Medium/dynamite

goog.labs.Promise instead of Kew

Closed this issue · 5 comments

Can be Kew replaced with a new goog.labs.Promise? Since I am using Closure code in Node.js, this would be fine improvement. Can I help somehow?

xiao commented

We will need to evaluate goog.labs.Promises, because the promise library's efficiency is critical for us, which is also why we implemented Kew. I can't promise this can happen soon.

Feel free to fork and try though. :-)

Can I kindly ask what sorcery you do with DynamoDB that average Promise implementation is not good enough?

dpup commented

It's not such an issue in Dynamite, more
https://github.com/Obvious/shepherd/

-- Dan

On Thu, Jan 9, 2014 at 10:27 PM, Daniel Steigerwald <
notifications@github.com> wrote:

Can I kindly ask what sorcery you do with DynamoDB that average Promise
implementation is not good enough?


Reply to this email directly or view it on GitHubhttps://github.com//issues/21#issuecomment-32005434
.

can you elaborate on why this would be an improvement? it seems like there
are only downsides.

  1. the goog.labs namespace is not really intended for production code
  2. dynamite will only run on a nodejs server, so a lot of the
    multi-environment abstraction that goog.labs.promise proivdes would just be
    dead weight
  3. the whole reason we have a promises/A+ spec is so that promise libraries
    can interoperate: whether or not dynamite uses kew shouldn't affect what
    promise libraries you can use.
  4. if we were going to change promise libraries, we would probably want to
    move closer to es6 native promises (
    https://github.com/domenic/promises-unwrapping)

On Thu, Jan 9, 2014 at 8:14 PM, Daniel Steigerwald <notifications@github.com

wrote:

Can be Kew replaced with a new goog.labs.Promise? Since I am using Closure
code in Node.js, this would be fine improvement. Can I help somehow?


Reply to this email directly or view it on GitHubhttps://github.com//issues/21
.

  1. Not yet ofc but will be.
  2. Dead weight? Why I should care on Node.js?
  3. It really can interoperate?

If Kew really can interoperate with http://promisesaplus.com then my
question was pointless.

On Fri, Jan 10, 2014 at 6:01 PM, Nick Santos notifications@github.comwrote:

can you elaborate on why this would be an improvement? it seems like there
are only downsides.

  1. the goog.labs namespace is not really intended for production code
  2. dynamite will only run on a nodejs server, so a lot of the
    multi-environment abstraction that goog.labs.promise proivdes would just
    be
    dead weight
  3. the whole reason we have a promises/A+ spec is so that promise
    libraries
    can interoperate: whether or not dynamite uses kew shouldn't affect what
    promise libraries you can use.
  4. if we were going to change promise libraries, we would probably want to
    move closer to es6 native promises (
    https://github.com/domenic/promises-unwrapping)

On Thu, Jan 9, 2014 at 8:14 PM, Daniel Steigerwald <
notifications@github.com

wrote:

Can be Kew replaced with a new goog.labs.Promise? Since I am using
Closure
code in Node.js, this would be fine improvement. Can I help somehow?


Reply to this email directly or view it on GitHub<
https://github.com/Obvious/dynamite/issues/21>
.


Reply to this email directly or view it on GitHubhttps://github.com//issues/21#issuecomment-32045178
.