Dean Hamstead

6.4K posts

Dean Hamstead banner
Dean Hamstead

Dean Hamstead

@PerlDean

Dean coding in Perl. Probably won't follow you. This account won't tweet or retweet politics.

Sydney, Australia Katılım Eylül 2011
103 Takip Edilen288 Takipçiler
Dean Hamstead
Dean Hamstead@PerlDean·
@ausstockchick Obviously the media will publish clickbait in either direction. Both are self fulfilling though.
English
0
0
0
58
that stock chick
that stock chick@ausstockchick·
How much the media narrative has shifted on the housing market. I just read that house prices could fall up to 10%, the biggest correction in 40 YEARS says a Morgan Stanley analyst. It wasn’t long ago the narrative was bullish. #ausbiz
English
55
10
187
9K
Dean Hamstead
Dean Hamstead@PerlDean·
@neetcode1 Whats the all-in-one part? all of it doesn't seem to be in one?
English
0
0
0
166
NeetCode
NeetCode@neetcode1·
I wonder how much damage this does to Google Cloud, let alone Railway. Has to be at least $1B for google long term. As someone who worked in gcp, I would describe the culture as the opposite of customer obsession. I have a few horror stories I’ll keep to myself, but let’s just say there’s many smart people who are just not incentivized to care about the long term success of its services.
Railway@Railway

We are working to restore the Google Cloud infrastructure that powers our dashboard, API, and internal network's control plane. We are in direct contact with Google Cloud's support team. We do not have an ETA at this time. We will continue to post updates on status.railway.com

English
31
30
1.1K
215.2K
John Carmack
John Carmack@ID_AA_Carmack·
It is a shame that the simple act of transferring a large block of data as fast as possible over the internet is not handled effectively by the primitive operating system calls. You either multiplex over parallel persistent TCP connections to combat head-of-line blocking and slow starts, or reinvent reliable delivery and flow control over UDP. QUIC has a lot going for it, but it is a large library (six figure LoC!) and conflates security and performance in a way I don’t love. There is also fundamental information about competition with other processes and link layer congestion that should be useful, but is unavailable to user libraries. You should be able to just write(really_big_buffer) and it is all taken care of for you.
English
147
146
3.3K
252.1K
Eric S. Raymond
Eric S. Raymond@esrtweet·
Announcing inventory release 1.0 A pain-reduction tool for Linux and *BSD system administrators. It lists all your packages from every installer, incuding: apt, pacman, dnf, zypper, apk, emerge, xbps, flatpak, snap, pipx, cargo, gem, yarn, pkg, npm, homebrew, and nix. It normally tries to skip packages in your system base, but you can override this. New in this release: Tested on apt, pacman, flatpak, npm, gem. gitlab.com/esr/inventory
English
12
15
207
14.6K
that stock chick
that stock chick@ausstockchick·
Today I had a massive win. You know those days when it’s like, I can take over the world. It was one of those days today! Details to follow. #ausbiz
English
16
2
110
6.5K
that stock chick
that stock chick@ausstockchick·
The treasurer is saying CGT changes have to be applied to all assets because investors would try and get around the changes by…. Set up a company Put their investment property inside the company When they sell, they’re technically selling shares in a company not property directly Those shares could still qualify for the 50% CGT discount #auspol
English
72
6
198
31.2K
Dean Hamstead retweetledi
Andrej Karpathy
Andrej Karpathy@karpathy·
Good post from @balajis on the "verification gap". You could see it as there being two modes in creation. Borrowing GAN terminology: 1) generation and 2) discrimination. e.g. painting - you make a brush stroke (1) and then you look for a while to see if you improved the painting (2). these two stages are interspersed in pretty much all creative work. Second point. Discrimination can be computationally very hard. - images are by far the easiest. e.g. image generator teams can create giant grids of results to decide if one image is better than the other. thank you to the giant GPU in your brain built for processing images very fast. - text is much harder. it is skimmable, but you have to read, it is semantic, discrete and precise so you also have to reason (esp in e.g. code). - audio is maybe even harder still imo, because it force a time axis so it's not even skimmable. you're forced to spend serial compute and can't parallelize it at all. You could say that in coding LLMs have collapsed (1) to ~instant, but have done very little to address (2). A person still has to stare at the results and discriminate if they are good. This is my major criticism of LLM coding in that they casually spit out *way* too much code per query at arbitrary complexity, pretending there is no stage 2. Getting that much code is bad and scary. Instead, the LLM has to actively work with you to break down problems into little incremental steps, each more easily verifiable. It has to anticipate the computational work of (2) and reduce it as much as possible. It has to really care. This leads me to probably the biggest misunderstanding non-coders have about coding. They think that coding is about writing the code (1). It's not. It's about staring at the code (2). Loading it all into your working memory. Pacing back and forth. Thinking through all the edge cases. If you catch me at a random point while I'm "programming", I'm probably just staring at the screen and, if interrupted, really mad because it is so computationally strenuous. If we only get much faster 1, but we don't also reduce 2 (which is most of the time!), then clearly the overall speed of coding won't improve (see Amdahl's law).
Balaji@balajis

AI PROMPTING → AI VERIFYING AI prompting scales, because prompting is just typing. But AI verifying doesn’t scale, because verifying AI output involves much more than just typing. Sometimes you can verify by eye, which is why AI is great for frontend, images, and video. But for anything subtle, you need to read the code or text deeply — and that means knowing the topic well enough to correct the AI. Researchers are well aware of this, which is why there’s so much work on evals and hallucination. However, the concept of verification as the bottleneck for AI users is under-discussed. Yes, you can try formal verification, or critic models where one AI checks another, or other techniques. But to even be aware of the issue as a first class problem is half the battle. For users: AI verifying is as important as AI prompting.

English
134
537
4.4K
844.5K
DHH
DHH@dhh·
Spent a few days on a rural internet connection. Real vintage stuff at 2mbit. Coming home to this feels like being catapulted back to the future! Not just the throughput, but latency: 1ms to the hop! And the modern internet being so CDN heavy means I actually get the 2gbit DL.
DHH tweet media
English
54
5
714
53.6K
TRÄW🤟
TRÄW🤟@thatstraw·
Which one do you use as your daily driver?
TRÄW🤟 tweet media
English
172
17
263
31.8K
Dean Hamstead
Dean Hamstead@PerlDean·
Big Pickle is a respectable model. It's been enjoyable to experiment with. It is enthusiastic and capable, but needs careful guidance and unit tests.
English
0
0
0
11
Anaya
Anaya@Anaya_sharma876·
If Linux is so powerful… why do most developers still use macOS?
Anaya tweet mediaAnaya tweet media
English
495
42
2.1K
210.9K
Dean Hamstead
Dean Hamstead@PerlDean·
@ausstockchick Imagine you're a family who bought thanks to the bank of mum and dad. Now your mortgage is underwater and so is the mortgage of your late 60s parents.
English
0
0
0
37
that stock chick
that stock chick@ausstockchick·
Do you think investors who bought real estate in the last year to negatively gear will be underwater soon? #auspol
English
21
1
9
4.6K
that stock chick
that stock chick@ausstockchick·
What fascinates me is reading the comments from the left who are celebrating the budget. Someone just re-tweeted my tweet and said Australia is taking back the working class. Honey, Australia just cemented you further into the working class than you have ever been before and it’s astounding some people can’t see that. People who built wealth via the old system are in complete shock. #auspol
English
166
86
1.4K
34.2K
ReliableSite Hosting
ReliableSite Hosting@ReliableSite·
@PerlDean Fair concern, but this isn’t ad hoc assembly. We’ve been refining our process for nearly 20 years. Our data centers are controlled for static and dust, and every server goes through testing and validation before it goes live. Hope that helps.
English
1
0
0
25
ReliableSite Hosting
ReliableSite Hosting@ReliableSite·
Behind every deployed server is a build like this - an EPYC 4545P assembled for performance, stability, and uptime.
English
2
2
27
1.4K
Dean Hamstead
Dean Hamstead@PerlDean·
@ausstockchick Rents will go up for anyone who can't buy. Immigrants for example. So it will rebalance toward cashflow rather than capital growth speculation
English
0
0
0
439
that stock chick
that stock chick@ausstockchick·
Can the rental market balance itself once negative gearing is removed? Removing means supply and demand both shrink together. We will see a shift in housing from investor ownership to owner-occupier owners. A softening in house prices. If the housing market softens. Investing in real estate just became pointless. Exactly as planned.
English
69
3
116
20.4K
Dean Hamstead
Dean Hamstead@PerlDean·
@ausstockchick Indeed. However the Australian mindset is to fins cheaper labor rather than to have more productive employees.
English
0
0
0
26
that stock chick
that stock chick@ausstockchick·
Have you ever thought about your level of productivity in society? Because Australia sure hasn’t. We’ve built an economy addicted to digging things out of the ground and watching house prices go up and called it prosperity. Meanwhile, we’ve built a public sector with layers of middle management, diversity coordinators, and consultants writing reports about reports. We tax work and innovation while sheltering passive wealth. And we wonder why the next generation can’t get ahead. The lucky country is running on borrowed time. The worst part is. People voted for this!
English
28
7
185
5.7K