Hacker Newsnew | past | comments | ask | show | jobs | submit | huqedato's commentslogin

That's smoke and mirrors. You can't logically predict the market. It never worked.

Sure you can predict the market. Making money off of it beyond the regular risk-adjusted return is what's hard. (And the prediction of this article is indeed based on that assumption.)

I completely disagree with the idea that 2025 "The (only?) year of MCP." In fact, I believe every year in the foreseeable future will belong to MCP. It is here to stay. MCP was the best (rational, scalable, predictable) thing since LLM madness broke loose.


I don't really understand the concept. What is the definition of "mini-framework" ? The author should have given a few examples.

I have the impression that he confuses "obscure" with "mini". Either framework or library..


I think concrete examples of this are tricky because these "mini frameworks" only exist inside of a companies' proprietary codebase. My understanding of the concept is its an abstraction layer built on top of a more general abstraction, except the more general version is well documented and well understood by the company (for an internal framework) or even overall developer community (for something open source).


Sounds too good to be true. I don't buy this. Every once in a while (several times a year) I bump in an article claiming to cure cancer.


Being outright dismissive without a corresponding article-specific argument is about the worst thing you can ever do on any forum.

This is not to be confused with dismissing with an article-specific argument (which you don't have).


Not the GP, but I'll bite. I'm skeptical too, so I read TFA.

---

They tried 9 bacterias and a 1 control group. Using n=3 * (9+1) = 30 mice they got this result:

> Most remarkably, E. americana demonstrated exceptional therapeutic efficacy, achieving potent tumor suppression and complete tumor regression (complete response, CR) following a single bacterial administration. The therapeutic kinetics revealed that mice treated with R. qingshengii exhibited initial tumor suppression up to day 5 post-injection; however, tumor re-growth was subsequently observed, suggesting that while this strain possesses antitumor activity, its therapeutic effects are not sustained long-term.

They claim "p < 0.0001" that in my opinion is a loooot of zeros for only 3 mice.

They end the experiment after 40 days, so it's not clear if the cancer would reappear after a a few months.

They tried again with 5 mice, and got similar results, so it doesn't look like a fluke, but it's a very short time to claim an "elimination" line in the title of the press release. The research article has a more neutral tone.

---

It looks like the idea is that these bacterias can survive without oxygen, so they are happy to live in the tumor that usually has a low number of capilar and blood and oxygen. IIUC the bacterias kill the nearby tumor cells, perhaps steal their food and also make the immune system go there and kill everything just in case. This sounds like a sensible idea, but it's too far from my area to be sure.


Fwiw, 3 mice is imho sufficient because these mice might as well be clones. They're not clones, but they're genetically very similar to one another, so variations in results are not expected.

Granted, the other concerns hold.


I confirm that. It had no idea how to use Deno v2+.


These are superb. I remember a similar magazine many years ago: Increment


Thanks!

Do you mean https://increment.com/programming-languages/? Huh, I never heard about it before - thanks for mentioning it! :)


Yeah, that one. I purchased the entire set with discount. Sadly they discontinued it.


Aham, tx. Good to know - I'll switch my projects to Deno.


you know Deno is VC backed right


Looking for a similar GEMINI.md


It might support AGENTS.md, you could check the site and see if it’s there


...and also bad engineers write bad code at small companies.


"Zero technical debt" - I doubt this is feasible in practice.


I'm not even clear on what it means, technical debt is non-deterministic in many cases.

To say it differently: if you wrote code that was perfect in time 0, that code may become legacy in time 100.

Are they saying you should continuously refactor all your code to cover the 'current user needs'?

I just think it's an oversimplification for those cases where you don't mind not covering the 0,001% of use cases.


Even if it is, it sounds highly ineffective, unless the only value you are delivering is source code.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: