Hacker News: vehemenz

New comment by vehemenz in "Ask HN: Alternatives to GitHub"

vehemenz — Mon, 17 Aug 2026 14:41:00 +0000

My org's GitHub Enterprise never goes down. The feature set is almost the same, though it lags a few months behind. At least you don't have to learn anything new.

New comment by vehemenz in "Seven books I keep close because I love them"

vehemenz — Fri, 14 Aug 2026 18:18:11 +0000

My convictions are pretty hollow because I'm demonstrating the pointlessness--or arbitrariness, if you like--of gatekeeping the ranks of the "well-educated," as the author tries to do here.

New comment by vehemenz in "Seven books I keep close because I love them"

vehemenz — Fri, 14 Aug 2026 17:56:59 +0000

> This of course is the cornerstone of Western culture and no well-educated person can be without a knowledge of what is in [the Bible].

Knowing what's in the Bible isn't sufficient. You need to understand the full genealogy of the Judeo-Christian tradition.

You need to understand Canaanite cults, the broad influence of both Zoroastrianism (via the Achaemenid Empire) and Hellenism, and how these cultures converged to form proto-Judaistic clerical class (or at least something like it), which eventually crystallized into the synagogue tradition and "modern" Jewish practice during the Hasmonean Dynasty.

New comment by vehemenz in "Delta"

vehemenz — Wed, 12 Aug 2026 20:03:48 +0000

If one builds something with an agent, the process itself is largely opaque. Preserving the transcript shows the reasoning of the agent and the decision process of the builder. This way everyone's on the same page instead of re-prompting same CLAUDE.md but with different results, and/or going through similar work over and over again.

New comment by vehemenz in "U of Michigan drops first-semester grades to ‘curb mental health crisis’"

vehemenz — Wed, 12 Aug 2026 14:14:29 +0000

Exactly. They're rejecting 85% of applicants because they're not qualified, but the 15% they admit are, by and large, not qualified either. I'm at one of Michigan's peer institutions, and we have the same problem.

New comment by vehemenz in "llama.cpp"

vehemenz — Wed, 12 Aug 2026 13:04:05 +0000

Context limit is far too small to do anything serious, tbh.

New comment by vehemenz in "Mark Zuckerberg attacks 'closed' AI rivals as Meta returns to open models"

vehemenz — Mon, 10 Aug 2026 15:23:09 +0000

This is unneedfully obtuse. Double hyphens are a stand-in for em dashes in most contexts.

New comment by vehemenz in "Auto mode is now the default in Claude Code"

vehemenz — Mon, 10 Aug 2026 11:52:34 +0000

I switched to auto mode after using manual mode long enough to understand how it works. That takes at least a few days of babysitting prompts.

Will new users feel intimidated by being in auto mode right away? More likely they won’t notice.

New comment by vehemenz in "Auto mode is now the default in Claude Code"

vehemenz — Mon, 10 Aug 2026 11:47:45 +0000

You mean the guy hired by the Metaverse guy?

New comment by vehemenz in "The Nixpkgs core team has disbanded"

vehemenz — Sat, 08 Aug 2026 14:55:07 +0000

Whatever the legitimate issues were, the letter undermines its own credibility with its overbearing and self-righteous tone. Aside from Determinate Systems and the Anduril bit, the complaints lack specificity. The Blueskyisms thrown around here are just as indicative of "cultural problems" as anything else, assuming these views are representative of the community.

New comment by vehemenz in "DeepSeek V4 Flash 0731"

vehemenz — Sat, 08 Aug 2026 13:17:04 +0000

I wondered this too. Also, the duration of the task depends on the quality of the prompt and the model used. I have a feeling a lot of these day-long tasks are bogged down by suboptimal tool use and on-demand python slop.

New comment by vehemenz in "The Nixpkgs core team has disbanded"

vehemenz — Sat, 08 Aug 2026 02:27:05 +0000

I’m not saying it’s decisively correlated, or anything on its own, but there sure are a lot of anime avatars in this community.

New comment by vehemenz in "Managing AI Coding Costs at Scale"

vehemenz — Fri, 07 Aug 2026 21:18:01 +0000

I've been using it for a week or so. The main draw for me is that I can keep my sessions in one database regardless of the model/provider I use. The webapp can access everything remotely, which is convenient when I'm on my phone.

I haven't gotten a chance to test the multi-agent capabilities, but the DeepSeek Flash prices are so low that I probably will soon.

New comment by vehemenz in "US strikes $1.2B deal to pay German firm to halt offshore wind projects"

vehemenz — Fri, 07 Aug 2026 14:11:10 +0000

Trump was elected because of rural/exurban white insecurity and white cultural decline. An election wasn't the cause of all this--it was more of an effect.

In other words, even if there's a feedback loop, an election won't magically fix everything.

New comment by vehemenz in "The web server deployment model breaks at hobby scale"

vehemenz — Fri, 07 Aug 2026 13:54:32 +0000

It feels like there could be an entire longform piece on how "modern" web development norms arose because hobbyist/student web developers learned everything in an ad-hoc, local environment without access to, you know, a development server or a mature software stack.

New comment by vehemenz in "Herdr is joining Y Combinator. The runtime stays open"

vehemenz — Thu, 06 Aug 2026 22:20:58 +0000

I've been using Herdr for a while now, but I'd like a multiplexer that's also a meta-harness that I can connect to remotely. I know others have setups that work, I just haven't quite figured it out yet.

New comment by vehemenz in "DeepSeek V4 Flash on a Single AMD MI300X"

vehemenz — Wed, 05 Aug 2026 20:43:01 +0000

I see your point, but Anthropic and OpenAI are not only less efficient, they have much higher operating expenses because of their massive AWS/GCP spending (due to not owning it) and have more capital expenditures than everybody else. It's not comparable to third-party providers making some profit on the margin.

New comment by vehemenz in "Zed DeltaDB"

vehemenz — Wed, 05 Aug 2026 19:33:26 +0000

Exactly. The more competitive the space is, the more important it is that products purposefully support 3rd-party providers. Now that people are increasingly switching from Anthropic/OpenAI for inference reseellers and on-prem compute, transparency and portability will be the new values, not being slightly better than the next agent or model.

New comment by vehemenz in "Antigravity 2.0"

vehemenz — Wed, 05 Aug 2026 19:07:14 +0000

Their marketing materials emphasize inter-agent communication, but that's hardly unique to Antigravity. Either way, it's not BYOK, so there's already a substantial reason not to use it if you're deploying multiple agents.

New comment by vehemenz in "DeepSeek V4 Flash on a Single AMD MI300X"

vehemenz — Wed, 05 Aug 2026 15:52:25 +0000

There's no evidence they're making money, and we already know from the projected datacenter capacity in a few years that there will be for more supply than demand, so the major providers will have to repay that debt. Even if they are making money on inference, it's nowhere near enough to cover the bill. It's a losing proposition either way, especially with Chinese models now in play.