I want to like this, but there's something conflicting about "thoughtful content" and being forced to throw it away (or have your post become extremely long).

If I put thought and effort into writing something relatively long-form (otherwise, is this just single-post twitter?), I don't want it to disappear, I want it to be archived when I post something new.

Still, doing stuff is cool, so great job! Also, I like the design of the site :)

Reading and writing on HN, I see (not always obviously) some very good and thoughtful posts. Most of these are relevant to the present in which they are written, and quickly get buried under new stuff. So I don't think it's a big stretch to just have old ones disappear

I do, as I like to bookmark stuff that I find interesting to share or look up again later. Having it disappear at a whim with the only hope being that it was archived such that I can access it makes the web a bit less meaningful to me. Finding out that something's gone after looking for it is already happening often enough, there's no need to increase that.

Thanks for the kind words :)

I get your point, but the idea is to strike a balance between something being meaningful and relevant. As the user, you are in charge of that - deciding what to keep and what to delete. That adds a taste of FOMO to the site that in my opinion makes it even more interesting.

> That adds a taste of FOMO...

If your aim is to avoid the dopamine, I think encoraging FOMO is a bit self-defeating.

Perhaps limiting posts to daily, weekly, or monthly might make sense? Letting people work on a queue with scheduled posts could encourage thoughtfulness.

It's writing. A text file. Archive it yourself.

Or do you mean archive it but leave it visible? That seems to not be the point of the website.

If the site has some git repo backing, it can remember the past.

Made me think fondly of the days I would update my .plan

https://en.wikipedia.org/wiki/Finger_(protocol)

I can't tell whether it's a coincidental invention and the author doesn't know about finger or it's a clone and an intentional omission.

Considering how long it's been since plan files have been relevant, the former is far more likely.

It's well documented even in popular tech books like the Doom book.

Direct experience isn't needed.

Related story. I had read extensively about Xerox Alto, Bravo, smalltalk-80 etc and it wasn't until my late 30s in Seattle at CHM that I actually saw one of the machines in real life. It was like meeting a celebrity I knew intimately about. All the things I had read about were there. Really marvelous.

I've heard of finger and remember finding it cool. The memory might've lived in my subconscious while making this site, but it wasn't something I had in mind while making this site.

No shame. Good ideas get reinvented and implemented frequently.

You can almost always find analogs to the most popular contemporary things dating back hundreds of years.

Hell look at European news parlors in the 1800s. The newspapers would be delivered by train and "influencers" would be waiting at the station to rush to a specific popular cafe where they would hop up on the table, read the headlines and give their hot takes. Audiences would follow them by coming around daily as merchants hung out near the best ones selling produce.

Sound familiar?

The deeper you look into the past the more it's just the same stuff reinvented. It's crazy how similar some of the stuff is

Did finger make it past the days when identity was by machine?

https://datatracker.ietf.org/doc/rfc1288/

The copy installed on recent MacOS is: $FreeBSD: src/usr.bin/finger/finger.c,v 1.36 2005/09/19 10:11:46 dds Exp $......

There was a short-lived site that I was presented, back in 2015, that had a similar idea. A user's 'home' page was a simple 4x4 grid and in each of the cells you could add pretty much any kind of content you wanted. Long form posts (truncated, on the home page, but could be clicked through to read in full in a modal), images (or galleries), videos, embedded content, etc. You had full agency to change any of the 16 things at any time.

There was also a "feed" page which just showed any cells that had been updated by users you follow, which made the feed page pretty much like any other social media.

I found it kind of charming that you had this very limited number of things you could be "saying" at any one time. You couldn't represent yourself with your entire history, but you could be quite expressive with just 16 cells of freedom.

I don't remember the site, nor can I find anything that seems to be what I remember, now. For some reason, I have the idea that it was killed off in a wave of "ephemeral" apps failing. But I don't know why I think that, so /shrug.

In any case, I find your idea to be charming in a similar way. I do like the idea of ephemeral representations, as opposed to a historical ledger of your entire usage with a site. And I also like the idea of, essentially, controlling your "headline" to such an extent that there are no (or very few) other options for people if they want to engage with your content at all. It's like "This is what I'm interested in, today. Please talk about it with me.", with an implication of "I can't focus on stuff from last week, even if you're only discovering it today, because of my new post."

For me, at least, that's a reasonable offering that I hope gains some traction!

Was this site you were thinking of (linking to avc post since it's been shuttered for a while)? https://avc.com/2015/06/pegg/

Woah! Nice find! I can't say for certain if that's it, but it certainly looks similar. Same minimalist (blank white) format, at least.

That said, I never saw the site I had seen on mobile, so I don't know if it looked more like a mortar layout on a phone. And no one ever demoed that kind of chat functionality. So it might be a similar thing? Or it might just be an evolution of what I had seen. Either way, Pegg looks pretty cool. Sorry to see it didn't take off (either?).

Isn't this (basically) what geocities was? A single "post" (webpage) that you can edit? Users were able to browse and search for other geocities pages.

I mean... you're eliding a lot of stuff with that "basically", but I suppose you could contrive that analogy?

Geocities didn't do a "global" or "following" feed with every user's content updates being represented as a reading list. Nor did it simplify the site's content management with "article"-based fracturing so that only relevant changes were presented to relevant parties.

Geocities is "basically" this other site in the same way that a peanut butter sandwich is "basically" a hamburger. Apps, as with entrees, can be so different as to be unrecognizable, just in their expression, much less their contents/functionality. But, by all means, if the Geocities framing helps you contextualize it better, more power to you!

You could have multiple pages on Geocities and push updates via FTP.

And it lives on in design at least through NeoCities [1]

[1] - https://neocities.org/

Technically this is not really a post, but more like a static page. If you call it a post that assumes I can create new ones, it has a sense of repetition and being part of a series.

There's also a CSS issue with links in the content, they are not readable with the default blue color.

Yeah, the home page when I loaded it had comments from the left column stretched over the page where I couldn't read it. They when I went to the join most of the text was white-on-white so I couldn't read it. I think finger has a better interface at this point.

Maybe it should be called a 'PUT' as it is idempotent?

Thanks for creating this website and well done.

I journal in long form documents rather than create lots of small files, so this website is suitable for me. I use GitHub's markdown editor or IntelliJ Markdown editor to update my journal. I use markdown headings and number them. Each journal entry goes to the bottom. GitHub's website has a table of contents button in the top left of README.md. You can search the headings which I think is helpful.

I think people could use this approach of markdown headings for each journal entry when updating posts on Pressn't

When I get to an arbitrary number of entries, I create a new GitHub repository and repeat the process and append an incremented number to the repository name. Then I share that repository of journal entries online and that journal only receives updates or occasional rewriting but never new entries. I link to old entries from future journal entries within the document (GitHub headings are linkable, if you hover over them you can copy a URL) or across repositories.

(Links in my profile)

Do you generate anchor links for each markdown heading? This might be useful to link into long form documents.

I just read through some of your “Ideas” repo. Congrats on being so prolific.

You can make a table of contents by using [TOC], and you can make a link to any heading using it as a #id link:

https://app.pressnt.net/post/4/#so-the-tldr-here-is

The markdown is based on python-markdown. Here’s the documentation on the TOC extension I’m using:

https://python-markdown.github.io/extensions/toc/

Thats pretty-damn creative

Nice idea.

The color for links seems to be the default in most browsers: blue for links and purple for visited links. These colors are extremely hard to read against the site's background color. Are you able to style links so they are easier to read against the site's background color?

Aha! I was wondered why people were commenting how the default link colour - which has been in existence for what, 30 years now? has suddenly become unreadable.

It's dark mode! The abomination that will never darken my doorstep, nor my monitor.

I'm definitely in a minority on this one.

The default link color wasn't made for dark mode, therefore it stands to reason that it will not have an optimal readability then. It hasn't suddenly become unreadable, the requirements changed. And under the new requirements it is not readable for many people. I wonder if calling a color scheme that arguably reduces eye-strain and increases battery life on modern monitors an abomination is a good idea. These kinds of stances usually result in unconstructive arguments.

Luckily we aren't constrained to a dark mode, or a light mode for that matter. So everyone can choose what they consider the best for them.

Actually, I prefer 100% of my sites to be in dark mode!!!

should be 'dark mode first'

https://i.imgur.com/NopbI8t.png

https://i.imgur.com/Eqjxo1U.png

Specifically, the site seems to follow the system's preference for light mode or dark mode. The link style against the dark mode background is very difficult to read.

Yeah, my dark theme is broken right now :( I’ll try to fix it later today

what it looks like: https://imgur.com/a/KGfQzva

Reminds me a bit of a social network concept of mine from years back: "Karachter"

"Imagine a social network where you only need, in fact you are only allowed, one friend. Imaging a messaging service so personal, so patient, and yes, so damned meaningful, that it only transmits one single character, to your one friend, each day. "

http://randomfeelsgood.blogspot.com/2014/02/announcing-karac...

One of these days we'll find that One Weird Trick For A Good Social Network.

Looks terrible on my phone. Social happens mostly on phones.

Yep, I agree. I’ll probably spend the next week improving mobile and Dark mode css.

Same, half the text is off screen.

It’s probably because someone made a post with a huge title lol.

I’ll add ellipsis overflow later

Oof, you are not wrong.

One nitpick about the responsive design—between width 1535 and about 1920, the Comments box appears in the left rail but it’s partly obscured. Seems like you should either adjust the media query to keep it at the bottom until there’s room to show it, or make the box itself a variable width.

Thanks for articulating that. I'm not a web dev, so all I noticed was that on my phone, the left-edge text was truncated, and zooming didn't fix it.

I don’t think this solves one of the huge problems with culture on the internet right now, which is stacks vs queues

with only a few exceptions[1], all social media, including HN, present information to you in a stack, most recent first, mostly with no option to change that. this forces everything to be shallow, surface-level and understandable to beginners, otherwise you don’t attract new viewers or followers

even beyond losing their attention spans, people are being trained out of learning things in depth. of building upon knowledge. it’s sad and it’s probably going to be damaging in years to come

[1] - youtube playlists are the only exception that jumps to mind, but even youtube is driving hard in the other direction, removing “sort by oldest” from channels and heavily pushing shorts

Could you expand on this? I struggle to think of how a site like HN or Reddit could work in a first-in-first-out fashion while also taking into account vote score. Perhaps you mean that a post's score should not decay over time and fall off the front page, or at least should do so more slowly? Or it should be able to be bumped to the front if there's new activity (comments/votes)? Would be interest to hear your thoughts

Not OP, but you could have a priority queue where you acknowledge each element as up/down/neutral and the change in score that you provide would affect it's location in other people's queues.

well there could be more scope for playlist-style content. this especially applies to reddit and HN (and twitter I suppose, although I’d probably exempt Twitter from my criticism in the first place as it’s supposed to be a bit ephemeral and shallow). users could collate content into “playlists” which would then be voted into a no-or-slow-decay alternative frontpage

in order to allow new content to rise, you’d either need a pretty complex algorithm, or just another “new” page

I had hoped Elon would work on things like this.

maybe he will. I doubt it though

See also /now pages[0] (a federated version of your gist points 1 and 2, i.e. creating and updating a post about what you are doing now)?

[0] https://nownownow.com/about

This is a really interesting idea to me. I've considered a site where you can only post once per day or per week, something like that - but this seems like a better approach. One interesting feature to add later might be the ability to see change history on other people's posts.

I wish pg or John Carmack had a page here so I could check what their post was.

People have compared this to .plan, but seem to forget that back then people often wanted to cram more and more in their plan. I had a job that would cat new data every 10 seconds, rotating my calendar and other info.

I don't like that information in general would be ephemeral (and screencast everywhere sucks), persistent URLs are one of the great things about the web (and archive.org...) but maybe you could show different visitors different URLs? (I'm reminded of jwz's site sending HN referrals somewhere different.)

To borrow another idea: choose your own adventure. The articles you've read (up/down voted?) previously alter the ones you get next time. Heck, feed every article into GPT and ask it to tweak it for the reader.

This made me think of AIM away messages. Probably some matrix of technical ability and age could determine your point of comparison.

I am building something similar (https://20-things.com) where the idea is to have the frontpage only display 20 things. No paging, no endless scrolling. Good luck with your project!

This is what .plan files we’re back in the day.

I like people trying stuff like this. The web is too large and too broad for everyone to hangout in the same place. There should be communities and sub communities that are distributed. We used to have forums, but maybe it'll end up being thousands of niche social media sites like yours. Kudos on the launch!

FYI, Got a 500 server error on https://app.pressnt.net/post/create/#edit-here-to-create-you...

Can you try again? You may have had bad luck as I broke something for like 5 seconds earlier today

How did you know about my bad luck? :)

It did work this time.

PSA: dark mode is kinda broken. I’ll work on it later today. The site respects system preferences so this is why it appears broken to some. Sorry about that.

I was wondering why I couldn't see the labels. If I double click to highlight the text, I can make out each one at a time.

https://app.pressnt.net/register/

I managed to join. https://app.pressnt.net/post/26/

site doesn't render right on FF

https://i.imgur.com/6N8j3SH.png

Lol that’s interesting. Turns put my dark theme is broken. I’ll try to fix ot after work today.

Be Real

Minutiea (founder is on HN sometimes)

we’ll see which one of these concepts really pick up

yes, there is a burgeoning trend of rejecting the digital social dopamine addiction as well as posting more authentic things. I think its a travesty that founders have to identify both the problem and the solution as if they are clairvoyant geniuses in both regards. But good luck.

Why one? How about, say, three posts. That seems it would still capture the spirit of the idea, and obtain the benefits of the constraint without it being overbearing.

What if someone is interested in several completely separate topics; rolling them into a post with unrelated sections seems poor.

Does the like count reset when you update? Seems like that could be a little dangerous if it doesn't.

Yeah you are right. I’m still trying to get my head around what the best ranking algorithm is. So far the factors I’ve thought of are:

- Update time

- Unique Visitors since last update

- Upvotes

- Upvotes since last update (I didn’t consider this one so thanks!)

I used this for a bit and I really like it. This could take off. It's fun!

People are okay with contributing content that's temporary as long as it doesn't take them a lot of effort, i.e. not all that meaningful, such as typing a sentence in a chat.

I kind of wish it didn't have editing. Or could only be appended, never revised. But still have upvotes or likes or whatever.

You have one shot at making a highly upvoted comment. go!

A post is not supposed to mutate. Supposed you have a post I liked and replied then you changed it completely; my reply would be totally out of context.

I'm reminded of two things: (1) XKCD's IRC channel, which had a bot that kicked you (with an exponential timeout) if you ever repeated something someone else had said before. I think the idea was to encourage originality. (I suspect it actually encouraged strings of random numbers on the end of sentences).

(2) Finding the perfect journal - the right cover, weight of paper, lines, everything - then never writing in it because it's too good to write in. "You only get one chance at this" is a high bar to clear. I like it, though.

I suggest making your website mobile-friendly. I can't read it from my smartphone.

Congrats on shipping!

Do comments get cleared when you edit? Otherwise the connect context gets lost, right?

That’s a good one. Context will indeed get lost…

I’ll consider clearing comments after edits

I wouldn't mind seeing a trend of BlogHN added to the ask, show, and tell posts.

Remember when this was called .plan? Pepperidge Farm remembers.

Seems completely broken on mobile?

Great project! Is it open source?

.plan files as a service

This is great.

So... finger?

Sorry to say it so bluntly, but this sounds like classic "tech solution to a people problem" to me.

At least to me, the restriction "only one post, but you can update it" seems pretty arbitrary to me. It's not clear to me why updating a post should lead to more meaningful content than creating a new post - or why following a stream of updates should somehow be less of a dopamine treadmill than following a stream of posts.

I think this is more of an experiment, or even an interactive art piece, than an attempted competitor to the mainstream. It doesn't need to be perfect, just thought-provoking.

More than anything, is what I’m spending my free time in. It does have some sort of “agenda”, but for the time being it’s nowhere near mastodon, twitter or even neocities.

I think this is potentially a good idea for a blockchain app. You could use the Ethereum blockchain as your backend rather than a database. That way you don't have to host it, and it wouldn't go down. You also wouldn't have to store usernames and passwords, users would just authenticate using their key pair through something like MetaMask. Users would also need to pay a small fee to send a transaction to update their post, possibly making their content more meaningful.

I only have this idea because the Ethereum "hello world" app I'm working on is something similar but even more reductive: just a single shared blob of text that anyone can update. I don't think it has utility, it's just to learn the technology.