Although I agree with most of what you write... you do realize that you use
a platform that goes a bit beyond what you describe above...lol
Yes and I believe the modern web was a big mistake which will be extremely difficult to undo.
We should ideally use specific client programs where possible instead of relying on the web, but the damage is already done and now it is "convenient" because the entire infrastructure is built around these obese browsers.
Ideally the web would have stayed static and we would be accessing services through specialized client programs such that the browser would only launch external programs and it would have been just as convenient without it being monolithic, but only if the infrastructure is structured as such. As an example, when browsing to retrobbs.i2p in an alternate reality, the page would redirect you to nntp://retrobbs.i2p:119 and the browser would launch an NNTP client and let it handle the rest.
I believe we need a small, lean and efficient web alternative that does one thing and does it well. Might as well solve the whole privacy fiasco while we are at it. The main hurdle would be to get others on board and get them to offer their websites in an alternative format (other than html/css/js), which will be extremely difficult.
Maybe due to all the smart phone hype some people might get on board if sites load and render in milliseconds instead of seconds.
Posted on RetroBBS
firstly: what is wrong with html ?
secondly: what is wrong with css ?
Both too complicated for my use case. I would also avoid HTML, simply because of the overhead. Look up deserialization performance of XML, then imagine that you would also have the overhead of having to keep track of incorrect HTML and correct it (fuzzy parsing). Using existing formats also come with certain expectations. A developer will think "It's HTML! I know HTML!", then be gravely disappointed because none of the modern features are implemented, leaving a bad aftertaste.
I will probably go for a simple data format that can be compiled down to an efficient and specific binary representation, which can then be deserialized fast and efficiently on end-user clients. The protocol in the place of HTTP would probably be a simple one too, think something like Gopher.
thirdly: what you describe sounds nice, but there is no profit in it, so
count out all commercial players. if you want text only, there are newsgroups.
There is no profit in it and that is OK. I would already be happy with it if a tiny community could form around it. I want basically a bit of rich text with linking. It should ideally become something that anyone could implement in a few weeks, not an entire lifetime. Stylistically it should look modern, yet simple, something like this: http://bettermotherfuckingwebsite.com/
Usenet is great, but it is interlinking and easy use of multimedia what really sells the web (which could be achieved with launching external programs...). One could probably retrofit a usenet client to do this but that is more effort than is worth.
I have a rough draft document that I am currently working on. I need to sort through all my ideas and come to a conclusion.
Posted on RetroBBS
-----BEGIN PGP SIGNED MESSAGE-----
well, sounds good to me, although i am not sure about the ratio between
effort and gain here.
We unfortunately don't have a real alternative to the web, so from my perspective it's worth a shot.
I also do not think it is _that_ huge of an undertaking, I'm keeping it simple and uncomplicated to give others the chance to write their own implementations. I want to avoid a one-implementation standard in the long run, if possible...
feel free to use /rocksolid/programming to publish your code, or apply for
your own forum if you want something with a bit more options than a thread.
Thank you, I'll let you know if I need something more than a thread. I think with the way Usenet does threading it shouldn't be an issue to discuss ideas independently of one another easily.
would be cool to have a converter for normal (=bloated) websites to a nice
clean text feed, stripping out all the annoying, unnecessary and dangerous stuff.
I am still deciding in which direction to take this, but this would be something I would also like to implement.
I'll use this opportunity to also post my public GPG key, to prevent imposters down the line. I don't know how likely that is, but I'll just post it here anyhow.
- -----BEGIN PGP PUBLIC KEY BLOCK-----
Comment: User-ID: crowbar <firstname.lastname@example.org>
Comment: Created: 26/10/2019 14:17
Comment: Expires: 01/01/2030 12:00
Comment: Type: 4096-bit RSA (secret key available)
Comment: Usage: Signing, Encryption, Certifying User-IDs
Comment: Fingerprint: A4460A9DE1AC2FB8DD21723427DBFF1D0324D351
- -----END PGP PUBLIC KEY BLOCK-----
-----BEGIN PGP SIGNATURE-----
-----END PGP SIGNATURE-----
Posted on RetroBBS
while i agree with your priorities, i think you need to define a bit more on which level you want to work. there is a lot of code already, addressing all three targets that you listed on different levels.
reinventing everything seems pointless imo, so you need to decide what protocols you are going to use to achieve your purpose, and at which point you will implement your own code.
i guess this will be part of your stacK:
then on top you will parse the content and do something with it.
just guessing here, really...
Posted on def2