HNHacker News
TopNewBestAskShowJobs

adrianwaj

2,066 karma · joined May 17, 2007

future dash secured dot com slash contact
submissionscomments
adrianwaj··on Show HN: A techno machine in one HTML file, with verifiable renders
I got my reader going by running Caddy and pointing my browser at: http://localhost:8080/reader.html

Imported about 750 feeds. Still buggy.

A lot of fiddling. Have to run PHP, PHP-FHM, install CURL, and messed around with configuration files.

Also needs a PHP proxy script to run alongside the HTML and ensure URI encoding.

Not easy done but can now disable CORS Unblock in the browser. Yet now these extra things have to run as services all the time.

Sidepage doesn't seem quite right for me at this stage but thanks for the ideas. It looks like it could do the same job, though. As long as an AI can provide detailed instructions and troubleshooting, it'll lower the barriers-to-entry.

I think the future may well be a modified browser with CORS Unblocking built-in (or some type of proxy alternative.) Have it offer 1000s of free "page-apps" from the get-go. Easy development and ubiquitous platform is the key. Offline button clearly marked and traffic/bandwidth data collection and window.

P2P and micropayments stuff could be interesting.

Overall, the downfall of feeds has either pulled people into the walled gardens or pushed them offline completely. Look at the new Commodore minimalist phone.

Yet having 1GB+ databases possible in browsers by default is something that could pave the way for some cool stuff. https://rxdb.info/articles/indexeddb-max-storage-limit.html

adrianwaj··on Show HN: A techno machine in one HTML file, with verifiable renders
Great point.

The unique thing about HTML apps is that there could be a "directory download" of say the best 1000 apps every month for people to run locally with imports also downloaded and security audited beforehand. That would be ideal.

GreasyFork checks for security problems and best practices in their scripts with community moderation a large component. https://share.gemini.google/VwomSQkChNWS

How I see it, there could be a code freeze period (quarantine) and if no issues arise, the HTML is bundled into the 1000 monthly thing.

adrianwaj··on Show HN: Chrome extension that gives you the TlDR from YouTube without the watch
No probs.

Yes. Ideally that box would be in a sidebar. On that note: Gemini Chat Sidebar, Sidebar with Gemini and the official Gemini in Chrome all come to mind, for example.

Also, have you seen these: https://greasyfork.org/en/scripts/by-site/youtube.com?site=y...

Seems like a crowded space, that could be a good thing if you differentiate well.

adrianwaj··on Show HN: A techno machine in one HTML file, with verifiable renders
At the moment, there is a toggle switch between own proxy and direct (requiring CORS Unblock extension,) but really having multiple methods can work.

Other mentions were using a CloudFlare worker proxy (needing Cloudflare account and Node.js), NeutralinoJS, Electron and Tauri.

Speaking of XML, there must be a way to push items out from a server using XML. That's an interesting solution and would let a remote server do all the polling and as such would mean sharing data with it. This could end up a paid option using micropayments and give it efficient performance in an ideal world.

adrianwaj··on Ask HN: Those making $500/month on side projects in 2026 – Show and tell
I like the idea.

I am building an RSS reader and like the idea of being able to tie articles back to verified individuals who have a crypto address and contact information. Even having encoded images would be interesting. "Oh, that's the guy with the funny duck picture on everything he makes." - But it's actually a steganography identifier.

So building a DNS for people... is just having an email address enough?

So imagine you are a freelance writer and at the bottom of an article you put some unique string or pic that leads people and agents back to a home page that contains a crypto address and contact details.

One application: even here on HN if some article gets over say 100 points, how would it be possible to invite the author into the forum for an AMA type discussion. You'd check the article for some type of identifier string first.

Also, with an RSS reader, an author or feed owner could be invited to claim a direct connection with their readership and end up collecting tips from them.

I'd use the term "Atom Reader," but someone may think I'm talking about a quantum physics laboratory. At least RSS has no similes.

adrianwaj··on Ask HN: Coding is a solved problem. What is left for experienced engineers?
"I have been an engineer for years. At this point, I do not see much value in sitting alone and building more projects just to prove that I can build software."

There is one massive thing positive about being independent: your code stays unique to you. If any AI is training off you (especially by stealth,) then it is making your solutions commodities, right?

I am building an RSS reader that I like a lot. But anyone that builds one after me, and especially if they use the AIs that I used, I presume will get there so much more quickly. If I have shared freely with the LLM, then from there it's open to anyone indirectly whether it is OSS or not. Also, the reader is an HTML page - as soon as it is public, anyone can get it.

So the playing field is being leveled for software, it's worthwhile thinking about new uses for software. For example, ways to tip writers and solving the ever present "micropayments" conundrum. Things can atrophy sure unless you look for new problems, new frictions and ways to improve life. Local-first Fediverse is a huge opportunity and enabling safe whistle-blowing while minimizing noise. Enable bravery, but not too much.

adrianwaj··on Show HN: Chrome extension that gives you the TlDR from YouTube without the watch
It's a good idea. Also just getting the top few comments of a video grants ketchup without any inference.

Have you seen DeepMark... https://news.ycombinator.com/item?id=49288896 ?

Looks well done - people may bookmark vids and want to use all their YT bookmarks as a source for the sauce. Get all the reading done in one shot.

Lastly, could a bookmarklet prompt an AI to summarize a video by pre-filling a prompt? I suppose that's what is being done here, but via hover windows.

adrianwaj··on Show HN: Dev tools in one static site, no framework, no server
Great idea, I just mentioned an idea for a site "thereisanHTMLforthat" [https://news.ycombinator.com/item?id=49425010]

Maybe you could host/archive such single pages for other people, especially if there was a way to automatically obtain the latest version from some known location?

You could collect stats, ratings and comments for such pages, like with GreasyFork.

Another idea would be to record the prompts that were used, or can be used to create these tools. For example, I just created one myself because I couldn't find anything else to do it. Probably quicker to do than keep searching for the right one.

https://share.gemini.google/Varmn9aYtGze (URL to JSON exporter from one small prompt)

adrianwaj··on Show HN: A techno machine in one HTML file, with verifiable renders
Good point. GlassBox also appears as single page. https://news.ycombinator.com/item?id=49421948

I am working on an RSS reader that is also single page and location-agnostic. Issue for me right now is CORS unblocking as that does entail some complexity to workaround. Current length of the HTML is 2500 lines / 101KiB.

Moreover, just like https://theresanaiforthat.com/ - a great thing to see would be something like thereisanHTMLforthat.com - because quality vibe-coding delivers!

Also, a single HTML can easily be security audited, right? How much damage can it do?

adrianwaj··on Ask HN: How do you keep up with HN these days?
Super cool if threads would indicate whether an author is participating.

("Author here") https://hn.algolia.com/?dateRange=pastMonth&page=0&prefix=tr...

The thread takes on an AMA (AskMeAnything) vibe.

I suppose it'd be possible to do that programmatically as a third-party service by checking for socials and then automatically notifying author of a thread - perhaps once it gets popular. Orange username?

adrianwaj··on Ask HN: How do you keep up with HN these days?
I often use my HackerTrain script.

https://greasyfork.org/en/scripts/589441-hackertrain-best-co...

Then with it installed, I go to https://news.ycombinator.com/bestcomments which gets reformatted. If I like comments there, I can read the underlying article having already setup some context beforehand. It's like going to a carefully selected film after having read the plot and spoilers - after reading many in a row. This keeps things interesting and allows for more efficient learning.

Because 2 hours is a long time to sit through something that doesn't really interest me. Online, 2 minutes is also a long time. If you're a senior manager, you use your direct reports to bring things to your attention. Introduces perspective, and seeing with 2+ eyes is a lot better than only using 1.

The rest of the time I use my other script: HackerBrain. https://greasyfork.org/en/scripts/588721-hackerbrain-sidebar...

Most of the time, a headline alone is pretty much useless and that's what vanilla HN overlooks.

adrianwaj··on Show HN: Deepmark – Search your bookmarks by what's inside them, not the title
"As of January 2026, custom search engines can no longer query more than 50 domains."

https://en.wikipedia.org/wiki/Google_Programmable_Search_Eng...

"All" you really need is a search engine, and then be able to customize it for each user with a set of URLs that they consider important. Use that to search/query for them.

Can a user run a conventional search?

How is it distributed? Where does the training take place? Are there minimum hardware requirements?

adrianwaj··on Where did the old web go? We followed 657,607 links to find out
For me the new web came about when if you wanted a website, you hired a web developer. As soon as it become DIY, without your own domain and didn't involve looking at HTML, the new web was born.

Today, if you own your own domain and fiddle with HTML then basically you're still part of the old web.

So basically, the old web required experts to help people get online, the new web negated them through homogenization.

The real experts from the old web basically became startup founders to monetize the homogenization and automation of their worldviews for the less tech-savvy (or specialized) folks.

People pine for the old web, but it was a churning, friggin' mess too. Search results were good though, but only when Google arrived. The standardization and ease-of-use from the new web has been a massive boon. I think things will change as people yearn for a wild-west, where there are new opportunities. The promise of a better life, seeing that the grass is greener, avoiding pain - those things always create golden ages.

Once AI favours people who can afford to pay for the best results.. or surveillance becomes too burdensome or dangerous, then watch this space.

The spirit of freedom always runs into trouble, when considering the notion of:

"Cut off the heads of others to make oneself appear taller."

Surveillance enables this. Decentralize again, but how?

ISPs - help or hindrance? Why/how can ISP charges be diverted to content creators?

adrianwaj··on Show HN: Deepmark – Search your bookmarks by what's inside them, not the title
Wouldn't mind seeing a "search inside your upvotes" thing for HN.

Good idea on the whole.. could end up being implemented by the browsers themselves over time.

I personally think pages themselves, if they're in the cache should be easily viewable when offline - even videos and music if they're in the cache too. Don't see why it hasn't yet happened with browser caches ballooning out so easily. "Cache Viewer with search." No need to depend on bookmarks or deliberate actions.

In fact, with the local-first Fediverse, I see friends and connected people building up copies of the web themselves and only going out to "publically fetch" when the "aggregate local cache" needs updating. It could be called "stealth communities." You could even tell a page to pingback if it is updated and then all stealth communities get notified of the change for either refetch or are pushed the diff patch.

Furthermore, I have been thinking about how to host a webpage locally but have the whole world be able to access it without performance degradation in any hug-of-death scenario - cascading stealth communities are the answer. I might only send out 5 diff patches, but those changes cascade down some pyramid structure that takes up the load.

I suspect this would have a benefit to censorship and anonymity too - I just haven't thought it through. The goal is to be able to change something from anonymous to identified and vice versa. It would be so cool to release some information anonymously, but be able to de-cloak once a threshold has been reached and safety-in-numbers could be obtained - especially after user feedback, which may be anonymous too. Smart contracts that can "auto un-redact."

With stealth communities, it would act as a decentralized archive.org alternative over time. I'm sure crypto could be worked into it too - micropayments for data exchanges. People's ISP fees - can't that be distributed to hosts and communities? That would definitely be an improvement and a viable monetization strategy, right? The whole ISP model could change. Start local and work outward.

ISPs know everything about their customers, and VPNs are conglomerated too from what I've read. Something to think about. It leaves people open to cyber gang-stalking by people up the chain in the system. Do you want to have your economic and social activity usurped while being blind-sided the whole time? Are there any places left where people can get and stay online anonymously? Why? When doesn't encryption work?

adrianwaj··on Ask HN: What are you working on? (August 2026)
This is a good idea, if you like top-level comments, also check out HackerBrain.

https://greasyfork.org/en/scripts/588721-hackerbrain-sidebar... (user-script created 2 weeks ago)

Comment reading is done with a sidebar from the listing pages without needing to go to individual item pages.

adrianwaj··on Hardware backdoors in some x86 CPUs
Another thing that can be done is more crowdfunded bounty programs.

5 here: https://cybersecuritynews.com/bug-bounty-platforms/

The issue are "bugs" could be randomly distributed, even across the same version number of some device.

Sample sizes and statistics come into play at that point.

Whistleblower protections are an avenue, but people can still be dealt with harshly (Schulte) and degree-of-protection can come down to motive. But motive itself is multivariate, is it not? A local-first Fediverse, with some type of "guaranteed anonymity" would work, though. But anonymity doesn't mean something can't be a hoax either, so it becomes a signal vs noise issue at that point. Yet, "nothing totally secure ever really is."

Source information can be embedded in the period of a printed sentence too.

A moral world is the answer to many problems. Something to ponder. People are said to only see the errors in their ways after death in the "hall of mirrors."

-- https://web.archive.org/web/20060102095331/http://maitreya-e...

Economic espionage is something to also think about. You may be doing everything right and that's what makes you a target. "If I'm not top dog, target anyone that is."

Perfect anonymity in crypto can also aid whistleblowing so that dump data = get paid.

adrianwaj··on What happens if an entire class of workers loses faith in their careers
I find technology's evolution to be a continual recursive loop of "intelligence building on intelligence" to create systems that extract increasing amounts of energy from the environment with increasingly smaller input. And to me coding has now been better integrated into the tech stack. But still a long way to go.

With coding, so far it has really pushed the use-case for English too and people have reverted back to using language for its more conventional patterns. Coders have stopped coding, so shouldn't that be a relief? If only we could stop using cars too! But code is gradually taking over the driver's seat.

Hunter-gatherer societies were very advanced in many ways just by being in tune with natural processes in order to support life. They didn't even need to drive cars! Long-live the camp fire!

adrianwaj··on The LLMs Problems
There are a lot of ideas on user-script sites with regards to what people are changing on their preferred destinations using client-side scripting. I mentioned that in my previous comment.
adrianwaj··on Show HN: Your next startup, ready in the blink of an eye
Seems to be a lot of overlap between ideas. I just ran a prompt:

"From GreasyFork scripts for YouTube.com - what are users most interested in seeing on Youtube, but requiring a script to run for that to happen?"

https://share.gemini.google/fgxPtnl5mYmr

The I asked it about DoorDash - a business featured on Blink.

There's a lot of intel that can be gleaned from a business' website and what people are modifying on it with scripts. I just published two scripts myself so it's been my focus for the last couple of weeks.

Probably a good time to build a YT competitor and offer RSS feeds for channels as they've just been cancelled and also somehow incorporate crowdfunding through micropayments. Tangent suggestion.

Anyhow, I think the flow for a business is - fill a hole in the market - obtain users - build their trust and confidence - ask what their problems are - solve them - $$$

PG: "When you go talk to users, don't (just) ask them what features they want. Ask them what problems they have. In the best case this will lead you to think of features they'll love but would never have thought to ask for. And expect to go through multiple cycles of ask-and-build."

https://x.com/paulg/status/2084655582938808691?ref_src=twsrc...

Anyway, IMHO the site seems quite wordy, would be great to see diagrams and the text visuals could have more contrast. Can create "optimal industry landscape maps"?

Would be interesting to have "I like this idea" and enable team-building right from the get-go.

Also, solving problems are only one dimension of a business case. It can also be "preventing the end of something good" and "improving quality of life." ~ Guy Kawasaki. Are both those problems to be solved?

How can you build up a database of reliable users?

Would be good to have people list prior employers and roles, and then send out email to companies and have them approve of that person's history there. Need to find real people and skills amongst the robots and scammers. Build up a public moderator log of users having their histories confirmed. https://greasyfork.org/en/moderator_actions If you say you've worked at IBM, someone should be able to confirm that from an @ibm.com email address.

Those moderator logs (see above) are crazy. So rare to see one like that in public. Must be a business in aggregating and automating them. Important, tedious, repetitive and relentless work.

adrianwaj··on Ask HN: What are you working on? (August 2026)
Just published two scripts for reading HN.

https://greasyfork.org/en/scripts?by=1628132 HackerBrain and HackerTrain

Absolutely astonished at how useful ChatGPT and Gemini have been in coding this up. Couldn't be happier. But there's a certain level of tech knowledge one requires to get a good result. The more you know what you want, from a variety of perspectives, the better the result.

adrianwaj··on Ask HN: What are you working on? (August 2026)
KM looks interesting. I just had this idea that some day people will be using some type of unified front end, which has input connections that change depending on what plugged into it prompting for input.

https://news.ycombinator.com/item?id=49151189

With regards to KM, I wonder if you could collect information in realtime upon entry into those sites rather than taking archives of prior activity. I suppose it'd add too much complexity without any promise of meaningful return... yet.

At some point it could allow input into new alternative services instead of the standard ones - it could look something like https://pico.sh/.

adrianwaj··on Atom is better than RSS, in ways that matter
"Gemin" I'd support an argument to bring back the "I" to computing. The gem in I.

Perhaps humans need their own APIs for others to then mash-up in the way they please?

For example, I think the goal is to merge all entry fields into a local one that manages and monitors output. I should be able to see and record a stream of everything I do online but I can't. That stream, being a representation of me, is what's valuable. Straits of Hormuz on a person-by-person basis.

How many entry fields are there that you use? email, video, social, crypto, HN, chat, apps... the list goes on. There's a lot of duplicated effort there. Why so many?

I think the whole "One device to rule them all" that I mentioned here [1] is about putting a layer around yourself and letting others interface with you on your terms.

[1] https://news.ycombinator.com/item?id=48912226

local-first >> individual first

adrianwaj··on Atom is better than RSS, in ways that matter
https://share.gemini.google/aHYGlJzhFHSx

https://imgur.com/PQICt6I

RSS or Atom - do they indicate whether they are providing a full or partial feed?

What is federated content?

Can the Fediverse be local-first?

adrianwaj··on Atom is better than RSS, in ways that matter
Both really need an explicit "Full/Partial" flag.

Over-polling is also a problem with both formats.

But the Local-First Fediverse is evolving. https://share.gemini.google/NHjiUHtpRTiJ

That's important and will break the walled gardens, and will change money flows I suspect as advertising revenue is more equitably shared with content providers, who themselves can be better paid for content produced and content promised. It's the 'ol centralization <> decentralization swinging back and forth again.

adrianwaj··on Show HN: Reply to Your Comments with an API
Is it possible to reverse the flow and pull feeds from all services into a single one for reading: like the old FriendFeed. The innovation could be unified feed sharing - see another person's unified feed that at first would seem like spying. The whole thing could even revolve around rss at first before the integrations.

The power of a service like this is slotting in your own alternatives to the underlying services. Like supermarkets building their own brands around bestsellers. You could wean people out of walled gardens. "What's the Mallary alternative to this?"

"When FriendFeed was active (2007–2015), major social networks provided open, unrestricted APIs that made aggregated timeline streaming straightforward. Modern platforms like X (Twitter), Instagram, and Reddit have locked down their APIs."

https://share.gemini.google/wsPvn2L4gZK1

Reverse the trend. "Use open standards but compete on implementation" ... from the old Sun Micro playbook.

See Surf:

"Combine people and posts from Bluesky, Mastodon, Threads, Flipboard, YouTube, and more to build custom feeds for your interests. Add a hashtag and others can join your feed!" https://surf.social/

adrianwaj··on Show HN: I was tired of opening 2 tabs for every HN link, so I made a userscript
I did something similar 2 days ago with ViolentMonkey if you only want to check high-level discussion without any new tabs at all.

https://greasyfork.s3.us-east-2.amazonaws.com/sokzpb2ztmatlh... (screenshot)

https://greasyfork.org/en/scripts/588721-hacker-news-comment...

I'd improve it by counting top level comments next to items but that would slow it down. Still need to make some slight changes like closing sidebar by re-clicking orange (beyond just right-clicking sidebar.)

It's very fast and speeds up thinking overall.

adrianwaj··on Show HN: Tilde Pay – Give your AI agent a bank account to pay for things
anonymous twin - if you're ever hacked, make sure the attacker can't identify users. Don't ask for emails or phone numbers. Provide users with 10 login strings that must be used sequentially. To login, use #1 string first + PIN and as soon as #2 string + PIN is used, #1 is automatically cancelled. They are not one-time logins. Can be used repeatedly.

Have an emergency passcode to be used for a new set of 10 after a fortnight time delay that neutralizes the previous 10. I am just making this up - I don't even know if any of this is truly necessary.

adrianwaj··on Show HN: Pilot Protocol – a network where AI agents find tools and each other
I wonder if an agent verification/authenticity/identity thing would work. I could have my doppelganger living in cyberspace doing my bidding without me ever necessarily knowing about its complete activity. Or just a swarm of branded agents building up a reputation that can be directly connected to the real world or at least verifiably connected to a domain name.
adrianwaj··on Show HN: Tilde Pay – Give your AI agent a bank account to pay for things
Homepage is nothing but "Sign in to continue" .. wow. Perhaps run an anonymous twin of the entire service too. See how they diverge over time. Expect massive legal and KYC stuff too. Maybe don't use a bank account anywhere and just accept direct crypto deposits?

Knowing where funds are spent can be very valuable market research. Useful in due-diligence. Best of luck.

adrianwaj··on Show HN: Pilot Protocol – a network where AI agents find tools and each other
Who owns the agents?
← PreviousPage 3 of 34Next →