Live community hangout #4 - Aug 31st, 10AM PST

[okdistribute]

We will have our fourth community call next Monday using Hangouts on Air! The entire community is encouraged to join and listen in, ask questions, and see where we're going next with dat.

Theme: DAT 1.0 ROADMAP! Where: Google Hangouts on Air Date: Monday, August 31st. Time: 1pm EST (17:00 GMT)

Topics

• containers in 1.0 We went into the German woods for #CCCamp2015, and came back determined for a dat 1.0. We'll talk packaging versioned data dependencies with versioned containers. https://github.com/maxogden/dat/issues/388
• changes format how does dataprotocols.org/SLEEP compare to what we need now for interfacing change feeds with external services and databases?
• dathub progress during the Europe trip, karissa found that dathub was greatly desired. Some thoughts on that.
• schemas how will we interface with dataprotocols.json files? can we make it easy to specify a schema with a dataset?

We'll be using Hangouts On Air

Youtube link (to watch) http://youtu.be/SaWMON7MB3A Gitter link (chat/ask questions): gitter web chatroom

Note that only Dat team members and others on the agenda will be invited to join the Hangout so they can broadcast voice and video. Hangouts only supports about 10 people max, but the youtube stream can support an unlimited amount of viewers.

[karthik]

Naive question. I recall dathub being tossed around as an idea in the early days. Is that now a serious prospect? i.e. a hosted dat backend?

[sallespro]

hello, first question, have you considered using docker and alpine to make images minimalistic ? https://github.com/gliderlabs/docker-alpine

[sallespro]

second question : why gasket instead of gulp ?

[sallespro]

wouldn't gulp bridge dat to a larger audience and enable more "de facto" standard transformations already in place, instead of creating a JSON only pipeline control ? third question : why not managing datasets using an environment like adventure-time, cloud based ?

[sallespro]

@substack made a quite interesting use of P2P in https://github.com/substack/braziljs-2015-talk/tree/master/example/torrent-log

changes feed injected in a torrent payload as a hash

[sallespro]

it seems to me schemas should be bound to the datasets, as meta info

[sallespro]

it is worth to get some inspiration from https://github.com/graphql/graphql-js to abstract schemas

[sallespro]

dear all, thanks for your time, ideas and consideration. i wish DAT could be targetting "not so scientific" use cases and be more driven to managing common used datasets like contacts, lists and also assume some more programatic use of workflow building.

i'll start some prototyping in this direction.

[max-mapper]

@sallespro I agree getting the general node community to use data pipelines etc would be really great for our project, I'm curious to hear if you have ideas on gulp integration and any other use cases you think would be good in the future :)