tomaytotomato

Wow, a real throwback to my earlier years exploring the Warez scene as cash poor teenager, which meant you had to get creative to find what you wanted.

Looking back in my 30s, I think part of my brain has atrophied because I can now purchase and download games and content with a few clicks on Steam and £40, which will be ready to play after dinner.

Before you had to think and search forums (both public and private) for DDL links, CD keys, cracks, passwords to unzip files.

Then once you had downloaded these multiple RAR/Zip files you then had to check and verify them for corruption or viruses. After that it was reading the sacred README.nfo file included in the crack zip. Which if you didn't specify to open in Notepad or Notepad++, it would open the Windows System information window.

The last stage before firing up a game was the patching/cracking phase; that could involved things like editing hosts files, unplugging or disconnecting from the internet, or modifying your firewall to prevent network requests for anti-piracy software. Then you ran your crack or copied a patched .exe file over into the game directory. Aye presto, you're ready to fire up (hopefully it didn't have a virus or rootkit)

...but wait there is no loading videos or splash screen (panic). Well that's because the pirates were saving storage space and removed unnecessary game videos and music assets.

<Play game> Bioshock

Ah, that was a nice nostalgia trip.

Also for any other nostalgia trippers, do you remember the first thing you always installed was a cracked version of WinRar (before 7zip became cool), running on a fresh install of Windows_7_SP3_BLACK_EDITION_XXX.iso

Yes and no

A social media feed implies 1(n) curated by 1 algorithm hosted on Facebook/Twitter/Instagram

What I was thinking is:

- foo.social

- bar.social (tech curations)

- java.bar.social (sub curated Java list)

All these DCS (domain content servers) would be polled by your own local client

Your client can then aggregate or organise how it shows this feed

e.g. I could have a trending aggregator for situations where a blog post is shown on multiple domains (sort of shows virality)

I've always liked this idea.

However I am not sure about "perma-shortlinks", for discovery on other sites as the means of networking and discovering content. It seems clunky to maintain as it requires a human or some automation to curate/maintain the links. If a blog removes a link to another blog, then that pathway is closed.

It would be cool if we could solve that with a "DNS for tags/topics" a - Domain Content Server (DCS) e.g.

1. tomaytotomato.com ==> publishes to the DCS of topics (tech, java, travel)

2. DCS adds domain to those topics

3. Rating or evaluating of the content on website based on those tags (not sure the mechanics here, but it could be exploited or gamed)

You could have several DCS for topics servers run by organisations or individuals.

e.g. the Lobsters DNS for topics server would be really fussy about #tech or #computerscience blog posts, and would self select for more high brow stuff

Meanwhile a more casual tech group would score content higher for Youtube content or Toms Hardware articles.

This is just spit balling.

Tempted to submit a Java app wrapped in PHP exec() :D

I'm guessing you like Gradle? :)

I would like to hijack the "high level language" term to mean dopamine hits from using an LLM.

"Generate a Frontend End for me now please so I don't need to think"

LLM starts outputting tokens

Dopamine hit to the brain as I get my reward without having to run npm and figure out what packages to use

Then out of a shadowy alleyway a man in a trenchcoat approaches

"Pssssttt, all the suckers are using that tool, come try some Opus 4.6"

"How much?"

"Oh that'll be $200.... and your muscle memory for running maven commands"

"Shut up and take my money"

----- 5 months later, washed up and disconnected from cloud LLMs ------

"Anyone got any spare tokens I could use?"

This gives MoE (Mixture of Experts) a whole new meaning, albeit a slightly darker one.

We live in a world increasingly becoming hypercapitalistic in every facet of life with problems and solutions being marketed together, all fed with algorithms.

"Have you got chronic sebhorric dermatitis, click this link to make it go away".

"You may be eligible for compensation if you bought a Volkswagen"

"Fight the corrupt fascist government, buy a gas mask here"

"Fight the corrupt socialist government, buy a year's supply of Iodine tablets"

Can the implementation used here for tone and pronounciation apply for Music?

It would be cool if a model could tell you if you are singing or playing a piece of music with the right intonation and other ways.

Over the last year I have seen some really nice exploration posts on HN with people poking around on:

- Splinter Cell

- Deus Ex

- Thief

- Civ

This is great work and will help tell the story of how these games were made.

It would be great if all games after a certain period of time were opensourced like some companies are doing:

https://github.com/electronicarts

https://github.com/bobeff/open-source-games?tab=readme-ov-fi...

UK government departments are slow and hostile to change, so I am skeptical that Palantir being parachuted in, would produce anything more than a CSV file with a few hundred rows in it.

How are Palantir so effective (as this article is alluding)?

From a cynical British perspective, when I think of government departments and civil servants. I think inefficiency, data siloing, politics and lack of communication between departments and also internally not communicating between teams. Not withstanding a lack of cooperating and willingness to change.

Did Palantir have a political mandate, or can they just cut through the bureaucracy or bypass it with technology?

I think Knapsack is an older word compared to Backpack, but it has connotations of being more rugged or tougher than a Backpack.

Thanks for this - its interesting how I have come to this conclusion as well.

My reworked approach is to return a list of results with a probability or certainty score.

In the situation of someone searching for London, I need to add some sort of priority for London, UK.

My dataset is sourced from an opensource JSON file which I am now pre-processing and identifying all collisions on it.

There are so many collisions!

Could I pick your brains and you could critique my approach? Thanks

I agree to an extent

My project is quite well documented and I created a Prompt a while back along with some mermaid diagrams

https://github.com/tomaytotomato/location4j/tree/master/docs

I can't remember the exact prompt I gave to the LLM but I gave it a Github issue ticket and description.

After several iterations it fixed the issue, but my unit tests failed in other areas. I decided to abort it because I think my opinionated code was clashing with the LLM's solution.

The LLM's solution would probably be more technically correct, but because I don't do l33tcode or memorise how to implement Trie or BST my code does it my way. Maybe I just need to force the LLM to do it my way and ignore the other solutions?

I updated my original comment to explain my reasoning a bit more clearly.

Essentially I ask an LLM to look at a project and it just sees the current state of the codebase, it doesn't see the iterations and hacks and refactors and reverts.

It also doesn't see the first functionality I wrote for it at v1.

This could indeed be solved by giving the LLM a git log and telling it a story, but that might not solve my issue?

I have been trying to use Claude code to help improve my opensource Java NLP location library.

However trying to get it to do anything other than optimise code or fix small issues it struggles. It struggles with high level abstract issues.

For example I currently have an issue with ambiguity collisions e.g.

Input: "California"

Output: "California, Missouri"

California is a state but also city in Missouri - https://github.com/tomaytotomato/location4j/issues/44

I asked Claude several times to resolve this ambiguity and it suggested various prioritisation strategies etc. however the resulting changes broke other functionality in my library.

In the end I am redesigning my library from scratch with minimal AI input. Why? because I started the project without the help of AI a few years back, I designed it to solve a problem but that problem and nuanced programming decisions seem to not be respected by LLMs (LLMs dont care about the story, they just care about the current state of the code)

I wanted to resist a pun, but that would be a drop in the ocean.

Seeding a large body of water with radioactive chemicals would take a ship quite a while.

Same, I found this video explainer here:

https://www.youtube.com/watch?v=Tnij_xHEnXc

Whenever I see Gaussian, I think of the Gauss gun from Half Life 2

Fair enough, I can add some screenshots but would that be more useful than just letting people click on the app they are interested in and then trying the theme?

Like the author I was fortunate enough to be exposed to Dilbert as a teenager, before I got caught up in the rush of the university-professional-yuppie-industrial-complex.

I found the Dilbert principle book in my parents downstairs cloakroom (wedged between magazines and other generic bathroom reading material).

At a superficial level I just read the comic strips in the book and laughed, I thought to myself - haha look at those poor corporate workers, that won't happen to me.

In a way it didn't happen to me vis-a-vis cubicles, suits and water cooler gossip, TPS reports etc.

However, in other ways it did happen to me, the frustrations of working with incompetent people, working in teams who brainwash themselves that they are making something useful or being productive, hilarious executive decisions made without any scientific or rational thought. (startup - https://youtu.be/iwan0xJ_irU)

I still like to add Dilbert comic strips to closing slides in presentations, my go to one is this, when we are discussing new technologies to use.

https://tenor.com/nJfQSXLP8am.gif

We are in the Dilbert universe, it just keeps changing

p.s. if anyone is looking for Saturday TV binge material, all of the Dilbert TV show is on Youtube here https://www.youtube.com/watch?v=nH7dgUq5Qe4

Indeed, it makes you wonder how prevalent this psychological malaise is in the world of teaching and academia.

I don't want any future children of mine, to have self loathing/pessimism or "woe is me" feelings taught by teachers or lecturers.

Self reflection yes, abstract and critical thinking yes, expressing feelings yes.

No - "sorry the world is burning, I think you should be sad about this and maybe reconsider being an Engineer".

The Cuban missile crises I would say was a lot less precarious than Able Archer or the 1983 Soviet nuclear false alarm alert - which was averted, by, ahem - an engineer!

https://en.wikipedia.org/wiki/1983_Soviet_nuclear_false_alar...

I think its a pessimistic outlook and I agree with the sentiment, but then I switch back to objectively how humanity has been historically and how far we have come, I can't stop thinking, "wow".

Being in my 30s I remember Y2K, OZone layer diminishing and a rogue comet coming to wipe out humanity, but it didn't. This is survivor bias just like the examples in the lecture around wildfires and Covid are surely survivor bias too.

My wife does not like when I solve problems instead of just acknowledge the problem and say "that's a shame/sad/terrible", but I can't help it, we as engineers are wired to do solve problems, not just acknowledge them.

Think of the Dog poo dilemma - most people will just point and say, "terrible someone has let their dog poo there". Then proceed to carry on with their day. My engineer brain says lets pick up the poo and then look at solutions to stop it happening again.

So when a crises happens I know there are lots of smarter men and women in my field and other areas, who won't just get sad about an issue and instead will start working their brains on the problem.

The apocalypse is delayed, permanently.

I am exploring various levels of prompting in Github Copilot at Org, Repo and Personal level.

Think of it as an exploration manual.

That could be a big potato battery bank?

According to google a 200g potato give off about half a volt (0.5v) and 0.2mah

    4000 tonnes = 4,000,000 kg = 4,000,000,000 g

    num potatoes = 4,000,000,000 / 200 = 20,000,000 potatoes

    volts = 20,000,000 x 0.5v = 10,000,000 volts (10megavolts)

current would stay the same at 0.2mah

I am not an electrical engineer, what could we do with this?

That "guy" on Youtube is Jamie Oliver, a world class chef who made a documentary on processed food in American diets.

https://www.youtube.com/watch?v=mKwL5G5HbGA

Some people find him a bit too preachy or wears his heart on his sleeve. However his battles for healthier food in schools is commendable.

RIP Scott Adams

I recently purchased his 2026 calendar for a family member who works in the consultancy world, they really enjoy it.

Starting my dev career in a big corporate telecom I used to attach Dilbert strips to the end of my presentations, sometimes people would laugh, others, normally execs didn't get the irony or commentary.

Then when I went to more modern and cool startups the same Dilbert comic strips still apply which I found hilarious.

A lot of our influences or heroes have faults and I hope we can all put them to rest and just remember Scott's great achievements with Dilbert and his many books on Management or Psychology.

I will just leave this scene from the Dilbert TV show, that describes the engineering curse:

"The Knack" - https://www.youtube.com/watch?v=g8vHhgh6oM0