Hacker News Reader

Comment by jottinger | Hacker News Reader

Dang, that hurt to read. I'm starting up a new news-ish site like the old TheServerSide.com, at https://bytecode.news, and I'm faced with the question of "how do I generate traffic in the face of AI and all the people willing to market, market, astroturf, market, market?" I'm not that kind of personality, I don't want to do tiktok or whatever the kids do, I'd far rather accept organic and slow growth over meteoric and unsustainable and undeserved success, even if "organic and slow growth" means failure in the end.

View on HN

Your comment reminds me of a quote by the Dutch artist M.C. Escher, that has helped in my lows

> How slowly one advances in a boat that does not float along with the stream in a specific direction! How much easier it is when one can connect with the work of great predecessors whose value is not doubted by anyone. A personal experiment, a construction whose foundations one must dig himself and whose walls one must erect himself, runs a real risk of becoming a humble hovel. But perhaps one prefers to live there rather than in a palace that has been built by others. (Escher on Escher – Exploring the Infinite)

That's an excellent quote and I love it. I think of it more as Thoreau, or in my own terms, but I LOVE that quote, which I'd never seen before.

Top comment of a frontpage post, you're not doing that bad at marketing.

That can be automated quite easily these days. Just make your bot/scraper scan for anything that blames something or otherwise fits narratives that can be used as a vehicle of your own promotion, and hit that.

P.S. Not necessarily implying that the grandparent resorted to that.

Please keep in mind that such a use is against the Guidelines[0] and will be downvoted and flagged rather quickly.

Since 2023 I always check the creation date of a user before I click on any link in their comment.

[0]: https://news.ycombinator.com/newsguidelines.html

>will be downvoted and flagged rather quickly.

You wish. It is becoming harder every day to find genuine comments or technical insights at the top of HN sections instead of blind love/hate for trendy topics.

It's turtles all the way down.

Even among the "genuine" comments most of them are just people spewing things because the monkey brain has determined that it likes it when the number in the top right goes up and has figured out what kind of things to spew to make it go up (dare I say encouraging this state of affairs is the reason some platforms choose to keep score in such a manner).

Does it really matter whether you're reading something written by an AI shill, a human shill or a human (or AI) who's repeating the shillery? You're being shilled at all the same.

Ooof! If I'd thought of that, I'd have kept my trap shut. :(

Art as expression, not as market campaigns, will still capture our imaginations - that's more my driving force than trying to sell more hotdogs.

I absolutely adore it when stumbling occasionally upon great musicians who have laughable numbers of listens on various platforms, dunno if that's mere chance or the algorithm trying to balance things out, but that surely brings some hope for the pure art.

Trusted news orgs around the world tend to have centuries of history behind them. If you actually did it right, it should long outlive you, with the speed of growth being largely irrelevant.

Looking at your site, you have a mission statement in your About page that roughly says "scrape other sources for anything relevant and interesting." Nothing about staffing, who you are, who your editors are, whether you even have editors. Every story seems to have the byline "DREAMREAL," which doesn't sound like a person.

It doesn't seem to me like you're interested in running a news org. It seems like you're dissatisfied with your ability to find things interesting to you in a single place and are trying to scratch that itch, probably in a mostly-automated way. I can sympathize with that, but a personal knowledge base with outgoing links to the original sources isn't a news org. By all means, share it. Maybe six other people in the world have exactly the same interests you do, but this is a far cry from journalism and you probably shouldn't frame it that way.

All fair, but: it's not a "news org." News is a business model, and, man, my business is feeding my family; this site's a passion project, fulfilling a need based on what I did back in the mid-2000s: curating stuff that developers roughly adjacent to my skills (i.e., JVM-adjacent) would find interesting. What happened, where, who, why, maybe how.

The site's actually not the main focus - the site's still being developed. The hardest thing about TheServerSide back in the day wasn't the writing or curating - although curation is hard if you're not just echoing press releases or READMEs - it was discovery.

Slashdot, freshmeat, RSS, IRC back then... all just being watched and participated in (I'm an actual developer first, after all) - talking to and with and watching ALL THE SOURCES. I had an OPML to die for, man! ... and I had to read it all, and filter it all, and ignore it when people talked about stuff that readers wouldn't care about.

So BCN's actually an infobot that happens to feed that discovery, in part - it's actually an information stream that yields factoids (infobot, woo!) and other services around that, and part of that yield is information about what's happening, so it HAS an RSS reader, it has github webhooks, it has everything I can think of as input streams, not as a comb filter but just as an information hub that can contain comb filters.

Yeah, there's no marketing; I'm no marketer. Don't want to be. And the presentation of the model is confusing, because I'm no marketer and because the readers don't and really shouldn't care, if it works. And the site's "in process" anyway - I mean, yeesh, it only has 50 posts so far! I'm working on getting there.

And "dreamreal" is a person - it's me. That's the IRC handle I've used for years, and that's actually one of the admin identities for me the bot had first, and it stuck. Easy to change, but honestly, it's not about me and shouldn't be.

But you're not wrong in the slightest.

ETA: it hit me after I wrote this comment that I was posting it under a name derived from my real name, to defend the use of "dreamreal" on the site to someone who posted THEIR OWN comment under "nonameiguess" - which would be an odd name to have in real life. :D

This is great and I wish you all the best. A byproduct of the content abundance age (because that's really what is is) is the expectation that not just growth but fast growth is everything is such a race to lower quality on the whole. It's pretty depressing but ultimately I suspect we will get sick of the lower level far quicker than we probably think.

Thanks. I decided long ago that I wasn't the kind of magnetic personality who'd be able to market a mass site - as a "face" I'm a very poor candidate, and I despise cults of personality, which seems to be the "go-to" for a lot of such things. It's the modern version of "smile more!"

I hope you're right. Not for my own sake, but for everyone's. I'm trying to do what I can to put stuff worth reading out there - and whether it's worth reading or not isn't actually mine to judge; if I make it "worth reading" by SEO terms, it's not actually worth reading all that much, being neutered and hedged to the point of milquetoast oblivion.

IMO.

I like your work but I'll echo others : the title font is too narrow and hurts readability. I've advise for a regular Garamond.

The entire site looks like typical AI design

This is also fair: it's definitely AI-assisted. I'm not a UI person; I have wonky color perception and synaesthesia, and a lot of experience has taught me that what works for ME in UI is what... uh... doesn't work AT ALL for other people.

So it's a problem: I can "hand write it" and have you say "mein Gott, who came up with THIS rubbish? Are they blind? Are they trying to blind US?" or I can have an AI spin something up that looks fairly bland and normalish, but fulfills the specifications I require, and have users go "this looks like a blog that pretty much anyone would write," except the content was always meant to be the differentiator.

And the UI is actually designed to be pluggable, anyway; what I work on is the backend REST services, and the UI goes through those services for all data - it has no access to the underlying datastreams, at all. Everything goes through https://api.bytecode.news, and the intent was that if someone was an awesome UI coder and wanted to show their stuff, they'd write their UI using CoolThing and I'd put it into the system at coolthing.bytecode.news, and if it totally dominated the "normal UI" I'd promote that to the default UI and just leave the other one out there as a reference (the UI you see is also hosted at https://nextjs.bytecode.news, as the "canonical url" for that implementation.)

That way, if someone's going "ooo I could make Struts sing with this" for some reason, well, the API endpoints are published with OpenAPI; go for it. Show the world.

The UI was always secondary to the concepts that drive the information system, and the content was meant to be more important than how the content looked. I'd rather deliver information than flash.

I didn't mean any disrespect. I just think AI design is the opposite of what you're saying it is. It's flashy and distracting and filled with unnecessary animations that often hinder readability. I think a much simpler design would be beneficial to the readability.

Whether that's accomplished with LLM tools or not

Also fair. I kinda thought the design for the UI was fairly normal - it looks like a lot of standard Wordpress themes would look, with an admin panel, a primary focused "recent post" and a list of other recent posts in descending order of attention. From a UX perspective, I thought it was okay; the animated circuit thing was my idea, but I tried to keep it VERY in the background, I wanted it to be fun and not a focus.

If you have ideas about what a proper UX would look like, I'm all ... eyes? ears? UX is not my thing and I am willing to consider breaking the molds to make something more awesomer if I can.

SEO, Reddit and Paid ads probably

I'm not going to optimize content for SEO - been on the other side of that, and I think it creates content that's bland and ineffective. Humans gonna human, machines gonna machine, and I'm not paying for ads. If humans want the site to succeed, it'll succeed. Otherwise, it won't. Such is life.

I kind of see what you are saying, but it reminds me of “if you build it, they will come.”

They won’t come, because they won’t even know about it. A more accurate aphorism would have been “if you build it and tell everyone about it, some of them might come.”

Humans probably don’t want the site to succeed, because they mostly don’t know it exists.

Perhaps.

How much SEO happens here on HN? How much do they spend to tell everyone about it? I'm guessing: Not much; maybe zero.

But people come here, anyway.

(That doesn't mean that it's capable of independently sustaining itself, but people do show up.)

> How much do they spend to tell everyone about it?

You could see every YCombinator investment as a kind of sidelong marketing for HN.

If you build it, and it's genuinely, uniquely useful, you won't be able to stop people from coming and inviting others to join.

Telling everyone about it is only necessary if you're indistinguishable from 100 different takes on the same thing, and trying to win a shouting match (this includes hurrying to shout the world down before competitors get a chance).

But as the saying goes, you are not in traffic - you are the traffic. The reason you need to shout is because of people like you shouting.

It is different. If you SEO, you are actively changing content toward worst and less interesting. That is not the same as making the decision between market and not-market.

This man will build a product no one uses and then complain about the world not being ready for his genius instead of using the tools that everyone else uses to give himself a chance at success.

In all honesty and writing it straight: this kind of success you're writing of would best be not achieved at all, as it's poisoning society and civilization.

I'm not talking the success of a Facebook or an Amazon. I think to successfully make an impact in terms of helping fellow humans, reach is a multiplier just as important as the quality of the product.

Hrm. If you're talking about me: I doubt it. I've been a programmer for 30+ years: I've built a LOT of products, some of which got used, some of which ... didn't. If I was going to complain, that'd be new.

And I know how to do SEO and why. And I'm just not interested. "Success" for me... well, I mean, I'd love to retire today based on the brazilians of dollars some mogul hands me for the IP behind the site, but, uh, that's not what it's for. SEO at the service layer? Sure, it has a mode for the bots, it tries to make scanning easy for the machines. But at the content level... nah, I have been watching the world tune information - one aspect of the psyop the OP talked about - for decades, and I can't bring myself to do it.

But also don't squander the traffic with default dark mode with low contrast ;) Use the device mode hint at least

The font for the headlines also looks 'off' to me. Letters too close together maybe

Hmm, It's supposed to be using the device mode. I'll look at the UI code and see if I can fix that. I use dark mode and my eyes work well with the framing, so any information here is welcomed. Will tune.

Also: one of the features of the underlying tech stack is that the UI is actually a completely separate module than the backend API. Different github repo, everything: the site is akshually https://api.bytecode.news and the UI at the "base url" is just one layer for that API. The goal was always to have multiple interfaces, all tuned for whatever the technology supported best: see https://enigmastation.com/2014/07/09/repost-some-of-what-id-... (from my personal blog).

BCN actually has multiple UIs for it, although I've been concentrating on the content rather than the front-ends - I'm not a front-end guy, so my UIs tend to be impenetrable. The model works. If it's my implementation the other front ends will suck from a UX perspective, because I don't see the UI the same way most people do (there are reasons, they're not important, bottom line is that I avoid UI.)

What would YOU suggest for a UI? I'm very much curious, because I would LOVE for the UI to sing and I would have no idea how to make that happen.

Ah that explains why it was so slow to load.

Not sure your reply was meant for me but: just make as much as static as possible

What's your load chain? I've seen some other people (who use Tor through Brave) say that it's slow to load, and I'm curious to fix that - the front end has a cached version that should be rendering pretty aggressively without datafeed updates.

"Dang that hurt to read, here's an advertisement!"

Why is this every 3rd comment on this site? Every single post has multiple comments that are

"I hate that, here's my solution"

Market-by-contradiction is huge these days. Almost as big as We're-Evil-So-What marketing!

> "I hate that, here's my solution"

HN comments have always been like that, or rather "here's my approach to circumvent that annoyance".

What's your alternative? I prefer that style over comments that stop already at "I hate that", as curiosity should be more than an expression of a dismissive opinion.

This is the second comment like that. It was honestly not intended - it was actually me trying to respond in my own context to the post, and WHY I was responding; "why" is more important to me than "what," and I have a hard time saying "this" without explaining why I felt "this" - or at least trying to explore it. I'm working on a site; reading the "this kind of thing is how people do things these days" actually made the concept of working on the site (which I hope has value) harder. That's what I was saying, or trying to; the "advertisement" was context, not intent, or else I'd have tried to post my own thing to HN saying "look at how ossum my seite are!"

Choosing to fail means you shouldn't begin the project in the first place.

Correspondingly, if you are beginning the project, you should not make choices that will result in failure.

I don't think i'm "choosing to fail," I'm choosing to accept the outcome given an effort to prevent it. Sometimes you try and it doesn't work out; I'm not committing the mortgage to the site or the effort, and I'd like it to pan out because I think its progenitor had a raison d'etre, and I was part of it when it was good and I think there's room for it now.

And if the moment's gone, well... that's the way it goes. That's not the same as "choosing to fail."

Reply on news.ycombinator.com