Notices by Karl Fogel (kfogel@kfogel.org)

Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Wednesday, 29-Jan-2025 11:48:01 JST Karl Fogel

Here's what I think is going on:

He's realized that the biggest lever for power at the federal level is the huge amount of money the national government (which doesn't have budget constraints in the sense that we normally think of them) throws at states and lower-down localities (who do have such constraints) to enable them to continue functioning.

The way the U.S.'s so-called federal system works is that basically nowhere actually balances their budget. Instead, everybody get more or less regular & predictable federal bailouts, in the form of highway funds, pandemic grants, Medicaid reimbursements, education grants, energy infrastructure & environmental improvement grants, etc, etc. The federal government's unique power to print money means it's not just like state government but larger; rather, it's a different beast altogether.

This means that all those court cases about "states' rights" and whatnot are ultimately cosmetic. In the end, if Washington wants you to do something, they'll just threaten your highway funds and then you'll do it.

So he's realized he can just shut off that valve (to make his threat credible) and then only open it up again for mayors and governors who cooperate and signal fealty in whatever ways he demands.

If we let him, that is. Non-zero chance that we do, but also that we don't. I wish there were someone with the vision and spine of a Nancy Pelosi high in the GOP, but I guess their filter has been pretty efficient at meeting those folks out lately.

In conversation about 22 days ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Wednesday, 29-Jan-2025 01:51:29 JST Karl Fogel
- Joey Rees-Hill
@jarhill0 Yup, you can! At the 53rd street station, there aren't even any ticket machines, so if you're not using the app (I'm not, because "I prefer not to install your app" is my middle name) then purchasing the ticket on the train from the conductor is your only option. Unless you want to get a monthly pass, but my usage patterns are too unpredictable to make that worthwhile.

In conversation about 22 days ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Wednesday, 29-Jan-2025 01:48:29 JST Karl Fogel
in reply to
- soaproot
@soaproot Heh! I thought about including some things like that from math, but I didn't feel I had enough expertise to be sure about what was and wasn't settled these days. Unlike other fields, math has gone through some cycles of taking care of deferred maintenance -- you know, things like re-establishing calculus on more rigorous basis than its originators had ever bothered with, and the adoption of ZF[C] as a foundational framework. Maybe there are other examples too (https://en.wikipedia.org/wiki/Foundations_of_mathematics#Foundational_crisis for those keeping score at home).
Math is special, quite unlike other fields in its hypertrophied attention to definitional rigor. To provide an example in mathematics, I would have had to get a Ph.D. first, and, you know, it was just one Fediverse post, so was it really worth half a decade in graduate school? Some might answer "yes" to that question, and, indeed, I'm not sure that the answer is "no", but... it was late and I just wanted to hit Submit.
In conversation about 22 days ago from kfogel.org permalink
Attachments
1. Domain not in remote thumbnail source whitelist: upload.wikimedia.org
  
  Foundations of mathematics
  
  Foundations of mathematics are the logical and mathematical framework that allows the development of mathematics without generating self-contradictory theories, and, in particular, to have reliable concepts of theorems, proofs, algorithms, etc. This may also include the philosophical study of the relation of this framework with reality. The term "foundations of mathematics" was not coined before the end of the 19th century, although foundations were first established by the ancient Greek philosophers under the name of Aristotle's logic and systematically applied in Euclid's Elements. A mathematical assertion is considered as truth only if it is a theorem that is proved from true premises by means of a sequence of syllogisms (inference rules), the premises being either already proved theorems or self-evident assertions called axioms or postulates. These foundations were tacitly assumed to be definitive until the introduction of infinitesimal calculus by Isaac Newton and Gottfried Wilhelm Leibniz in the 17th century. This new area of mathematics involved new methods of reasoning and new basic concepts (continuous functions, derivatives, limits...
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Tuesday, 28-Jan-2025 23:59:28 JST Karl Fogel
in reply to
- harper
@harper Tried. A fight broke out over who gets to chair. I didn't expect that.

In conversation about 22 days ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Tuesday, 28-Jan-2025 23:57:01 JST Karl Fogel

Each conductor on Chicago's Metra Electric commuter train has a uniquely identifying hole punch shape. Many of the shapes are familiar to me now. I don't know this conductor's name, but I always think of her as Ms. Mandelbrot.
In conversation about 22 days ago from kfogel.org permalink
Attachments
1. Chicago Metra Electric ticket, with three holes punched by the conductor, each hole looking like a two-dimensional view of the Moon sliding behind (or on front of) the Earth.
  https://kfogel.org/media/2b977e2c-b4e9-48ae-b086-c179d6f5aaab/PXL_20250128_145136116.jpg
Embed this notice
Paul Cantrell (inthehands@hachyderm.io)'s status on Monday, 27-Jan-2025 13:35:44 JST Paul Cantrell
in reply to

Sitting here trying to make a satire transposing Andreesen’s dumbassery into a nontechnica realm, and this is the best I’ve managed:
“A world in which trees are all destroyed by squirrels, who then themselves die out — logically, necessarily — is a world in which nut production goes through the roof, absent the squirrels to eat them. Because zero divided by zero is infinity, we then live in an infinite forest, beyond all imagining.”

In conversation about 24 days ago from hachyderm.io permalink Repeated by kfogel
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Sunday, 26-Jan-2025 23:31:40 JST Karl Fogel

This paper about DeepSeek-AI lists a zillion authors, and the first listed author is "DeepSeek-AI":

https://arxiv.org/abs/2501.12948

This is like a multidimensional version of when a company buys its own stock.
In conversation about a month ago from kfogel.org permalink
Attachments
1. Domain not in remote thumbnail source whitelist: arxiv.org
  
  DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
  
  We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrates remarkable reasoning capabilities. Through RL, DeepSeek-R1-Zero naturally emerges with numerous powerful and intriguing reasoning behaviors. However, it encounters challenges such as poor readability, and language mixing. To address these issues and further enhance reasoning performance, we introduce DeepSeek-R1, which incorporates multi-stage training and cold-start data before RL. DeepSeek-R1 achieves performance comparable to OpenAI-o1-1217 on reasoning tasks. To support the research community, we open-source DeepSeek-R1-Zero, DeepSeek-R1, and six dense models (1.5B, 7B, 8B, 14B, 32B, 70B) distilled from DeepSeek-R1 based on Qwen and Llama.
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Saturday, 25-Jan-2025 05:32:42 JST Karl Fogel

A pattern I've noticed:

For many fields of specialization, the core field-specific concepts that outsiders assume are well-defined and unambiguous to specialists are instead, to those within the field, often fuzzy, ill-defined, highly subject to debate, and in some cases even considered to be of questionable ontological utility.

Examples: "species" for biologists; "languages" and "words" for linguists.

In conversation about a month ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Wednesday, 22-Jan-2025 23:49:29 JST Karl Fogel
in reply to
- Aram Sinnreich
@aram I don't think I buy this. Was someone a collaborator for living in Germany in 1943 and using the postal system to send domestic mail? Or buying a train ticket? I think for most if its users Twitter is like a kind of utility. I wouldn't be alarmed at my babysitter using Twitter any more than I would if they shopped at Whole Foods. That's not to make an equivalence between Billionaire M and Billionaire B -- they're not equivalent -- but rather just to say that when someone runs a ubiquitous service, and any individual usage of that service represents an infinitesimally miniscule quantum of association between the user and service, then that association is not automatically amplified by the magnitude of the owner's odiousness.

I don't like Twitter either; I wish people wouldn't use it; I avoid it myself now in order to not contribute to the network effect. But that doesn't mean that when someone uses it that are suddenly a Nazi collaborator. If you manage to avoid all such people, it will inevitably mean shrinking your social circle drastically --and many fine & strong resistors will turn out to be outside that circle.

In conversation about a month ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Sunday, 19-Jan-2025 05:09:08 JST Karl Fogel
in reply to
- tef
@tef Blocking a road doesn't reduce the credibility of all future roads.

In conversation about a month ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Saturday, 18-Jan-2025 11:19:20 JST Karl Fogel
in reply to
- soaproot
@soaproot Well, the size of SGMC doesn't matter here. It's just about the arrangement of authenticated and authorized customer entities they offer in their online service.

In conversation about a month ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Saturday, 18-Jan-2025 06:34:14 JST Karl Fogel
in reply to
- soaproot
@soaproot This is technically RBAC, sure, but the trouble is that RBAC sort-of-kind-of implies assigning a user to (one or more) roles within one over-arching organization. Like, you're employed by Soulless Global MegaCorp Inc, and when you log on to the corporate network, you have certain access permisions due to having certain roles within SGMC, Inc.

PAMT is a little bit different because there isn't one authority granting the roles. The user is associated with different organizations, each of whom has their *own* organizational account within the service provider (whether that org account has its own dedicated login account or not doesn't matter), and each of those orgs might assign that user a role within their org's "area" or "domain" (or whatever you want to call it) within that service provider.

I feel like if I say "RBAC" to someone who is entirely familiar with RBAC, they're not necessarily going to think of PAMT. It's sufficiently different to need its own label.

In conversation about a month ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Friday, 17-Jan-2025 03:53:17 JST Karl Fogel
in reply to
- Dave Lane :flag_tino: 🇳🇿
@lightweight Ah, you've just been having better luck than we had, okay. I'm glad! I just wish I knew how to replicate your success. The suggestion to test with mail-tester.com (made in your follow-up reply) is helpful -- thank you.
In conversation about a month ago from kfogel.org permalink
Attachments
1. Domain not in remote thumbnail source whitelist: www.mail-tester.com
  
  Newsletters spam test by mail-tester.com
  
  mail-tester.com is a free online service that allows you to test your emails for Spam, Malformed Content and Mail Server Configuration problems
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Thursday, 16-Jan-2025 13:33:58 JST Karl Fogel
in reply to
- bignose
@bignose Sure, that describes the feature. I'm looking for the label, not the description, though. Not every service supports this way of working. For those that do, I want to be able to say, for example, "Oh, Digital Ocean supports PAMT" or "FooBar Inc doesn't do PAMT, so you just have to use a role account."

In conversation about a month ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Thursday, 16-Jan-2025 13:29:54 JST Karl Fogel
in reply to
- Dave Lane :flag_tino: 🇳🇿
@lightweight The big question is deliverability. When I've tried this, I've ended up spending lots of time trying to convince -- sometimes successfully, sometimes not -- large providers that the emails coming from our service were not spam (which they 100% were not, by the way). The burden of those interactions was very high, but the price of not engaging in them was that our users could not send email to anyone with, e.g., a Hotmail address.

How did you solve this? Or do you not have this problem? Or do you have it but tolerate it?

We knew about SPF/DKIM/DMARC, and AFAIK implemented everything correctly. I'm sure that helped -- our delivery rates would have been much worse without that. But still, some providers are just eager to reject, and yet sometimes people you want to reach are using those providers.

In conversation about a month ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Thursday, 16-Jan-2025 13:24:39 JST Karl Fogel

Is there a word for this pattern?

Some online services handle the "multiple users per organization && multiple organizations per user" situation in the following way:

A user logs in with their individual account and then chooses (typically from a menu in the upper right) which "team" they want want to currently be acting in. A team might correspond to a company, or to a department within a company, or whatever: it's some organization that the user is associated with. So instead of people logging in with a role account that represents the team, they log in as themselves and then "wear a cloak" that lets them act on the team's behalf (within whatever permissions the team admin has granted to that particular member). Naturally, a user can be associated with multiple teams, and those teams don't all have to know about each other -- maybe only the user knows them all.

Is there a name for this pattern? I'm going with "PAMT" ("Personal Account Multiple Teams") for now, but if there's already a widely-used term I'd like to know it.

In conversation about a month ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Saturday, 04-Jan-2025 03:38:42 JST Karl Fogel

The delights of FOSS citizenship:
I just saw a program I wrote 25 years ago, cvs2cl, referenced as an optional dependency by Gource, which is itself not young but which happened to be on the front page of Hacker News today for some reason.
And the Gource documentation doesn't even need to link to cvs2cl's home page, because it can assume that cvs2cl is available in your OS distribution's packaging system (which it is, if you're on any of the major distributions).
Warms the cockles of my hearty-heart-heart on a cold January day.

In conversation about 2 months ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Wednesday, 01-Jan-2025 08:03:04 JST Karl Fogel
in reply to
- sebsauvage
@sebsauvage I will seek that closing paren for the rest of my days...

In conversation about 2 months ago from kfogel.org permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Wednesday, 01-Jan-2025 02:52:04 JST Karl Fogel
in reply to
- Evan Prodromou
- tlariv
@tlariv @evan @tlariv @evan Quite right. The imaginary unit i had not been invented yet, and therefore did not exist.

In conversation about 2 months ago from gnusocial.jp permalink
Embed this notice
Karl Fogel (kfogel@kfogel.org)'s status on Wednesday, 01-Jan-2025 02:48:18 JST Karl Fogel

These days, whatever I'm doing, I just automatically ask myself whether I should be using SQLite. Opening a jar of pickles with a tough-to-turn lid? We've all been there -- try SQLite! Retrieving a ball that rolled under the couch? Sounds like a job for SQLite. Car engine running a little low on coolant? It wouldn't be doing that if the manufacturer had used SQLite...

In conversation about 2 months ago from kfogel.org permalink

Before

Public

Notices by Karl Fogel (kfogel@kfogel.org)

User actions

Following 0

Followers 0

Groups 0

Statistics

Feeds