Show HN: Tilion – Stealth Browser Infrastructure for Agents
twitter.com
twitter.com
Please don't use HN primarily for promotion. It's ok to post your own stuff part of the time, but the primary use of the site should be for curiosity.
I’m having success right now with Reddit and YouTube. I think HN may be over for good work unless you are protected/blessed/not downweighted by mods.
I love your promos video - how did you put that together? Vibing with the retro internet story.
Also i am building something with similar goals: https://duetbrowser.com/dub-cli
I’m very glad you got on the show page. Many of my recent posts that got the same vote/time rate with similar min stories on show never escaped new - indicating flag or mod suppression - even tho there’s zero rational reason and no good reason for anyone to do that. Good work should be shown not hidden by personal bias of a few
I’m not a noob - i got the biggest show HN of last year, but the system, or the experience of some accounts including mine has radically changed in the last few months. So far the mods are saying they don’t know anything about outright downweighting, but have not shown data that supports that. What I’ve seen very much supports that
Good luck with your post!
Of course we want to see “good work”; we're working constantly to develop new ways of identifying it. And the moderators are not going to down-weight work that is “good”; if we did that we'd be hurting HN and YC. Most of what it submitted to Show HN, we don't see, because so much is submitted that it's impossible to review it all.
What we are looking for is evidence of credibility, substance, depth and novelty, and it's actually not very hard to convey that if it exists; you just need to make make the effort to write a good post that tells the story of why and how you built the project and what you overcame and learned along the way.
What doesn’t scan is i don’t make shit. I regularly hit front over the last ten years. Now the posts that get community signal are held back, and not a single one of my high-quality works has hit HN front page in six months. Why are you doing this?
Also i suggest for posterity you prevent the comments in our thread here from being flagged invisible. Let’s keep it open and fair. The sole exception to the above ‘ban’ was so egregiously flagged and misrepresented in comments it led to GitHub disabling the projects’ repo. Mods did not step in to protect or police the bad comments. Night and day with before.
Edit: edit fair enough on looking for a new ways to identify signal, but to be honest it seems like you guys already have a good signal of interestingness. I mean it worked before this year, and it also seems the ability to consistently hit the front page should be a good signal of ability to generate interesting content, but you seem to be penalizing that in my case?
Thanks for the back and forth
Further, HN exposure only has any value if the interestingness level of content that makes it to the front page is high. If the interestingness drops – and there have been many complaints this year that the interestingness has dropped due to AI code generation and AI-generated writing – then the value of HN drops for everyone, including everyone who wants their project to be on the front page, including you.
> “System is already perfect, if your work is not getting up it must be shit”?
I didn't say the system is perfect, I literally wrote “we're working constantly to develop new ways of identifying” good work. And I didn't say work (by you or anyone) that doesn't make it to the front page is “shit”.
The question that matters is what is interesting to the HN audience.
Yes, you had some big hits on the front page, and we gave you plenty of support with that. It's also the case that after a while, your subsequent projects were minor variations on the same theme, and some members of the audience found them repetitive.
It's always the case that when an individual developer submits multiple variants of a similar concept, their exposure has to be reduced, because repetitiveness is bad for HN and also for the developer. We've often seen that developers who submit similar projects over and over get ignored or resented. It’s unclear which is worse, but neither is good.
Projects that are interesting, novel and offer some gem of insight for the audience to learn about will always have place on HN.
That refers to the comedy series of last year (https://news.ycombinator.com/item?id=46205632 https://news.ycombinator.com/item?id=46326588#46329337 Etc), not any of this year’s CLI/tooling innovations, so any penalizing of the former should not apply to the latter.
Saying the TUI browser, operable browser REPL for agents, microfilm viewer, AOT language, and others are minor variations of a “theme” is gross misrepresentation that has no idea about the depth and quality of those, and how they are all distinct - and is not an appropriate reason to hold them back.
The comedy posts were definitely variations on a theme, the work this year is distinct and not minor variations on a theme.
Resubmissions of high quality work (that hasn’t yet hit the front page) consistently achieves front page placement, unless interfered with. When you hold back my posts, you invite more resubmissions for work which, if unmolested, would have already hit the front page. Why are you guys doing this?
I imagine you get vocal email complaints from folks who don’t want to see my work on the front page for whatever reason. Is that true?
> Projects that are interesting, novel and offer some gem of insight for the audience to learn about will always have place on HN.
My work has all that - as can be seen by looking at the work, looking at the organic reaction it gets before it’s prevented from hitting the front page when it should, and by looking back at how I consistently hit the front page over 10 years here - but you guys are holding it back now, why?
I think you’re repeatedly and falsely trying to make it about creator quality when that’s not justified - because it’s not a question in the case of my work here. It already is very quality. Something else is going on, what is it?
> Something else is going on, what is it?
The world is changing fast and everyone has to adapt.
> A project that is built with AI in a few days or even weeks does not cut it anymore
That does not characterize my work - it’s typically months of work.
But you’re correct that I haven’t been writing detailed intros. I create stuff people can use and I prefer that experience to speak for itself, but … it’s necessary to intro it here, I think.
I have emailed you just now for that for reference.
> You’re not being downweighted or flagged or penalized.
Thanks for saying that, but where is the data to support that? It does not match my experience here.
The thing is my posts are being seen: six or more times over the last few months a new project I post will get 6 to 8 points on the new page, but it won’t graduate to show or front even though that puts it well above the lowest on the page. Or if rarely it gets to show it’ll remain in the dead zone of position six and below where votes barely accrue.
That suggests to me either consistent flagging or downweighting you’ve baked in or extended from the downweighting dang says here was happening for the comedy followups end of last year: https://news.ycombinator.com/item?id=46326588#46329337
I have been using this website for a long time and it doesn’t feel to me like it’s simply a case of you need to clear a higher bar.
I feel the raised bar is part of it - but maybe you can look at the data of what I’ve posted, and whether it has been flagged consistently and by who or somehow caught in some unintended downweighting. It really does feel like that. Maybe you can have a look?
Thank you for the back and forth
If a project has already been on HN and has set off a flamewar, you don't get a better outcome by submitting the same project under a different URL the following week.
You take the feedback and think about how to improve the product or the messaging, then try again when it's changed enough to be worthy of another shot at the front page.
also, the post last wekk was about an open soruce engine, this is the infra layer that enables 1000x of what that did. on benchmarks regarding stealth, speed and usability, we now beat all major providers. what
hn is goated precisely because ppl here are real, and the last weeks feedback helped us grow a lot. we are hoping that exponentiation keeps happening.