Hacktivity around the Collaboratory


#1

The Collaboratory team will be present during the whole summit hackathon to help you achieve any hack that involve integration with the Collaboratory or its underlying web services.

If you plan to try to integrate one of your tool, you can get in touch here. We will provide extensive help for starting from scratch using Python or NodeJS and AngularJS. We can provide guidance with most other web technologies as well.

Example Hacktivities:

  • integrates your existing web application or a part of it within the Collaboratory
  • bootstrap a new app
    • to create a data visualization of your existing dataset
    • that run experiment in our datacenter and publish the result in the Collaboratory storage
    • that let HBP users access some of your service
  • authenticate user to your existing application using the HBP OpenID Connect service
  • learn how to create a collab to describe your Platform, SP, or favourite tool

If you have any potential idea, please introduce it by answering to this topic.


#2

Domenico Guarino and I would be interested in working on a couple of things related to the Neuromorphic Computing Platform:

(i) replacing the grouping of jobs by Projects with grouping of jobs by Collabs, i.e. to have an app for submitting jobs to the NM platform which could be embedded into any Collab, and which would show only jobs belonging to that Collab.
(ii) adding interaction with the Provenance Service.
(iii) AngularJS code review - we’re quite new to working with Angular, so a quick review of our code by one of the Collaboratory team would be helpful.


#3

That sound really great! We can work on that with you, for sure.

On the topic, do you know how far we are from being able to run job with HBP accounts? I mean without too much manual accreditation.


#4

You can run jobs on the Neuromorphic Platform now with HBP accounts (and since a couple of months). No manual accreditation needed (in future there will be a quota system, but not yet).


#5

Ah, you will have to show me how to add tasks and run job there during the summit :sunny:


#6

If I have some time during the hackathon, I would like to play a bit with Operational Transform library (eg. share.js) in order to create a live editing experience for scientific tasks.

The idea is to:

  • read a data file from the collab storage,
  • live edit it,
  • and publish a new version of it.

Optionally have a Provenance proof of the linkage belonging to all live editor contributors.
I’m looking for a good use case to demonstrate it. I would like to find some data where there is a clearly added value by allowing simultaneous edition. Any idea?