HackerLangs
トップ新着トレンドコメント過去質問紹介求人

oblio

27,500 カルマ登録 13 年前

投稿

Ask HN: Why are Gemini CLI and Claude Code TUIs so terrible?

5 ポイント·投稿者 oblio·7 か月前·4 コメント

コメント

oblio
·昨日·議論
How do you even imagine I can reply to your comment?

In what universe was your comment a high quality comment? Especially for a place like HN.
oblio
·一昨日·議論
Online zealotry is just blowing off steam. True political action happens in real life and most of it is just mobilizing people to do what they know deep down is right but real life or laziness stop them from doing

And of course it can fail. But just saying "nothing works" and "everyone is the same" is even worse, it's just guaranteed loss.
oblio
·一昨日·議論
The existence of Telegram doesn't negate the fact that WhatsApp is the world's most popular instant messaging platform, and the others aren't even close.

And Telegram is a lot more developed and has a much larger percentage of the global instant messenger marketshare, compared to Github vs CodeBerg.
oblio
·一昨日·議論
Ah, you're a hobbit, I see?

https://64.media.tumblr.com/tumblr_mcqo6lloZ41qgzdhjo15_r1_2...

"Keep your nose out of trouble and no trouble will come to you".

I used to be a hobbit, too, when I was 19.

It does NOT work. Politics is everywhere, we're social animals.

The only winning move is to do whatever we can to protect democracy and pick the least damaging idiot, and if we accidentally pick the most damaging idiot, get them out of power as quickly as we can.
oblio
·一昨日·議論
This is the worst approach to politics. I would use harsher words.

Yes, people are corrupt. The world isn't binary. Some people are more corrupt than others.

Otherwise Argentina would be Germany and Brazil the US.

People matter and differences matter.

In politics a smart person chooses the lesser evil. Intelligent people don't vote saints, they vote for people that burn their house down less often.
oblio
·一昨日·議論
The cost of surface level rewrites has crashed. Which will probably cover 80% of cases. Caveat emptor on which side your project falls.
oblio
·一昨日·議論
Edsger W. Dijkstra:

"Program testing can be used to show the presence of bugs, but never to show their absence!"
oblio
·一昨日·議論
AI <<can>> do a lot of things, but does it actually do that without an exhaustive test suite (which legacy software generally doesn't have, and it can never be 100%, anyway)?

Between context collapse and hallucinations, how likely is it that the end result isn't slightly polished slop that misses lots of crucial details?
oblio
·一昨日·議論
Why does every country in the world have to look like the US?
oblio
·3 日前·議論
> That sounds like an extremely obnoxious judgement.

Or just a cultural difference.
oblio
·3 日前·議論
> I mean, how many CEOs of major German companies are non-German?

A quick headcount says Mercedes, Adidas, Bayer,...

https://www.consultancy.eu/news/1605/one-third-of-executives...

It pays off to do the actual research before passing judgment.
oblio
·3 日前·議論
It's worse than that. If Apple were banned, someone else would fill their spot.
oblio
·3 日前·議論
The ones I'm talking about had the most active unions.
oblio
·3 日前·議論
I highly recommend reading "The Box", about the history of the shipping container.

Longshoremen literally retired early and were paid pensions out of corporate profits from container related productivity increases.
oblio
·4 日前·議論
LOL. Go do more research if you have issues with their approach. HN really likes to assume that everyone except themselves are idiots.
oblio
·4 日前·議論
In Europe it's done through EuroNCAP: https://etsc.eu/cars-will-need-buttons-not-just-touchscreens...

It has already started asking for physical buttons for key functions to give manufacturers the top safety rating, and it's working. Buttons are coming back.
oblio
·4 日前·議論
Ok... let's say the profession is destroyed. Who maintains all these ever increasing piles of code? Who figures out the requirements, who creates specifications? Who solves issues that are reported by users?

LLMs/turtles all the way down?
oblio
·5 日前·議論
Extraordinary claims require extraordinary proof.

The steam engine was invented in 1765 (+/-) but employment crashed in the textile industry starting in 1815, 50 years later.

LLMs were practically invented in 2017, 9 years ago.

Of course, software is software and we live in a faster age, but guess what, human needs and desires are infinite so requirements keep going up.

At this point I have 0 trust in purely agentic coding (i.e. the 0 supervision approach). YouTube randomly stops displaying subtitles, Gemini CLI (before its shutdown) and Claude Code suffer from constant refresh glitches, the Gemini and Claude mobile websites are synchronous, like we're living in the 60s (if I minimize Firefox both Claude and Gemini stop answering and Gemini until about 1 month ago would also lose my prompt - this is the kind of thing people were fired for in the past, etc). If this level of buginess is the best agentic coding can produce with unlimited state of the art mega corp tokens, software itself will collapse faster than the job market.

Claiming that there will be no more SDEs of any kind, worldwide, in max 2 years, is an extreme position.

I'm not even an SDE. If they're turkeys, AI boosters are pheasants :-p

And I know why the legitimate AI boosters do it. They were genuinely impressed by what they saw in agents, especially since many hated programming anyway. They extended the line into the future and assumed complexity was linear. These people looked at LLMs and agents and didn't see what was there, but what could be (frequently from a position of major SciFi consumers through their entire lives, as all techies are).

Optimism is a positive, extreme optimism is a great flaw.
oblio
·5 日前·議論
The real test for your theory is what people do when they're not getting paid. I have never in my life heard of a large forum/subreddit/WhatsApp group for cleaning toilets. There are no toilet cleaning meetups.

Some, actually many, jobs are just shitty.
oblio
·5 日前·議論
Keep in mind that a good chunk of people here are basically AI boosters: they work for AI companies, sell AI courses and training and certifications, own shares or otherwise benefit from AI companies doing well. Oh, also, lots and lots of marketers. Gitlab kind of exposed this a while ago, their marketing division had alerts based on HN keywords, their marketers were literally expected to be on-call to these keywords and even had a sort of lite SLA to respond to those keywords (stuff like Github, they were supposed to insert themselves in the conversation).

To the direct point: I've used AI for a lot of things. It's good, you can definitely build some things faster. It's far from perfect.

Also, if we were sooooo close to AGI, why would Anthropic and OpenAI continue hiring SDEs like mad, which they are? Why would Claude & co have embarassing bugs years after release (the glitchy scrolling bug, their websites being fully synchronous in 2026!!)? Etc.