Real world 2.0 (AKA port a unix tool to the web)

[ninjabear]

I can't remember where I read this, but something along the lines of all big websites are just an evolution of something we already do all the time.

For example,

• skype is telephone 2.0
• twitter is basically SMS 2.0
• email is snail mail 2.0

Theres a long list of linux utilities that do "stuff" like this,

wall : its basically twitter find : really old web search locate : google comes along rsync : cloud drive sed : news aggregators/parsers? traceroute : google maps? maybe a stretch finger : facebook

So the idea is to create a web based version of something a nix utility already does or did.There must be loads of these which don't have a web implementation in a literal sense, with more having a non literal successor.

[MatthewJWalls]

I think google maps probably replaces, you know, maps, rather than traceroute.

[MatthewJWalls]

I really like the idea of data structures as a service.

Memcache is a hashmap as a service. RabbitMQ/ZeroMQ are queues as a service. Are there any data structures that haven't been packages as a service with an API that would be useful?

[ninjabear]

I'll reel some off;

Semaphores as a service LRU cache Linked list

Having started that the problem with these is we'd have to write it in something like C (or rust)

[ninjabear]

Frequency tracker (as in a dictionary word->int) which is surprisingly annoying to generate and manipulate imho

[ninjabear]

added proposed-veto because it will be mainly soul searching for things people haven't already done (there are existing fully formed ideas here already)