Skip to content

Instantly share code, notes, and snippets.

@tchak
Last active August 29, 2015 14:05
Show Gist options
  • Star 1 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save tchak/33733f152afb4029ac0d to your computer and use it in GitHub Desktop.
Save tchak/33733f152afb4029ac0d to your computer and use it in GitHub Desktop.
Distributed Web Storage

Some random follow up thoughts on todays discussion with @jo_liss, @terzicigor and @dgeb

After more thoughts it seems to me that on the backend it’s almost unavoidable to have commits built in to the database, or at last to have some kind of a persistence layer on top of the database. Basically if someone or something changes the database and do not update "commits table", the system will become inconsistent. So are we trying to invent a new database?

On most backend apis there is a serialization/presentation layer on top of the database, which means api objects != database objects. It means what we were describing today, is not a drop in replacement for REST apis in the form they exists today. Firebase is a “dumb store”, this is why they can be smart about synchronization.

We had only little talk about multiuser scenarios. Or even anonymous users. Seems like a lot to think about here. There should be a way to subscribe to some subsets of data with the right authorizations. What happens if authorizations changes over time?

Given that, I would like to hear your thoughts on who we are building this for? The way I understood the scope of the project is “a general purpose distributed storage for the web”. So lets assume we can build it, we should promote it to "general purpose web developers” right? Seems like more than technical challenge, it is a “change the way people build web sites” kind of challenge.

Also, just like @jo_liss was asking yesterday “what can we do in order to make software developer profession popular?”, we should ask ourselves “what make’s a framework/technology popular?”.

If the answer is "the need to solve a particular problem", it seems there is a lot of work to do to convince web developers that they actually have a problem. Lots of people in the industry still not sure about SPA. And I can not blame them. Offline is seen by most like some kind of gadget. DHH official position is - in a few years there will be no offline. We are going to have 4G all over the world. Seems unlikely, unless we ignore some large portions of the world...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment