Comment by ToJans
4 years ago
Mostly web, as that's my area of expertise.
- Start of the movement from cloud/SaaS computing/storage to edge devices/client-side/on-prem.
- Amount of new successful SaaS orgs will decimate. The only 2 SaaS markets that will grow will be highly vertical/specialized or very big orgs offering commodity. They both will use value based pricing, but for the first growth will be sales-led and prices will be high, and the second will use product-led growth, and prices will be a race to the bottom of the market.
- Return to aspx/cgi-bin/single php/... files, as more and more services will simply bind together existing third party services.
- Return to server-side default first.
- People will start discovering P2P and start using it.
- Some variant of SQLite that you can run from the browser and update client-side will emerge. This might result in the re-discovery of offline web apps.
- A lot of AI driven content will appear, resulting in too much content and the need for a circle of trust (i.e. what are some good websites). Generic search might become less useful with things like ChatGPT, and something like the old webrings concept will probably emerge, although it might be differently implemented (i.e. a protocol, social media thing, ...)
bonus (as it's related to what I'm doing on a daily basis): something like the Metaverse will start emerging, but it will be web based using WebXR, in splintered/mostly gated communities. AR on mobile phones will become more accessible to consumers via phone, and a common way to interact with these things will emerge this year.
> People will start discovering P2P and start using it.
I recently tried to get into P2P for web apps, only to discover that P2P is increasingly a lie.
100% of my users are unable to connect to 100% of my users, due to CGNATs and IPv4/IPv6 mismatches.
I know the percentages are better for some projects.
Maybe once we are fully IPv6 and carriers drop their NATs.
Won't happen next year, if ever, especially the latter.
We are working in decentralised hole punching with QUIC to help enable P2P.
Does it work when both users have symmetric NAT with randomized ports?
WebRTC provides an abundance of hole punching, but as far as I've learned, nothing gets through symmetric NAT with randomized ports.
Except for relays (TURN), which isn't P2P.
1 reply →
> Some variant of SQLite that you can run from the browser and update client-side will emerge
Has already been tried:
https://en.wikipedia.org/wiki/Web_SQL_Database
It failed because of being a browser built in tech with insufficient support and no route to standardisation. Newer wasm based approaches have had a good year in 2022 with absurd-sql, sql.js and upstream wasm/vfs work so I think it's actually pretty close to just having a drop in library with good browser support