Has anyone proposed CommonMark over HTTP? Like, just shove markdown in a GET response with Content-Type: text/markdown And let the client decide how to render it. It'd be like Gopher, but modern.
Does Gemini[0] look interesting to you?
The "Cheap" Web
131–140 of 230 posts
Re: The "Cheap" Web
#132Has anyone proposed CommonMark over HTTP? Like, just shove markdown in a GET response with Content-Type: text/markdown And let the client decide how to render it. It'd be like Gopher, but modern.
“But but but.. how do we shove ads everywhere and track the targets?”
Re: The "Cheap" Web
#133Has anyone proposed CommonMark over HTTP? Like, just shove markdown in a GET response with Content-Type: text/markdown And let the client decide how to render it. It'd be like Gopher, but modern.
Does Gemini[0] look interesting to you? [0]: https://en.wikipedia.org/wiki/Gemini_(protocol)
Gemini isn't over HTTP and it's not Markdown.
Re: The "Cheap" Web
#134> Cheap to maintain: Most webpages should work indefinitely without falling over. Simple HTML will do this, so long as a browser that supports that version of HTML and CSS is still around. A static Go binary serving dynamic HTML will do this. I struggle to imagine Python, Ruby, etc being able to accomplish this though. That's to say, I'm not sure that's a good rule or could use some reworking to make it more achievab…
Once upon a time side bars and tool bars were made up of iframes that could talk to each other. Those sites wouldn't work today for "security reasons"
Re: The "Cheap" Web
#135Has anyone proposed CommonMark over HTTP? Like, just shove markdown in a GET response with Content-Type: text/markdown And let the client decide how to render it. It'd be like Gopher, but modern.
Does Gemini[0] look interesting to you? [0]: https://en.wikipedia.org/wiki/Gemini_(protocol)
Gemini is like modern Gopher, but it is also very opinionated. The protocol does not support extensions and revision, it refuses HTTP, mandates TLS, etc...
That's not the same as having browsers support the markdown content type.
Re: The "Cheap" Web
#136Earlier quoted context omitted.
Don't visit the website
I hope we can have a more reasoned debate than this. This topic is really important and frankly there's so much to it I hope that hearing others I can learn more. When I read some of the thread above what I am hearing is: "I don't want people to be able to run programs on my computer without asking explicitly". Which is exactly what the modern web and modern web browsers enable. If you learned computing in the 90's y…
"Don't visit the website" is as simple as it needs to be.
This is HN. 90% of the time, the nitpicking here isn't about computer security, bandwidth efficiency, free speech, or whatever the high-minded principle of the day might be. People nitpick because they don't like something about the website, full stop.
It it wasn't the design, it would be the ideological bent of the writer. If not that, it would be annoyance as to why they put up an email signup form or use an analytics script, as if those aren't present on any of the other websites that are featured on the front page every day.
I'd wager that a good chunk wouldn't be happy unless they can extract your article into their homemmade RSS reader.
Re: The "Cheap" Web
#137Earlier quoted context omitted.
I mean, that's because Facebook Pages are cheap. Businesses don't want to spend money on anything they don't have to, and people aren't much different. The argument for learning to write even basic HTML only to upload it even to something like Amazon S3 (free) and direct a web address to it (not free) or spend money on hosting to not have to learn how to deposit files into an S3 bucket is a LOT of time to spend to ac…
The other conclusion I've come to is that web designers way overcomplicate stuff. So I don't want to do a Facebook page. What are my relatively low-cost options? Well, I can have a really basic HTML site but that's a little late-90s. OK, a Wordpress site then. Now I'm dealing with introductory rates that shoot up after a year unless I just use a standard VPS and now I'm having to do a lot of stuff myself. And, as far…
> The other conclusion I've come to is that web designers way overcomplicate stuff.
I mean, there's a floor right? The floor being that you need a server that can respond to HTTP requests, yours or otherwise; you need said server accessible from an address to which you can map a DNS record to; and you need HTML files that can be served in response to that request. Few of these are free, but many can be had for very cheap.
> So I don't want to do a Facebook page. What are my relatively low-cost options?
Amazon S3 with Cloudflare in front of it and a domain name. Probably about $15 per year unless you manage enough traffic to get billed by Amazon.
> Well, I can have a really basic HTML site but that's a little late-90s.
I think that's a little reductive. HTML sites can be just about anything and they can be quite fancy if you're willing to put the time in to learn something like Jekyll. As a side hustle I maintain a couple of basic HTML websites for local businesses, and just bill them for the annuals + the time I spend tweaking things for them, which thanks to my workflows, isn't much.
> OK, a Wordpress site then. .... I'm willing to spend some money but I haven't really found an answer I love.
I mean, that's just the platform-ification of the Internet described. There's a floor of technical requirements and software experience required to publish on the Internet, and the vast majority of people like yourself do not possess it and aren't interested in acquiring it. That's fine, but it's akin to saying "I want to print a newsletter but don't want to learn how printers work." You've selected yourself out of your own goal.
HTML, CSS and a bit of JavaScript can make dynamic and beautiful websites. But it's not without a learning curve, and if your response to that is "well I want to do this without learning HTML/CSS/JS" well, that's the central conceit of Facebook pages, Shopify websites, and Wordpress.
Re: The "Cheap" Web
#138This is why we should be using decentralized XMPP servers to chat since they actually run on potato hardware unlike other chat options.
Google Chat on my 2008 Blackberry (XMPP client) was the fastest IM experience I've ever had. I finally understood the meaning of "Crackberry" when I had that setup. The combination of the instantaneous speed and keyboard was something I haven't experienced since.
Re: The "Cheap" Web
#139Reading through I got this: - Cheap = a pain to read Good and modern visual design exists explicitly to deliver information to a person in the most accessible way. If you show a Japanese website to someone who grew up in Ireland, they will be overwhelmed. Not catering to different cultures, like the website suggests - only makes it worse. - Cheap = a lot of spam Having low cost of equipment and almost unlimited anony…
"Good" modern visual design barely exists at all according to that definition. Most modern web design exists to jam marketing in your eyeballs while drip-feeding information at the minimum rate to keep you from closing the tab, which is basically the opposite of "delivering information in the most accessible way".
Also: The YouTube homepage uses twice the memory of the NicoNico homepage to show fewer thumbnails, and has basically been unchanged for years, while YouTube can barely resist keeping their layouts the same for a single month. Japanese web design has quite a few issues with being stuck in the 2000s but I'll take consistently bad over rolling the A/B testing dice every week to see what page I get.
Re: The "Cheap" Web
#140Earlier quoted context omitted.
They're hard to find because they're not really linked anywhere. We used to have Web directories arranged by subject headings that tried to be mostly comprehensive and put some serious curation effort into that goal, such as DMOZ.org - but there's no modern equivalent to that. People like to complain about how the whole SEO issue has made search results useless as of late, but that if anything is downstream of the ov…
That's so true, it's almost like a return to the beginning (Yahoo Web Directory). We need to find a way to curate content that scales but in a way that eliminates the incentive for bad actors to try SEO-like approaches to game the process. That feels impossible without a feasible alternative to the ad-based revenue model.
Directories died because
a) Websites change hands frequently and enshittify rapidly.
b) Nobody was getting paid to check in on websites and make sure they were still on-topic
c) Most 'good' websites don't update every day, and people hate RSS and email signups. So activity and engagement on those sites are completely overshadowed by the bottomless pit of content that is social media.