This isn't sustainable.
Arguably margins were too low is some parts of semiconductor space, but we still need lower cost per transistor over time or were in trouble.
What the US needs is more available and unregulated and unrestricted funding for startups and not what currently exists which is a huge un-American moat blocking any new businesses in the tech space.
I've taken to buying SACDs when possible. The format supports higher dynamic range, but that barely matters. The mix is the bigger issue and SACD mixes are often better.
Note you need an SACD player.
And also note this only applies for playing on a proper HiFi or with good headphones at least.
In your car, etc you probably want the compressed mix.
I vaguely recall reading about a double-blind test with the DSD stream from an SACD converted to 44.1kHz 16-bit PCM.
Even without proper dithering, listeners could not tell the difference between that and the SACD[1], but could tell the difference between that and the CD version of the same album.
I read that as: SACD customers expect a better mix.
The format is only relevant in that it requires audiophile level dedication and money to use the format in the first place. Not dissimilar to vinyl before its recent boom.
I have an SACD setup, but for what I want to listen to, everything is out of print and secondary market is insane. Players can be found relatively cheaply at thrift stores (many don’t bluray and multi-CD carousels support it with digital output).
Bingo. MS has so many strengths that should make them relevant — a billion or so Windows installations, ~~Office~~ ~~M365~~ M365 Copilot is still the de facto productivity suite, Azure data centers, the OpenAI deal — and they just can’t get out of their own way because their strategy is “leverage those strengths to cram Copilot down peoples’ throats”
They have no taste. And no aspiration to taste. The industry is moving too quickly for the quasi-monopoly strategy of forcing users to buy their product.
Books will be written about how Microsoft had an amazing strategic position and failed at AI because they never prioritized an actual great product.
No, Teams classics was a much better product. This is javascript slop all over. From unresolved updates to inconsistent state across the screens, it's insanely bad.
Didn't the first 365 copilot lauch have a whole rollback as they belateded realised the rag setup would often ignore file access and permissions, so queries like "List the highest paid members of x team sorted by salary" would just work etc?
The combo of rushing with a technology that isn't very easy to control, understand or securely limit is just mad to me.
I really dislike that I can't customize it with permanent config files, similar to how I can configure a regular GPT model agen. I guess it's probably because it's in the fancy word they use for "beta".
I haven't really used any other Copilot product in a while since they were so bad compared to our other corporate options, but I'm rather impressed with Cowork inside it. Exactly because we can actually use it without breaking any EU laws.
I doubt an LLM is calculating withholding. I presume 99.9% of the actual logic will still execute in QuickBooks or Paychex etc. Lots of this sounds like cross system orchestration against well defined APIs.
Yes, there's still danger, it could use the APIs wrong, but humans can use the GUI wrong too
Also you have to remember the basics of statuspage messages: Its always just elevated error rates. Even when the error rate is elevated to 100%.
"We are observing elevated error rates when accessing objects stored in the affected region. Impacted customers may experience increased latency or intermittent failures while retrieving debris adjacent data." /s
I work at a less innovative place, and I see out product managers coming with prototypes, at least solid mock ups rather than just a jira.
They socialize it with potential users, they iterate, they find missing requirements, it's pretty powerful.
The net result is we're building better features faster.
We need to match the tool to the uncertainty we're facing.
The "just prototype it" thinking addresses "feasibility uncertainty". It surfaces blind spots and helps people tangibly reason about what the product looks like. It's a great exploratory tool for incremental ideas.
But it doesn't address the the larger uncertainty that startups are faced with: "market uncertainty" (or pmf). It doesn't answer "should we be building in this the first place?" That's where writing as a tool of thought is most powerful -- it helps you crystallize what problem we're actually solving.
The "just prototype it" culture (which is being promoted these days because Claude Code makes it easy) risks answering the wrong question, or at least the right question but in the wrong order. You end up with organizations that are incredibly fast at building things that no one should have built.
Ironically sometimes you need to start from a lower resolution (i.e. writing a doc). Prototyping too early is premature optimization.
I really agree with a lot of this but also think it may be hitting a bit too hard. It may be most applicable to engineer founders.
My anecdote is that, after a few stings with non-technical founders, a doc etc will not improve the chances to reach PMF and prototypes that they can understand can improve the chance.
Outside of the startup context, I have also seen prototypes (hand written way back when that was a thing) resonate with FAANG directors much more than brainstorming.
I am very much for not just vibe it, and the biggest risk of prototypes is they lend to just directly launching broken systems to production. But I think this is a different topic than reaching PMF.
I prefer prototyping to slides. The reason is it helps me understand the problem and edge cases better. Getting AI to build means you could potentially understand it even less than if you put the slides together.
Hiring talent that is passionate about delivering a quality product is more important than ever considering there are so many ways to take shortcuts now that might not be obvious until later.
Can confirm this in my portco's and a couple other peers (one of whom previously founded a major threat intel platform).
If you have product-minded Engineers and engineering-minded PMs, you can merge the two into a single function and remove much of the friction surrounding requirements, prototyping, and launching MVPs.
A couple of these products are already being deployed by F100 security teams as we speak. I also know of one F10 that's building it's own entire security platform from scratch with a team of security engineers working directly with one of the foundation model vendors.
Too many people on HN are divorced or too OOTL from some of these initiatives and then get blindsided during layoffs.
What matters now is DOMAIN EXPERIENCE. Do you understand good development principles and the problem your ICP is trying to solve and how pricing, packaging, and procurement is structured? I don't need a code monkey, process sloths, and queens of the calendar. I need domain experts who can actually execute.
reply