https://www.phillipcarter.dev/
Maintainer of MorbJS https://www.npmjs.com/package/morb
- phillipcarterNot to discount physical infrastructure, but the world is quite digital these days and being at the absolute top of the software + associated techs economy is nothing to sniff at.
- I don't think it's about dreaming to be Google. K8s is pretty easy to set up now with a hosted cloud platform if you start with it, and helm takes care of pretty much all your infra needs. Migrating to K8s is what's awful. From there, the docs have most everything you need to know and there's an abundance of helpful information online that covers most problems you'll run into.
- By recently you mean a decade ago yeah? I mean it’s fair that it was only a half-decade (.NET 5) when it was genuinely complete enough, but lots of stuff was in good shape when it was called .NET Core.
It sounds like you’re projecting the problems of an existing .NET shop onto the shape of a startup without all that baggage. I can assure you, having worked with many customers running new business on newer .NET, it hasn’t been a legit technical concern since about .NET Core 3.
- Not really? Having come deom a TS + Go startup it’s pretty trivial to wire up domain objects across each language and define a clean API boundary with some enforcement at build time. And Go was a far better choice for the backend than TS for some lower-level memory considerations.
- Is it though? Backends can be any language and there's a lot more variety there -- TS+node, Go, Python, Java. It's just .NET that's largely ignored for no real technical basis.
- Rider is your option there, it's better than Visual Studio (I used to work on VS).
- > I wonder how much of Meta the corporation is a scam waiting to crumble.
I imagine almost none of it. Social networks solve connectivity problems that people want solved. Talk to some "casuals" who aren't in tech about how they find out about new restaurants, social trends, arts and crafts, places to go visit, etc. and the answer is Instagram or TikTok. And FB does the same but for older generations.
Ads are also a fundamental revenue pillar in this world. You can layer in relevance ads for a product to anyone, at any time, for any topic. If something exists and people pay attention to it, there's a way to make money advertising around it.
There's ... certainly deeper questions to be had about if this stuff is actually good for us, but in the mean time, it's very real and worth a lot of money.
- Would be curious about this in more detail since I’ve not normally seen a JS app be significantly slower due to adding autoinstrumentation over the years. There’s obviously an overhead but aside from the occasional bug I’ve never seen it be significant to impact user experiences or cost to serve.
That said if your goal is basic performance metrics and nothing more, then tracing is overkill. Don’t even need an SDK, just monitor the compute nodes async with some basic system metrics. But if your goal is to narrow down behaviors within your app on a per-request basis there really is no way around tracing if you value your sanity.
- I didn’t agree with the article at all.
- Wouldn't be a gary marcus post without congratulating himself for tweeting something that all but confirms a disaster that hasn't happened yet.
Anyways, OpenAI is not in profit-seeking mode, and there's no economic incentives to do so right now.
- > that it's not on anyone's land
Oh you can bet that, if we assume this happens in 10 years, various countries will absolutely do a "land grab" up high. There is no escaping it.
- Not necessarily? It's been the basis of one of the major ways people would query their data since 2023 on a product I worked on: https://www.honeycomb.io/blog/introducing-query-assistant
The difference is this feature explicitly isn't designed to do a whole lot, which is still the best way to build most LLM-based products and sandwich it between non-LLM stuff.
- I'm "team Anthropic" if we're stack ranking the major American labs pumping out SOTA models by ethics or whatever, but there is no universe in which a company like them operating in this competitive environment didn't pirate the books.
- Chat for teens.
- I don't want edgy 4chan-bot in my codebase, so there's no reason to adopt this when there are many other great coding models available.
- Uhhh, I'm pretty sure DeepSeek shook the industry because of a 14x reduction in training cost, not inference cost.
We also don't know the per-token cost for OpenAI and Anthropic models, but I would be highly surprised if it was significantly more expensive than open models anyone can use and run themselves. It's not like they're also not investing in inference research.
- Sure. But Gartner is Gartner. They are a trusted brand and you can’t “hold them wrong”. And they also do a great job of being pulled into a meeting with the throat-choker to gaslight them into being fine with the decision.
Deep Research doesn’t do this, and even if it could today, human trust systems take a very long time to build.
- I wouldn’t bet on automated deep research until they figure out a business model that gives people a throat to choke. Enterprise software is a world where it’s more important to have another human you can blame for when you fuck things up than actually making a good decision. What incentive is there for an exec to say, “well I ran a deep research and it seemed good enough to me” when their boss demands an answer as to why $VENDOR was a bad choice?
- Purchasing decisions. If Gartner doesn’t claim you’re a Leader, then a massive chunk of your addressable market is not unlockable for you. This may be fine for now, but eventually when your investors demand accelerated revenue curves (and you’re not an AI coding tool), then you’ll be talking to Gartner and praying they place you high. Full stop.
Separately, they offer consulting with their analysts. A lot of these consultants are quite knowledgeable. They also are usually there to help a leader make a purchasing decision.
- Yeah, nah. The enterprise software market is nowhere near close to being upended by AI, and Gartner has their tendrils deeply wrapped inside of it. Small companies like Netlify which are barely in use by this market are not a canary in the coal mine.