Hi Andy
How uniform are apache projects in terms of documentation structure, release,
roadmap, etc?Is an apache goal one where for example, one browse to each project
http://spark.apache.org/
http://kafka.apache.org/http://storm.apache.org/
http://hbase.apache.org/
http://bigtop.apache.org/
...
and expect to find certain links in common? For example - pointers to ASF,
download links, ...
This doesn't seem to be the case. Each project has its own style of html and
way of introducing the interested user to the project.
Given each project has it's own style in introducing what the project is
about,I imagine ASF members use both github and JIRA to ascertain a project's
health.
That is - it's fairly easy to go to the following github
linkshttps://github.com/apache/sparkhttps://github.com/apache/kafka
https://github.com/apache/storm
https://github.com/apache/hbase
https://github.com/apache/bigtop
and get a sense of forks, favorites, stars, contributors.In the same way -
going to JIRA - one could get a sense of issues burn down, release schedule
etc.There's also probably a way to judge a given projects usage within other
ASF projects if you were to cull
maven dependency information.
Both github and JIRA have good API's. Do ASF members receive summary reports
showing trending of contributors, issues, releases, etc (if not it seems like
they should)?
For Gearpump, we certainly want to progress along accepted criteria ASF uses to
judge health.I think we also want to focus (based on the teams comments) on
making areas of contribution crystal clear,with easy ways to onboard a
developer. This will also imply very clear roadmaps and establishing a
predictable release cadence. Judging from some of the more popular projects, it
looks like the committers take great care in providing a 'getting started' page.
http://beam.apache.org/getting_started/
http://spark.apache.org/docs/latest/quick-start.html
https://ci.apache.org/projects/flink/flink-docs-release-1.0/quickstart/setup_quickstart.html
http://hbase.apache.org/book.html#quickstart
On Tuesday, April 5, 2016 2:10 PM, Andrew Purtell <[email protected]>
wrote:
This is great.
Note the list of three things can evolve over time as the podling makes
progress.
In my opinion, the most important thing an Apache project can do is release
software. While it is true there is a mantra here which goes "community over
code", communities need code around which to form. Therefore I submit to you
that the most important task the Gearpump podling can currently accomplish is
an Apache branded release of the Gearpump code, conforming to our release and
IP policies.
Also, for what it's worth, consider at the Foundation level there are ~280
projects and ~40 podlings, each with their own specific set of technical issues
and goals, of which the Board and Membership cannot possibly understand all in
detail. Try to place yourself in such an oversight role and think about how you
would judge the health of any given project. What measures would you consider?
Community building? Policy conformance? Releasing? Governance?
Anyway I await the outcome of your discussion oh _the_ list of three things for
tomorrow's report.
On Tue, Apr 5, 2016 at 9:03 AM, Kam Kasravi <[email protected]> wrote:
Three most important issues to address in the move towards graduation:
1. Develop a tag line that describes gearpump. For example others
- Apache Flink is an open source platform for distributed stream and batch
data processing
- Apache Spark™ is a fast and general engine for large-scale data processing
- Apache Storm is a free and open source distributed realtime computation
system.
- Apache Beam is an open source, unified model and set of language-specific
SDKs for defining data processing workflows that may then be executed on top of
a set of supported runners
2. Focus on use cases that leverage akka strengths (location transparency,
remoting, code provisioning, dynamic deployment)
3. Target areas within apache ecosystem where gearpump can provide
significant value.
- Many real time data processing engines are exploring reusable pipelines
including spark (ml pipelines), akka-streams (blueprints)
- Other real time data processing engines are targeting a common data model
DSL which allows different execution engines (apache beam and 'runners')
Show original message
On Tuesday, April 5, 2016 8:51 AM, Kam Kasravi
<[email protected]> wrote:
Three most important issues to address in the move towards graduation:
1. Develop a tag line that describes gearpump. For example others
- Apache Flink is an open source platform for distributed stream and batch
data processing
- Apache Spark™ is a fast and general engine for large-scale data processing
- Apache Storm is a free and open source distributed realtime computation
system.
- Apache Beam is an open source, unified model and set of language-specific
SDKs for defining data processing workflows that may then be executed on top of
a set of supported runners
2. Focus on use cases that leverage akka strengths (location transparency,
remoting, code provisioning, dynamic deployment)
3. Target areas within apache ecosystem where gearpump can provide
significant value.
On Tuesday, April 5, 2016 1:06 AM, Weihua Jiang <[email protected]> wrote:
I agree with Sean and Vincent.
To graduate, we need to create a healthy community. To me, a healthy community
has:
1. Enough contributors who actively contribute to this project.
2. Enough users who actively using this project in their environment. And they
are willing to interact with community for issues and features. Their requests
can be handled in a timely way.
So, to me, pre-conditions to graduation shall include:
1. Fully follow Apache project best practices. That is, doc style, process,
releases etc. And it is better to have 2 or more releases before graduation to
get everyone familiar with the new process.
2. Code quality. It is better if we can meet certain code quality metrics
before graduation and integrated in Apache infrastructure. E.g. Code coverage,
auto-checkin test, etc.
3. Contributor friendly as Sean mentioned.
4. User friendly as Vincent mentioned.
在 16/4/5 下午3:08，“Vincent Wang”<[email protected]> 写入:
>Hi Xiang,
>
>I think your third point is kind of related to the first one. We should
>also reduce the challenges for the users who want to try Gearpump, better
>documentations, easy-to-use API and more external connectors.
>
>Thanks,
>Huafeng
>
>Best regards
>Vincent Wang
>
>2016-04-05 14:14 GMT+08:00 Sean Zhong <[email protected]>:
>
>> Here is my thoughts about pre-condition for graduation:
>>
>> 1. We should setup the environment to reduce the challenges for new
>> contributors to make contribution. Currently, it is not very easy for new
>> developer to make contribution, we need a environment like this:
>> a. Document clearly on contribution process so that new developer know
>> exactly how to submit an issue, a PR, and ...
>> b. Better and more documents to walk new developers through to help them
>> better understand Gearpump without spending too many time on source code.
>> c. Clearly identity the scenarios that Gearpump works best, and why
>> Gearpump is good at this.
>> d. Have a clear document on the milestone, and the timeline, so that the
>> community can form a consensus on what is coming next.
>> e. Gradually form a convention of SLA for issues and questions. The
>> committers need to take ownership so that all issues and questions are
>> taken care in a timely manner.
>> f. Move project discussions offline to online, and publish them on the
>> mail list.
>>
>> The goal is to serve new contributors best so that they feel it is easy to
>> get started, and be comfortable in the community.
>>
>> 2. More examination and adoption on various use cases by different
>> companies. We need more examinations in real production clusters of huge
>> size, and having complex network environment.
>>
>> 3. A busy and noisy user community and developer community.
>>
>>
>>
>>
>> On Tue, Apr 5, 2016 at 11:01 AM, Sean Zhong <[email protected]> wrote:
>>
>> > Hi Andrew,
>> >
>> > Got it! Let's discuss it here.
>> >
>> > On Sun, Apr 3, 2016 at 1:13 AM, Andrew Purtell <[email protected]
>> >
>> > wrote:
>> >
>> >> Sure.
>> >>
>> >> If by 'collect thoughts' you mean yourself spending time to think,
>> great.
>> >>
>> >> If by 'collect thoughts' you mean to talk with others on the project
>> >> before replying with a summary or conclusion, let me recommend not doing
>> >> that and instead have that discussion here on dev@.
>> >>
>> >> > On Apr 2, 2016, at 5:39 AM, Sean Zhong <[email protected]> wrote:
>> >> >
>> >> > Thanks, Andrew,
>> >> >
>> >> > Let me collect some thoughts before replying you.
>> >> >
>> >> >> On Sat, Apr 2, 2016 at 1:59 AM, Andrew Purtell <[email protected]>
>> >> wrote:
>> >> >>
>> >> >> Greetings ...,
>> >> >>
>> >> >> (What should be the nickname for your community? Gearpumpers?
>> >> Gearheads?
>> >> >> (smile) )
>> >> >>
>> >> >> It's time to file the first podling status report, due up on
>> >> >> http://wiki.apache.org/incubator/April2016 by April 6, this coming
>> >> >> Wednesday.
>> >> >>
>> >> >> I have started a draft for you. Let me ask you: What do you think are
>> >> the
>> >> >> three most important issues for you to address before graduation? I
>> >> left
>> >> >> that blank. If you would like to see additional changes in the
>> report,
>> >> >> please discuss.
>> >> >>
>> >> >> Gearpump
>> >> >>
>> >> >> Gearpump is a reactive real-time streaming engine based on the
>> >> >> micro-service
>> >> >> Actor model.
>> >> >>
>> >> >> Gearpump has been incubating since 2016-03-08.
>> >> >>
>> >> >> Three most important issues to address in the move towards
>> graduation:
>> >> >>
>> >> >> 1.
>> >> >> 2.
>> >> >> 3.
>> >> >>
>> >> >> Any issues that the Incubator PMC (IPMC) or ASF Board wish/need to be
>> >> >> aware of?
>> >> >>
>> >> >> The rights holder of the Gearpump copyright filed a CCLA including a
>> >> >> Schedule B granting the Gearpump codebase to the Foundation. We are
>> >> >> awaiting assistance from Infrastructure on INFRA-11435 to perform the
>> >> >> import.
>> >> >>
>> >> >> How has the community developed since the last report?
>> >> >>
>> >> >> All of the initial committers/PPMC have set up with Apache accounts
>> and
>> >> >> Apache JIRA accounts.
>> >> >>
>> >> >> Discussion among developers has started on
>> >> >> [email protected]
>> >> >>
>> >> >> How has the project developed since the last report?
>> >> >>
>> >> >> The JIRA for the podling is active at
>> >> >> https://issues.apache.org/jira/browse/GEARPUMP and seeing activity.
>> >> >>
>> >> >> Date of last release:
>> >> >>
>> >> >> No release yet.
>> >> >>
>> >> >> When were the last committers or PMC members elected?
>> >> >>
>> >> >> No new committers or PMC members elected yet.
>> >> >>
>> >> >> <<<
>> >> >>
>> >> >>
>> >> >>
>> >> >> --
>> >> >> Best regards,
>> >> >>
>> >> >> - Andy
>> >> >>
>> >> >> Problems worthy of attack prove their worth by hitting back. - Piet
>> >> Hein
>> >> >> (via Tom White)
>> >> >>
>> >>
>> >
>> >
>>
--
Best regards,
- Andy
Problems worthy of attack prove their worth by hitting back. - Piet Hein (via
Tom White)
