spot-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mark Schoeni <markscho...@gmail.com>
Subject Meeting Minutes for Project Wide Meeting on 2018-07-27
Date Sat, 28 Jul 2018 18:23:35 GMT
Hello Everyone,
Thank you for those of you who joined the meeting yesterday. I apologize
for the bumpy start, next time should be smoother now that the limits of
Hangouts are clear. Below you will find notes taken from the meeting. I
will find a permanent location for these meeting minutes (Any suggestions
would help). By the way we had a pretty good turn-out with 14 people.

Thanks again everyone

- Mark Schoeni

--------------------------------------------------------











*Highlights - Bi-weekly (every 2 weeks) Project wide meetings will be
scheduled. - Discussion over how we can build a better community.- Increase
visibility of Apache Spot users.- New Branch to develop a better Front-End
user experience.- Easier Installation procedure and documentation.-
Straight forward performance tests for grading machine learning
algorithms.- All new Features should be Jira tickets before development
start to allow for feedback.- Move all discussions to the dev list
(dev@spot.incubator.apache.org <dev@spot.incubator.apache.org>).Actions -
List all pull requests and order them by priority- Create a working roadmap
in Confluence- Bi-weekly (2 weeks) hangout meetings on Friday from 3:30 -
4:30 (ET)- Developers should create Jira issues for features they would
like to addNotesTopic - How to build a better community - Increase
visibility by requesting platform users to release statements on how
they’re using the platform.- Branch to build a better Front-End user
experience- Need to develop an open-source experience that does not rely on
external visualization platforms- However, should still be able to plug in
external visualization tools such as Arcadia Data.- Will plan a sprint
around this development.- Would like to present the updates around the time
of Strata (Sept. 11)- Simplify the ODM table and make it more user
friendly.- Create performance benchmarks for new algorithms and machine
learning models- Current models have a lot of false positives.- Need to be
able to compare the performance of models against the same benchmark
dataset.Topic - Roadmap - List all of the Pull Requests out there and set a
priority- Committers need to approve the pull requests. Though some can be
approved by one person, it would be better to have more discussion over the
PRs (over +1s) before merging.- Would like more feedback on PRs before
merging.- Features cannot be merged without a matching ticket in Jira- A
ticket should be generated before the feature is created to allow for
feedback from the community- Short-term to bring the community back to life
we should focus on features over PRs*

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message