Operating Space post archive

I started this forum to support my blog.

http://overflow.space
https://jasonsackey.wordpress.com/

That's some of the sort of stuff I'll be posting here. I've mostly focused on software-related topics, because that's the field I work in, but that not supposed to limit the forum topics here in any way.

There's one main 'Technology' subforum. I'll sort out a more fine-grained structure later, when there's a bunch of content that needs sorting out.

I considered dividing the forum into Software and Hardware, but then where would a general 'VR' or 'Apple' thread go? I also considered Computers + Other tech. And Peace and War.

For now, with whatever information you wish to contribute, make threads as general or specific as you wish. A thread can be about a company or a product class or a particular product. A programming language or a programming paradigm.

I started to use some comment threads on my WordPress blog as linkblogs. I'll do that here now.

To do:

Make a frontpage for this site that includes articles/blog posts plus some sort of feed linking to latest forum content. Maybe this'll use WordPress, maybe not. The RSS widget for wordpress.com doesn't cut it.

• ☐ Make a front page - blog + forum activity
• ✓ Set a maximum forum content width
• ☐ Write introduction to the site (in progress!)

---

29/6:

• ✓ Remove date of birth fields from user profiles

Towards automating crossposting to here from Twitter. OS threads should be able to 'follow' twitter threads. Here's my proposed workflow:

- I tweet about internet ads
- I link that tweet on the OS online advertising thread
- Now my OS thread is 'following' the twitter thread
- My OS thread automatically gets my tweet replies crossposted
- (other folks' replies don't, not without opting in...)
- stats on replies and retweets get mirrored into the thread

So we could use a bot on this forum, one with a special capability for marking its posts with certain metadata. In the UI, tweet posts should be visually distinct from regular replies.

Extra metadata needs to be stored. Perhaps in new DB fields.

https://about.google/

Our mission is to organise the world’s information and make it universally accessible and useful.

Curtis Yarvin, Urbit founder:

Google’s mission statement is ‘organize the world’s information.’ What that assumes is that the world’s information is a shitshow, a pile of crap, and we need to pay a bunch of people $300,000 a year to go in and sort it out. Our idea is that the world should organize its own information.

https://sparktoro.com/blog/less-than-ha ... n-a-click/

"Less than Half of Google Searches Now Result in a Click"

https://news.ycombinator.com/item?id=20690414

reaperducer wrote:

cyrusshepard wrote:No, more people don't click, because they've taken the answer from your website and displayed it right in their search results.

It's for this reason that's I've stopped embedding micro data in the HTML I write.

Micro data only serves Google. Not my clients. Not my sites. Just Google.

Every month or so I get an e-mail from a Google bot warning me that my site's micro data is incomplete. Tough. If Google wants to use my content, then Google can pay me.

If Google wants to go back to being a search engine instead of a content thief and aggregator, then I'm on board.

"Why Google's new search results design is a dark pattern"

https://mobile.twitter.com/johnny_makes ... 5655079936

https://urbit.org/

A project I've referenced several times. Strange, ambitious, promising....

... and according to some, evil (!)

'Is Urbit evil?' playlist:

https://www.popehat.com/2013/12/06/nock ... t-matters/
http://distributedweb.care/posts/who-owns-the-stars/
https://twitter.com/glenweyl/status/1161764907840413696

This (2016) could use a sequel:

https://urbit.org/posts/common-objections-to-urbit/

Current (2019) official explainer:

https://urbit.org/primer/

Parallels between Urbit and Red

https://en.wikipedia.org/wiki/Red_(prog ... _language)
https://en.wikipedia.org/wiki/Rebol

Red is a under-development programming language. It's an open source successor to Rebol, a niche, high-level, peculiar language and proprietary cross-platform runtime. Rebol and Red borrow some aspects from the Lisp tradition, like 'homoiconicity' ("A language is homoiconic if a program written in it can be manipulated as data using the language" -Wikipedia).

Its killer feature is ease of creating domain specific languages.

Rebol was used as sort of OS-in-an-OS. It had its own desktop environment, and networking stuff.

There's a lower-level version of Red, Red/System is meant for, among other things, developing operating systems. It was guided by a sort of 'lean software' mindset. "Rebol is a rebellion against software bloat."

So, embryonic open source programming languages that challenge the prevailing paradigm.

Urbit's languages, Nock and Hoon are functional. Rebol/Red are classed as functional languages too.

Absentee founders
- Urbit was started by Curtis Yarvin, who ceased involvement in 2019.
- Rebol was created by Carl Sassenrath, who worked on the Amiga OS kernel, but as far as I know he's not actually involved in Red, which is the spiritual successor.

Crypto assets
- Urbit ID is a scarce namespace which functions as an ID system with a potential for sustaining long-term reputation-bearing IDs, preventing spam. They've called it digital land, to make an analogy with Bitcoin as digital money, and it runs on Ethereum. Tlon has sold off parts of it to fund the project. Now they're offering bits as rewards for contributing to the project in other ways, see https://grants.urbit.org/ -- you can get 'stars' (critical IDs that let you spawn personal IDs, to give to your friends, or sell, or whatever) for writing Urbit code, or hosting Urbit meetups.
- Red's gotten into the cryptocoin game too: https://www.red-lang.org/2017/12/leapin ... chain.html

As Red aims to be a fullstack programming solution, we naturally look at what are the potential next big platforms we should support. Since last year, we are watching and studying the various blockchains, especially focusing on the ones supporting smart contracts, Ethereum being the leading chain in that domain.

Red has a crypto token on Ethereum, to help incentivise/reward development of the system. But this was added quite late into the project's development lifecycle, to the displeasure of some fans. The ICO was a success.

https://www.red-lang.org/2018/01/

Both languages are high-risk-high-reward propositions. You could play it safe with JavaScript, Python, FANG stocks, or get involved early on potentially critical, revolutionary computing infrastructure by learning weird unfinished new languages that no one will pay you to use, except in obscure crypto tokens that might be of great value later... or not.

@jasonksackey wrote:#PersonalServer thread

Freedombox was the first project that got me interested in the concept: https://freedombox.org

https://twitter.com/jasonksackey/status ... 2092075008

Interesting that some of these projects offer hardware, others are software only. Hardware means you have something more substantial to sell, maybe it's easier to charge higher prices. Helm is the most cashmoney-premium-priced offering in that thread, and the one with the most visually-interesting hardware (the L-shaped box).

Freedombox is conceptually tied to the physical box. It's supposed to live in your house, because as per Eben Moglen's seminar, possession is 9/10ths of the law, and that's the best place where possessions can be legally protected under law (US in particular, but other places too).

Urbit devs have expressed the possibility of special hardware to run the system's VM. The cloud is the best place to run a ship now. I use my laptop for now.

Holochain is meant to run everywhere: phones, home servers, laptops. Cloud too?

Owning multiple devices is common. But sharing devices is common too. 'Personal computers' are often household computers, or family computers. How our crypto-secured private OS future will work with normal human social patterns is yet to be seen.

Online advertising and its consequences have been a disaster for the human race.

Good sources:
- https://adcontrarian.blogspot.com/
- [more coming. Post suggestions!]

Recent research from Norwegian consumer rights activists about unlawful (contra GDPR) data sharing from popular apps to adtech networks:

https://www.forbrukerradet.no/out-of-control/

My Twitter thread:

https://twitter.com/jasonksackey/status ... 6529024003

Reporting on this issue tended to focus on the involvement of dating apps, a justifiable choice, dealing as they do with particularly sensitive bits of user data: exact GPS coordinates, sexual orientations and preferences, questions about drug usage, and such. The clickbait incentive is there too.

A software pattern for making a distributed and consistent ledger, and the basis for many brilliant scams, chief among them being Bitcoin:

https://overflow.space/2017/12/03/notes ... coin-scam/

A testament to our greed and gullibility.

Does the tech have any non-scam value? Ethereum also has a blockchain. Urbit ID runs on it.

link didn't work - got to wordpress, but: 'OOPS! THAT PAGE CAN’T BE FOUND.
It looks like nothing was found at this location. Maybe try a search?'

I accidentally posted that to the wrong place, then deleted it! It's this https://conflux.game.blog/ps5-vs-xbox4/

Jason,

I use your name in vain - and link to your hypertext post - in this draft:

https://docs.google.com/document/d/1ivR ... sp=sharing

I have some rewriting to do: the 'Story' modality is going to be renamed 'Linear Narrative', and the Pattern Language section will be greatly expanded.

Comments welcome!

Dil

Great stuff! Looking forward to more on Pattern Languages. I clicked through to the PL Institute site and have started exploring that too...

The famous programmable, terminal (and GUI) plain text editor.

https://www.vim.org/

Similar programs:

- emacs - comparible text editor, with a lisp programming environment. Considered close to an OS, with a large suite of of add-ons
- nano - another lightweight terminal editor
- ed - older, more primitive Unix editor
- vi - ancestor of vim
- neovim - popular vim fork

---

How do you configure the font, font size, and theme (colorscheme) on gVim on Windows?

Open the Startup Settings: on the menu bar Edit -> Startup Settings.

Add this:

My speculative piece:

https://overflow.space/2018/02/19/love- ... computing/

And an edutainment/propaganda game:

https://monstermatch.hiddenswitch.com/

MonsterMatch simulates a dating app and shows you how dating apps really work.

"Twitter and Instagram are arguably the best dating apps around"

https://qz.com/quartzy/1753286/why-twit ... ting-apps/

Some alternative OSes:

1. Urbit
https://urbit.org/primer

Your Urbit is a personal cloud server. A simple, private, general-purpose virtual computer on an encrypted P2P network.

The name 'Urbit' refers to a 'clean slate' computing stack that comprises a new OS (Arvo), new languages (Nock, a minimalist functional assembly; and Hoon a symbol-infested Lisp (sort of?)), a new P2P networking protocol (Ames), a new filesystem (Clay), etc. Familiar computing paradigms reimplemented from scratch, functionally. And some Ethereum infrastructure (the 'Azimuth' PKI).

2. BTRON
http://tronweb.super-nova.co.jp/btron.html

Japanese system architecture with some interesting history.

The Japanese government planned to introduce the Matsushita PC in its schools, but the Office of the United States Trade Representative objected, claiming that the plan constituted market intervention and threatened Japan with sanctions; not coincidentally, the former official of the United States Trade Representative office who issued the threats against the Japanese government, Tom Robertson, had been offered by Microsoft the very lucrative position of being their Tokyo-based director for government affairs in Asia.

https://en.wikipedia.org/wiki/BTRON

3. TempleOS
https://templeos.org/

Retro 16-colour OS, a solo project built as an act of Christian worship, including Bible-themed games, a new C-like language (HolyC), a divination program to speak with God, and decidedly no networking capability.

https://thenewstack.io/the-troubled-leg ... rogrammer/
https://rationalwiki.org/wiki/TempleOS
http://www.codersnotes.com/notes/a-cons ... -templeos/

4. kOS
http://www.kparc.com/

K is a very terse language used with time-series, in-memory database platform kdb+, used in finance. kOS is its author's OS project, which uses K. Hopefully it's not just vapourware.

At Iverson College in 2013 he demonstrated the new graphics layer, z – 9Kb of code to replace the X Windows system. For the first time we saw the kOS desktop, solid black with a Tolkienesque legend top left: one system/all devices. Arrayed on the right edge, the icons of five kOS apps. He launched the text editor app and then wrote a new one, working out the key callbacks in front of us and explaining them as he worked. As he defined each callback the new app acquired it: no compile, load, install cycle. In eight lines of K he had replicated the core function of Notepad. At this point, with the new z layer in place, kOS weighed 62Kb.

http://archive.vector.org.uk/art10501320

http://kx.com
http://wiki.c2.com/?KayLanguage

The title of this thread could be appended with 'vs. wikis vs. Q-and-A sites vs. mailing lists vs. chat' and so on.

All these systems are fundamentally similar. They all let us put content online, so they're all usable for publishing and discussion.

They all have their defining peculiarities which make them more or less appropriate for certain sorts of publishing and discussion. But forums have declined in popularity while certain alternatives have thrived. I will argue that the forum format has certain advantages, and it's a bit of a shame that they're a dying species. But they are dying for specific reasons which are worth discussing.

Here's an important essay for background reading:
https://web.archive.org/web/20190113153 ... ws_racket/

It discusses online content about games, but much applies to other topics. This was written when the online space of writing was spread over various forums, blogs, and news sites. Before centralised social media's present utmost dominance.

One question for here is: what combination of which features of forums/blogs/email/messenger pigeons/etc. should I use to make Operating Space the best possible place to read about and discuss technology? This is a matter for not only theorising, but research and development. I'll build stuff and run experiments.

Hacker News is a fine site, an uncomplicated community blog for tech news and various interesting topics. But one important distinction I want for here is: one thread per topic. HN has repeat threads, which is dumb but hard to avoid when posting a reply doesn't necessarily bump an old thread up to the top of the HN frontpage. A forum doesn't necessarily enforce a single-thread policy, and not all forums even have this as a strong rule or custom. But a forum makes it a reasonable possibility. This also touches on the issue of moderation, more on that later.

https://en.wikipedia.org/wiki/Ackermann_function

My JavaScript implementations:
A faster version without recursion:
Even with small inputs, it churns though a massive number of computational steps, and requires a large amount of memory, on the way towards reaching the output.

I first learnt about this function via Urbit.

https://urbit.org/docs/learn/hoon/hoon- ... ackermann/

(I'm doing Hoon School now.)

I thought the statement that it is "not primitively recursive -- meaning it can not be rewritten in an iterative fashion" was supposed to mean that it couldn't be written without recursion, with a loop only. But obviously that's not what that means, or the code above would be impossible. I guess it'd be impossible to do with a for loop where we decide upfront how many iterations we'll go through. (And I know that's not the only way to use the for loop.)

It's a great example of a brief, simple program that's runs in some finite time, and (potentially) consumes a shit-ton of resources. Uselessly. See also: Bitcoin.

It's a clear demonstration of why shared computing systems can't be a free-for-all, they need to be rationed. I wonder if the earliest timesharing systems were robust enough to handle (i.e. not get completely buried, and be able to 'cleanly' stop/crash/interrupt) implementations of Ackermann.

https://en.wikipedia.org/wiki/Dartmouth ... ing_System

That's 1963.

One supposes if you try it on Ethereum you'll be burning a lot of 'gas'.