cbsmith

Agreed, though this appears to be much more than open weights.

Missed opportunity to say "Kinds of kindness". ;-)

Yeah, but only the Rust guys are right. ;-)

What could possibly go wrong? ;-)

I kind of like the whole Paramount acquisition as something that is just going to drag them down.

Anyone else immediately think "Clippy for developers".

I thought we all moved over to Niquests and/or aiocquic and HTTP/3 these days.

"my bet is that close to no one knows what the acronym BDD stands for".

That's disconcerting. Software engineers ought to know.

As a guy named Chris Smith, I can tell you, people do absolutely "match people on first and last name", and it causes me no end of trouble.

> I don’t know if that’s true. More schooling does not mean more trustworthy or careful doctors. And I suspect some negative trait selection.

The medical boards are well aware of this, which is why the standards are far more than just about having more schooling.

I am familiar with Friendman's dissertation, and while the economic claims it makes are solid, it isn't a take down about the medical impact of the boards. It really doesn't say one thing or another about them. Let's put it this way: if you are going to inflate the cost of a service, it's hard if the quality of the service is terrible and easily replicated by someone else.

Everything old is new again. ;-)

The article concedes that there is a minimum power consumption. Thinking that therefore defines your generation strategy (e.g. I will have plants that constantly generate that minimum amount and the rest I'll handle with separate peak plants) is a myth.

...because they have other goals beyond reduce carbon emissions.

Better a financial barrier than a physical one. If satellites and spaceships are literally smashing in to each other, I have a hard time interpreting it as anything other than a regulatory failure.

Are there other OSS LLMOps projects that have overtaken you? I couldn't think of one.

As someone who has jumped from one industry to another: I'm sorry, but domain expertise isn't much of a moat. Yes, it takes a while to learn a particular industry/business, but it doesn't take that long, and worse still: one advantage that LLMs have over us humans is they have such a broader inventory of human knowledge at their disposal. I've literally used LLMs as product managers for new domains, and while I see the rough edges that LLMs have, they always have significant domain expertise.

I don't think that's the moat.

Confirming it was fixed and it worked well.

Installing collected packages: sortedcontainers, boolean.py, urllib3, typing-extensions, tqdm, tomlkit, tomli-w, tomli, tenacity, stevedore, shellingham, ruamel-yaml, regex, PyYAML, pyparsing, pygments, pycparser, platformdirs, pip-api, packaging, packageurl-python, msgpack, mdurl, marshmallow, MarkupSafe, license-expression, joblib, idna, h11, filelock, defusedxml, click, charset_normalizer, certifi, annotated-types, annotated-doc, typing-inspection, requests, pydantic-core, py-serializable, pip-requirements-parser, nltk, markdown-it-py, jinja2, httpcore, dparse, cffi, anyio, rich, pydantic, httpx, cyclonedx-python-lib, cryptography, CacheControl, typer, safety-schemas, joserfc, bandit, pip-audit, authlib, safety, velonus Successfully installed CacheControl-0.14.4 MarkupSafe-3.0.3 PyYAML-6.0.3 annotated-doc-0.0.4 annotated-types-0.7.0 anyio-4.13.0 authlib-1.7.2 bandit-1.9.4 boolean.py-5.0 certifi-2026.4.22 cffi-2.0.0 charset_normalizer-3.4.7 click-8.3.3 cryptography-48.0.0 cyclonedx-python-lib-11.7.0 defusedxml-0.7.1 dparse-0.6.4 filelock-3.29.0 h11-0.16.0 httpcore-1.0.9 httpx-0.28.1 idna-3.15 jinja2-3.1.6 joblib-1.5.3 joserfc-1.6.5 license-expression-30.4.4 markdown-it-py-4.2.0 marshmallow-4.3.0 mdurl-0.1.2 msgpack-1.1.2 nltk-3.9.4 packageurl-python-0.17.6 packaging-26.2 pip-api-0.0.34 pip-audit-2.10.0 pip-requirements-parser-32.0.1 platformdirs-4.9.6 py-serializable-2.1.0 pycparser-3.0 pydantic-2.13.4 pydantic-core-2.46.4 pygments-2.20.0 pyparsing-3.3.2 regex-2026.5.9 requests-2.34.2 rich-15.0.0 ruamel-yaml-0.19.1 safety-3.7.0 safety-schemas-0.0.16 shellingham-1.5.4 sortedcontainers-2.4.0 stevedore-5.7.0 tenacity-9.1.4 tomli-2.4.1 tomli-w-1.2.0 tomlkit-0.15.0 tqdm-4.67.3 typer-0.25.1 typing-extensions-4.15.0 typing-inspection-0.4.2 urllib3-2.7.0 velonus-1.0.0a1

(venv) $ velonus scan data

Traceback (most recent call last):

  File "~/venv/bin/velonus", line 5, in <module>

    from shield.main import app

  File "~/venv/lib/python3.12/site-packages/shield/main.py", line 15, in <module>

    from shield.commands import auth, config, scan

  File "~/venv/lib/python3.12/site-packages/shield/commands/scan.py", line 32, in <module>

    from scanner.pipeline import ScanPipeline

ModuleNotFoundError: No module named 'scanner'

Wow. I remember the Toronto PET Users Group. That takes me back.

He he. If you think the costs for an education will go down because they're seeing fewer applications and they're getting less funding from other sources, I have bad news for you.

This is a bizarre article, and it kind of reminds me of the concept that the only truly good software is the software you've never used.

Also correct. I see the problem as particularly rampant with the video sites.

I was going to say: "in what way is the problem restricted to health?"

It's kind of a comprehensive guide to all the ways that restricted shells don't.

Yup. Sounds right. Interestingly though, you can use the new input cover with the older frames.

More like the reverse: if you want the new battery you need a new bottom.

Ask yourself why airlines deliberately overbook flights while venues do their best to never overbook events, why are subject to identify verification before a flight, and why you go through the TSA before you board... and you'll start to understand why airline seats are maybe not such the perfect metaphor for concert seats.

> If only the federal government were allowed to pursue this case, it would have ended when the administration changed.

This is more why DOJ cases should remain independent from the executive branch. Politically controlled prosecutions means justice is intrinsically unequal. Having states be independent is helpful, but not in this regard.

> Because artists don't always want to extract the maximum money possible from their fanbase?

I think that's both true and not. The larger truth is that trying to maximize the extraction during a single ticket sale is incredibly short-sighted of an artist. Having fans attend shows is a very effective way to grow your fan base and your brand, and that brings so much more lifetime value for an artist than you'd ever get from a single ticket sale (except for maybe on your retirement tour --and even then).

> They don't seem to mention the most obvious reason: the same companies profit from both the primary and secondary market.

That's not true. Ticketmaster has a monopoly (or near-monopoly) on the primary market. On the secondary market they have a fraction of the market; the dominant players are StubHub, SeatGeek, and Vivid. Furthermore, most of the revenue from primary ticket sales goes to the venue and the artist/promoter, and they are usually completely disintermediated from the resale market.