and mechanisms to suit
141-a lot of needs, so, just like other vehicle engines, there will be mostly a
142-collection of standard configurations (themselves constructed of software
143-modules as usual), with a host of tweaks to get what you want out of them.
144-
145-
146-#### galaxies v. conventional resource space
147-
148-as i mentioned in the motivation section, the network stack that runs "the
149-internet" is one of many sets of conventions and transmission media, and it is a
150-centralised one, with very little actual choice, just a collection of mostly
151-profit-oriented [dividuations](link here).
152-
153-- twitter architecture changes such as "verified" profiles and algorithmic
154- suggestions
155-- facebook UI being used for emotional experiments
156-- snapchat failing completely at its sole security promise
157-- CDNs ghettoising requestors
158-- an overall trend of
159- "[bullshit minimalism](http://idlewords.com/talks/website_obesity.htm)"
160- obscuring a swath of throwaway "single page" javascript, written for the
161- primary purpose of providing data to markup
162-
163-in most of the above cases, basically the only recourse you have is to be a
164-complaining consumer. you're not even a customer, because you don't have a
165-paying arrangement (a condition for enfranchisement in a market-controlled
166-system). and anyway, making a lot of noise in the same chamber controlled by the
167-authority you're attempting to sway, or making insignificant consumption choices
168-at the mercy of whatever "app" is trendy at the moment does nothing to actually
169-increase your agency in any way.
170-
171-it's not about "network effects". most of the hand-wringing over the implied
172-market capture is a neurosis created by capitalist winner-take-all
173-systems. you+your real friends is a good enough "market share" for anybody. the
174-options just haven't been good or flexible enough yet.
175-
176-
177-#### spacecraft v. browsers
178-
179-to be honest, these days a browser is a cross between a television and a cold
180-war european border. its default use case is "consume content", and getting data
181-out of it for your own use (as opposed to sending it to another consumer outlet)
182-is a complex process of organising scripts to avoid falling afoul of CSP,
183-exfiltrating data through a URI, and parsing it once delivered to you. even
184-though you can see it and hear it through the browser window, you're not allowed
185-to make it yours (for your own protection).
186-
187-
188-
189-
190-a spacecraft has privacy and safety on the inside. yes, outside of it is an
191-environment hostile to nearly all carbon-based life, with crazy radiations going
192-everywhere. and yes, if you spring a leak, the privacy goes right out. but all
193-of that is true already.
194-
195-and in a spaceship, you can go **anywhere**.
196-
19728 … ## prototypes and examples
19829 …
19930 … - [git-ssb](https://github.com/clehner/git-ssb)
20031 …
rationale.mdView
@@ -1,0 +1,181 @@
1 …+# rationale
2 …+
3 …+## motivation
4 …+
5 …+### the ghost of cyberspace
6 …+
7 …+the existing internet's not really space. it's not really a place you can
8 …+travel.
9 …+
10 …+"the" internet, the one ruled by ICANN, a host of peering and consumer-facing ISPs who play
11 …+resource games with each other, certificate
12 …+
13 …+this internet that is talked about these days is a small sliver of all of the
14 …+other things the initial decentralised network project is and was, even at that
15 …+time. that sliver used to be called the "world wide web". it is no longer
16 …+worldwide--it is balkanised and corporatised, it is regulated and stratified,
17 …+and its hierarchy is incredibly firm. you know that some things will be hidden
18 …+and some will be visible when you v
19 …+
20 …+and that's just it--"the" internet, as it is now, is a collection of
21 …+financialised behemoths (not even metaphorically, this is the age of the server
22 …+farm) who package up code and content to serve to consumers. sometimes they're
23 …+"content creators", but they don't benefit the way an actual "creator"
24 …+might. this line was written in 2016 CE and we're still digital sharecropping.
25 …+
26 …+even if you're not a marxist (or a telekommunist ;), you would probably like to
27 …+own what you create, right? this is literally a world where people work for
28 …+scraps (or sometimes for nothing at all) doing things like
29 …+[complete menial videogame tasks](), write
30 …+[articles mainly designed to steal attention](http://www.theguardian.com/media/2016/apr/10/twitter-ev-williams-medium-content-fast-food)
31 …+(often instead of actually informing),
32 …+
33 …+this kind of network is not much of a space, and certainly not a safe space to
34 …+be. if it's any sort of space at all, it's like the railsea of
35 …+[china mieville's novel](https://en.wikipedia.org/wiki/Railsea): it approximates
36 …+an ocean, but all you can do is switch rails (and companies and governments are
37 …+out to put a price tag and patrol on every inch of it).
38 …+
39 …+with the advent of increasingly decentralised, private, independent (in the
40 …+sense of capacity) communications, we've been given a chance to revisit the
41 …+conception of 'cyberspace' as a separate performative domain.
42 …+
43 …+actually, you know what? scratch the 'authoritative' passive voice. here is what
44 …+happened: a bunch of really thoughtful people did an unbelievable amount of work
45 …+for free and gave it away honestly. that continuing work, in the form of
46 …+protocol designs, code, tests, server time, commentary, conversation, and jokes,
47 …+creates new spaces the way the internet did before existing social/economic
48 …+empires and edifices colonised it.
49 …+
50 …+we've taken it calling it 'cypherspace', since cryptography is one of the
51 …+pillars on which it rests.
52 …+
53 …+cypherspace is a place whose entryways and underpinnings lie outside of
54 …+monolithic server farms, name authorities, global singletons, or any other
55 …+hierarchy dwarfing any of the people who visit or inhabit it. its limits are
56 …+defined solely by the choices of those who work within it, and the consent of
57 …+those around them.
58 …+
59 …+while all of this is great news, decentralisation, and the egalitarian agency
60 …+that lies beyond it, can't be a purely network layer movement--it needs to
61 …+happen at the application layer also. it needs to put people at its center.
62 …+
63 …+
64 …+in order to do this, we need to re-evaluate some old architectural assumptions.
65 …+
66 …+
67 …+### old assumptions
68 …+
69 …+what the "server" has to offer in the client-server model is meaningless in a
70 …+decentralised space. basically, a server only gets you a few things:
71 …+
72 …+- uptime
73 …+- processing power
74 …+- packaged data (in other words, an API)
75 …+- a location (in some field or geography of reputation, throughput, legal
76 …+ regulation)
77 …+
78 …+decentralised systems typically provide these to every person within them. the
79 …+actual requirements behind uptime (availability, eventual consistency) are
80 …+provided by asynchronous and aggressively egalitarian propagation protocols,
81 …+like gossip or torrent. processing power is provided by inexpensive
82 …+general-purpose hardware (notably, driven by open standards). typically, a
83 …+person who has guaranteed access to a computer has more computing power than
84 …+they can use.
85 …+
86 …+the place of APIs, data packaging, and location in a decentralised space is the
87 …+exactly what the spaceship design aims to address.
88 …+
89 …+in a spaceship, the components of the imbalanced client-server model that are
90 …+kept out of your control are transformed into components that sit inside your
91 …+spaceship.
92 …+
93 …+- engines instead of APIs
94 …+- galaxies instead of application stacks, server clusters, and other large-scale
95 …+ structure
96 …+- spacecraft instead of browsers or "apps"
97 …+
98 …+#### engines v. APIs
99 …+
100 …+recall an API is basically a public commitment by a service provider to do a
101 …+certain set of things if you give it the right request. usually this is "package
102 …+data", but since the results can range from "update record" to "audit this
103 …+human's state of mind" to "kill somebody", the whole "data" thing has gotten
104 …+loose from a purely informative conception. (and anyway, physicists recognise
105 …+this sort of distinction as purely conventional.)
106 …+
107 …+but there's no provider in a decentralised space! there's just information in a
108 …+space, and you can either see it, or you can't. (and with an async request, you
109 …+can just plan to do whatever is needed in the eventuality.) moreover, what you
110 …+do with the information that's been shared with is honestly up to you.
111 …+
112 …+in that case, what you need is some programming that moves you through data (or
113 …+moves data thru you ;). that's what an engine is. you may say "get me another
114 …+page of this", or you may say "warp 8 to the happening jams"; it's all about the
115 …+mechanisms you want to build in.
116 …+
117 …+saying that, you can expect a fair number of common types and mechanisms to suit
118 …+a lot of needs, so, just like other vehicle engines, there will be mostly a
119 …+collection of standard configurations (themselves constructed of software
120 …+modules as usual), with a host of tweaks to get what you want out of them.
121 …+
122 …+
123 …+#### galaxies v. conventional resource space
124 …+
125 …+as i mentioned in the motivation section, the network stack that runs "the
126 …+internet" is one of many sets of conventions and transmission media, and it is a
127 …+centralised one, with very little actual choice, just a collection of mostly
128 …+profit-oriented [dividuations](http://p2pfoundation.net/Dividuation).
129 …+
130 …+here are some examples:
131 …+
132 …+- twitter architecture changes such as
133 …+ ["verified" profiles](http://anildash.com/2013/03/what-its-like-being-verified-on-twitter.html)
134 …+ and
135 …+ [algorithmic suggestions](http://www.forbes.com/sites/theopriestley/2016/02/06/twitters-algorithmic-timeline-switch-is-all-your-own-fault/#796e152331b6)
136 …+- facebook UI being used for
137 …+ [large-scale emotion manipulation experiments](https://www.theguardian.com/technology/2014/jun/30/facebook-emotion-study-breached-ethical-guidelines-researchers-say)
138 …+- snapchat failing to live up to
139 …+ [the privacy feature that constituted its sole offering](http://www.networkworld.com/article/2999980/security/snapchat-now-has-the-rights-to-store-and-share-selfies-taken-via-the-app.html)
140 …+- CDNs ghettoising requestors by
141 …+ [forcing them to perform repeated labor to view content if they come from a suspected VPN or Tor exit IP](https://trac.torproject.org/projects/tor/ticket/18361)
142 …+- an overall trend of
143 …+ "[bullshit minimalism](http://idlewords.com/talks/website_obesity.htm)"
144 …+ obscuring a swath of throwaway "single page" javascript, written for the
145 …+ primary purpose of providing data to markup
146 …+
147 …+in most of the above cases, basically the only recourse you have is to be a
148 …+complaining consumer. you're not even a customer, because you don't have a
149 …+paying arrangement (a condition for enfranchisement in a market-controlled
150 …+system). and anyway, making a lot of noise in the same chamber controlled by the
151 …+authority you're attempting to sway, or making insignificant consumption choices
152 …+at the mercy of whatever "app" is trendy at the moment does nothing to actually
153 …+increase your agency in any way. worst of all, it reinforces a frame where
154 …+attention, oration, rhetoric, and other tactical expressions are the main tools
155 …+of change, where "social capital" is yet another resource to be accumulated.
156 …+
157 …+it's worth it to mention that "network effects", a supposed driver of monolithic
158 …+centralised services, are meaningless at the scale of lived experience. (almost
159 …+by definition.) most of the hand-wringing over the market capture implied by
160 …+that concept is mostly a neurosis created by capital-dependent, winner-take-all
161 …+systems. "[you and yours](https://en.wikipedia.org/wiki/Friend_of_a_friend)" is
162 …+a good enough "market share" for a fully-lived experience. the options just
163 …+haven't been flexible or satisfying enough yet.
164 …+
165 …+
166 …+#### spacecraft v. browsers
167 …+
168 …+to be honest, these days a browser is a cross between a television and a cold
169 …+war european border. its default use case is "consume content", and getting data
170 …+out of it for your own use (as opposed to sending it to another consumer outlet)
171 …+is a complex process of organising scripts to avoid falling afoul of CSP,
172 …+exfiltrating data through a URI, and parsing it once delivered to you. even
173 …+though you can see it and hear it through the browser window, you're not allowed
174 …+to make it yours (for your own protection).
175 …+
176 …+a spacecraft has privacy and safety on the inside. yes, outside of it is an
177 …+environment hostile to nearly all carbon-based life, with crazy radiations going
178 …+everywhere. and yes, if you spring a leak, the privacy goes right out. but all
179 …+of that is true already.
180 …+
181 …+and in a spaceship, you can go **anywhere**.

Built with git-ssb-web