💾 Archived View for gemini.bunburya.eu › newsgroups › gemini › messages › 875yq0rdrr.fsf@haraya.loca… captured on 2024-05-26 at 15:26:28. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2022-04-28)
-=-=-=-=-=-=-
From: rtr <rtr@haraya.invalid>
Subject: Re: In the news
Date: Mon, 31 Jan 2022 07:43:04 +0800
Message-ID: <875yq0rdrr.fsf@haraya.local.net>
David <david@arch.invalid> writes:
On 30/01/2022 01:22, rtr wrote:
> David writes:
>
>> Article in The Register about Gemini with lots of links
>> https://www.theregister.com/2022/01/27/gemini_protocol/
> Also what's with this fella:
>
>> A post this week gained a lot of traction on Hackernews forums, when a
>> software engineer calling themselves "マリウス" – that's "Marius" to
>> gaijin – called it "solutionism at its worst". They argued in a blogpost
>> that Gemini is an answer to a problem that doesn't exist and encourages
>> a bunker effect, excluding people who use ordinary web browsers, perhaps
>> due to accessibility issues.
> I'd argue that gemini is an answer to a problem that does exist. And
> frankly, I don't think it was ever an issue for geminauts whether people
> would or would not be excluded from gemini. It exists by itself as an
> independent means of obtaining information. For that alone, it is really
> great.
Indeed! Suppose Gemini built on HTTP(S) instead of the current TOFU
approach, and used AsciiDoc, MarkDown or whatever as a default text
format ...
- Would http(s):plain/markdown have a catchy name, say Mercury?
- Would such beautiful clients exist like Amfora and Lagrange?
- Would there be concerns about privacy that do not exist with Gemini?
- Would there be a tangible community around it, a Wikipedia page?
This newsgroup?
I doubt it. Or is it just luck, that Gemini became quite popular?
Maybe it's as simple as that: a blog post by an influential hacker
once got enough clicks on Hackernews. At least that's how I got to
know of Gemini, if I remember correctly.
Honestly, I don't think it's just luck. I think the reason why gemini
became ``popular'' is because it tried to do something different rather
than expand on HTTP(S) for the nth time. I think this is the reason why
the author's arguments missed the entire point of gemini. It was never
about ``giving access to everyone'' it's about giving them a choice.
The reason why I'm using gemini is because it's convenient for me to
keep a blog in gemini than in a normal LAMP/LEMP stack. My VPS just has
a gemini directory that I rsync from my local machine. The thought of
having my blog being accessible to everyone and their grandma never
crossed in me. If I wanted that I would have hosted my blog in HTTP
instead of gemini.
> I took the liberty to skim this fella's blog here:
> https://xn--gckvb8fzb.com/gemini-is-solutionism-at-its-worst/
> It seemed to me that he is a developer of something called
> Superhighway84 which seemed to be a USENET clone from I read about it.
> On that, I find it funny that he is talking about Gemini being ``an
> answer to a problem that doesn't exist'' when he himself have created
> ``an answer to a problem that doesn't exist'' with his Superhighway84.
Superhighway84 seems similar to Usenet regarding the organisation of
messages with groups, posts and threads, but data transfer is
different:
- The data of the messages is kept locally in a database (OrbitDB).
- Changes to the database are propagated to and from connected peers.
- Peers are connect via the IPFS protocol.
- There is no central server (or few servers as in Usenet?).
- When installing IPFS on your machine, the software comes with a set
of pre-configured peer addresses to connect to initially
(bootstrap).
So there's no single point of failure, which is impressive in itself.
Are there other notable uses of IPFS, would sci-hub be one (or is it)?
I think the idea underneath is the same. It's just that its using
IPFS instead of NNTP. USENET is also decentralized, it's just that most
of us don't bother to install INN and just connect to a USENET
provider. Also, IPFS is a weird hodgepodge protocol.
Funnily enough, I was reading a lot about IPFS a few days ago because I
thought it's a neat protocol with all the fancy modern web crap that
they've pulled in their website. But then I came across this guy's blog:
https://fiatjaf.com/d5031e5b.html
Which promptly killed all of my enthusiasm with IPFS. It seems like a
half-baked protocol that sounds neat in theory but is quite bad in
practice. I don't want my files having anything to do with something
like that.
--
Give them an inch and they will take a mile.
--
gemini://rtr.kalayaan.xyz
Parent:
Re: In the news (by David <david@arch.invalid> on Sun, 30 Jan 2022 11:02:47 +0100)
Start of thread:
In the news (by David <david@arch.invalid> on Sat, 29 Jan 2022 19:07:21 +0100)
Children:
Re: In the news (by Dan Purgert <dan@djph.net> on Mon, 31 Jan 2022 10:16:17 -0000 (UTC))