airavata-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vivek Bhatia <viv...@gmail.com>
Subject Re: Plan for community buidling
Date Tue, 22 Apr 2014 21:21:13 GMT
Sorry I forgot to provide an example so doing that now. So based on
following comment from Suresh in the email related to the discussion we are
having regarding GFac's role.

So I should write one plugin which works well for a single execution and
has been builtin checkpoint recovery at critical steps and then framework
should help me deal with multiple threads of these executions, logging,
recovery, call-back and so on. There should be a guideline of how the
contract between the framework and extensions.

Theme - GFac 2.0
Epic - PlugIn for GFac
Story - As a User I would like to build checkpoint recovery at critical
steps in the plugin.
Task - blah, blah...

The component for this is GFac and others that we touch.

Hope this helps!

-Vivek




On Tue, Apr 22, 2014 at 2:12 PM, Vivek Bhatia <vivb77@gmail.com> wrote:

> Hi Suresh,
>
> Yes, I agree we should segregate major architectural changes. Ideally, we
> can view these as 2 separate areas. There are others as well but can start
> with these.
>
> 1. Work streams - These are projects/work items that every one works on.
> We need to establish a structure for these. For example following is one
> way to organize our work items.
>
> Themes - This is a high level view of tangible work/product/feature. This
> is sometimes also called a HLF (High Level Feature)
> |_
>    Epic - The themes are generally broken down into one or more Epic. An
> Epic is a block of requirements that have not been broken down on
> rationalized into stories.
>          |_
>             Stories - These are brief statements for product requirements
> or use/business case. These are one level below the Epics in other words
> one or more stories under an Epic
>                      |_
>                         Task - Tasks are at the most granular level and
> are a discreet piece of work. They define some effort or work that can be
> completed to satisfy the task. These roll up into a story.
>
> JIRA provides the capability for us to be able to build a structure like
> this. This will enable us to determine what and how much work we need to do
> to achieve what we are planning to achieve. This is a simplistic view and
> obviously more needs to be done but if we can implement this to begin with
> it will be a good start. For example, later on we can also track these work
> items using labels once the structure is established.
>
> 2. Architectural Components - These are architectural components that we
> work on as part of work items. We can define these in JIRA in the component
> field such as XBaya, GFac, etc. I noticed that we are already doing this.
>
> I can help  with some of this effort but I do not have edit/create
> permissions on JIRA. I tried assigning a JIRA to myself but wasn't able.
> Additionally I wanted to log a JIRA for the common commands but couldn't do
> it. I think I will need permissions to do that.
>
> Thanks,
> -Vivek
>
>
>
> On Tue, Apr 22, 2014 at 1:39 PM, Suresh Marru <smarru@apache.org> wrote:
>
>> Hi Vivek,
>>
>> I think this is a great suggestion. I agree and second everything you say
>> below.
>>
>> Should we also consider segregating major architectural changes,
>> incremental development tasks and bug fixes?
>>
>> Do you have any suggestions for a JIRA Workflow? If you already do not
>> have the right privileges on airavata jira, we can req
>>
>>
>> On Mon, Apr 21, 2014 at 8:57 PM, Vivek Bhatia <vivb77@gmail.com> wrote:
>>
>>> Hi All,
>>>
>>> re: community building and suggestions. I completely agree with the
>>> following indicated in Suresh's email. It is always a good idea to think
>>> through the bigger picture first and break down the work into smaller
>>> chunks and file JIRA's for it. This is a very common industry practice and
>>> will help us in several ways such as provide a high level structure for the
>>> JIRA's, help other contributors understand the bigger picture and pitch in
>>> into the effort, help us evaluate work/milestone for each feature etc.,
>>> additionally could also help identify what our roadmap is so that we can
>>> publish that out to perspective community/users for Airavta. We might be
>>> doing this already and it might be a good idea to take another look at this
>>> to see where we need to put more emphasis on, which is what I think the
>>> objective of this effort/email is...
>>>
>>> 1) The current core developers should spend more time in described
>>> requirements and clearly scoped improvements to JIRA. As developers, we
>>> tend to enjoy writing in java than in english. But I feel, the time we take
>>> of our own coding and writing well defined requirements will boost the
>>> community building.
>>> 2) JIRA tasks - Currently the developers are adding issues on what they
>>> are working. This is undoubtedly helping to track commits to JIRA, but as a
>>> good development practice, we should add as many tasks as possible, and
>>> then when we start to work on an issue, we should assign it to ourselves
>>> and start coding. This way we know the active development areas ahead of
>>> time and community can if possible align.
>>>
>>>
>>> my 2 cents here...
>>> -Vivek
>>>
>>>
>>>
>>> On Fri, Apr 18, 2014 at 8:09 PM, Suresh Marru <smarru@apache.org> wrote:
>>>
>>>> Hi All,
>>>>
>>>> I want to revisit the community building thread from over two years
>>>> ago. Any concrete steps we can take now?
>>>>
>>>> Eran, thanks for sharing some of these concerns in a post-hangout
>>>> discussion today. Can you please share some of your suggestions on this
>>>> thread?
>>>>
>>>> Suresh
>>>>
>>>>
>>>> On Sun, Feb 5, 2012 at 11:39 AM, Suresh Marru <smarru@apache.org>wrote:
>>>>
>>>>> Thanks Ross for initiating this important conversion and for Chri's
>>>>> suggestions on OODT.
>>>>>
>>>>> Good to see some new community requests recently, it will be nice to
>>>>> get some feedback as well. So please speak up, both good and back feedback
>>>>> will be equally recieved. We would like to know how we can help lower
the
>>>>> barrier to use and contribute to Airavata.
>>>>>
>>>>> In addition to what Marlon already mentioned, I can see some
>>>>> improvements we can make.
>>>>>
>>>>> 1) The current core developers should spend more time in described
>>>>> requirements and clearly scoped improvements to JIRA. As developers,
we
>>>>> tend to enjoy writing in java than in english. But I feel, the time we
take
>>>>> of our own coding and writing well defined requirements will boost the
>>>>> community building.
>>>>> 2) JIRA tasks - Currently the developers are adding issues on what
>>>>> they are working. This is undoubtedly helping to track commits to JIRA,
but
>>>>> as a good development practice, we should add as many tasks as possible,
>>>>> and then when we start to work on an issue, we should assign it to
>>>>> ourselves and start coding. This way we know the active development areas
>>>>> ahead of time and community can if possible align.
>>>>> 3) Add all the test cases to  be improved to JIRA, yes one more JIRA
>>>>> suggestion, but I feel this is important.
>>>>> 4) Improve architecture diagrams, data models, schema documentation,
>>>>> E-R diagrams what ever makes community to understand the code better.
>>>>> 5) Improve usability. Invite HCI usability experts to criticize at
>>>>> same time give suggestions to improve.
>>>>> 6) Airavata primarly caters to Scientific use cases, but as we
>>>>> realize, its fully general purpose and useful in many facets of other
>>>>> application areas. We should actively synergize and engage with workflow,
>>>>> messaging system and hadoop related projects.
>>>>> 7) Start developing web interfaces/gadgets to Airavata back end
>>>>> services and actively work with projects like Rave.
>>>>>
>>>>> Couple of brainstorming ideas:
>>>>> * Should we actively participate in Google summer of code? this not
>>>>> only helps us to break down the tasks, it also makes us think the next
6+
>>>>> months of roadmap. If we are lucky, we might get good code contributions
>>>>> too. Ross, Chris, Any directions on how to proceed on this?
>>>>> * Invite Airavata to be used for capstone projects in programming and
>>>>> HCI courses? Answering student questions will improve our FAQ's greatly
and
>>>>> as above we might expand community to both faculty and students.
>>>>> * Reach out to technical writers to seek their help in improving
>>>>> documentation?
>>>>> * How to address Marlon's comment on making the community feel that
>>>>> they need not write code to be part of the project and be pro-actively
>>>>> contribute to its future directions?
>>>>>
>>>>> Any others?
>>>>>
>>>>> Thanks,
>>>>> Suresh
>>>>>
>>>>> On Jan 31, 2012, at 12:13 PM, Mattmann, Chris A (388J) wrote:
>>>>>
>>>>> > Hi Marlon,
>>>>> >
>>>>> > Both of these are great suggestions and yes we can immediately cite
>>>>> a synergy with OODT as well and some
>>>>> > pilot projects. Getting the conversation on list will be great for
>>>>> the other direct contacts, but it's something we
>>>>> > struggled with originally in OODT and something that can be worked
>>>>> through.
>>>>> >
>>>>> > Great suggestions.
>>>>> >
>>>>> > Cheers,
>>>>> > Chris
>>>>> >
>>>>> > On Jan 31, 2012, at 6:18 AM, Marlon Pierce wrote:
>>>>> >
>>>>> >> -----BEGIN PGP SIGNED MESSAGE-----
>>>>> >> Hash: SHA1
>>>>> >>
>>>>> >> We've been recruiting several groups to participate, and I expect
>>>>> an increase in communications on the list from java cyberinfrastructure
>>>>> developers from Iowa State and University of Minnesota.  We also have
met
>>>>> with Chris Mattman and others from Apache OODT, which is doing
>>>>> complementary things.  We have discussed pilot projects with OODT, so
I
>>>>> think this is something we can do immediately to broaden the community.
>>>>> >>
>>>>> >> Two issues I have seen: 1) we tend to get contacted directly
by
>>>>> collaborators instead of through the dev list, so we need to encourage
(or
>>>>> insist) that more traffic goes on airavata-dev; and 2) we have many
>>>>> collaborators who are not java developers but who have valuable
>>>>> requirements, usage scenarios, feedback, complaints, etc that also need
to
>>>>> go on the list. We need to make it clear to the second group that there
are
>>>>> many ways to contribute besides submitting code patches.
>>>>> >>
>>>>> >>
>>>>> >> Marlon
>>>>> >>
>>>>> >>
>>>>> >> On 1/31/12 8:55 AM, Ross Gardler wrote:
>>>>> >>> First off, I've been a little remiss in my duties as a mentor
here.
>>>>> >>> Appologies for that and thanks to Chris for keeping things
moving.
>>>>> I
>>>>> >>> hope to find more time to spend on this project in the near
future.
>>>>> >>>
>>>>> >>> I would like to see the project members discussing how we
can go
>>>>> about
>>>>> >>> building community diversity in the project.
>>>>> >>>
>>>>> >>> What simple actions can we take to raise awareness (over
and above
>>>>> the
>>>>> >>> lower barriers and make releases items in the board report)?
>>>>> >>>
>>>>> >>> I'm particularly interested in hearing from people who are
lurking
>>>>> >>> here but not yet contributing. What is stopping you from
>>>>> de-lurking?
>>>>> >>> How can we help you take those first initial steps?
>>>>> >>>
>>>>> >>> For those active in the project how do we communicate the
value of
>>>>> >>> Airavata to the rest of the world? Are there any often requested
>>>>> items
>>>>> >>> that people can work on as a first step into the project
community?
>>>>> >>>
>>>>> >>> Any other ideas?
>>>>> >>>
>>>>> >>> My goal is for us to come up with 3-5 concrete actions that
we can
>>>>> >>> include in our next board report.
>>>>> >>>
>>>>> >>> Ross
>>>>> >>>
>>>>> >> -----BEGIN PGP SIGNATURE-----
>>>>> >> Version: GnuPG/MacGPG2 v2.0.16 (Darwin)
>>>>> >> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>>>>> >>
>>>>> >> iQEcBAEBAgAGBQJPJ/gqAAoJEEfVXEODPFIDHZAH/i6Zna9sHcis0GLEfTfotrzO
>>>>> >> l2feAQMbG2I6IO/BOxM8lXtVPbjJGE7DhiFuskbjaommDl+v5Y83UP1lPUTkUIZy
>>>>> >> 1qVCSlIY/7R0ey9ogYA4Yq4rOM7vC+udGlXM5c3Hob/uboctT5io573jx7nGBlqw
>>>>> >> V857RAgbbJdXBVecr25FdEh0jU+It7oJGksERBJnH01EJEvQFof9/1GeuGmnJou4
>>>>> >> rd+LZJZNIhjXa1ZL/uR9BP7kPkMpk4dKVW6xq5d1pg2gJzU9/RE75DYY8r+fsRum
>>>>> >> fUc37om165goIqSHjgq5kRfQdIAHliMwyH/cpp5yjd7a68hASkg5evHo2WxX+l8=
>>>>> >> =6TAH
>>>>> >> -----END PGP SIGNATURE-----
>>>>> >
>>>>> >
>>>>> > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>>> > Chris Mattmann, Ph.D.
>>>>> > Senior Computer Scientist
>>>>> > NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
>>>>> > Office: 171-266B, Mailstop: 171-246
>>>>> > Email: chris.a.mattmann@nasa.gov
>>>>> > WWW:   http://sunset.usc.edu/~mattmann/
>>>>> > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>>> > Adjunct Assistant Professor, Computer Science Department
>>>>> > University of Southern California, Los Angeles, CA 90089 USA
>>>>> > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>>> >
>>>>>
>>>>>
>>>>> -----BEGIN PGP SIGNATURE-----
>>>>>
>>>>> iQIcBAEBAgAGBQJPLrC8AAoJEHmz9P1hfdutJHUP/RnvhhB7m3N4p/FrSPh365Cu
>>>>> XRkN+aHF9bs2wEJohx0py/DJBD5Zpy1MiVisa9m0lBesnIJ1ZcEY2ox8lJOoMQQB
>>>>> HN0IMyeo9/miNFWGAqpBdxIDsBSo4GhI76KQQhPt/ui0MVmQBP/FePGkFaqTS8JK
>>>>> sbDR+BYMv/nZcYxJpFfHdPepiETyXqw29RZUF3SWKeeyDyLWiix23qE3KLiCIFlF
>>>>> kIdiWgNUq/5p6WaOkWuLWhh90tuKMbYVgaA02XbvqoI4ovrxWcSDsSxoaYos8T/1
>>>>> OuubfJqRHgUXP1bMkFifYIYjQxMDN8hg0GAsD/wBy/CWxIHX+UBUc4C8+PjWHmnW
>>>>> 8i87bDnVELNNX9gX/GRGkeJQLaW7gUVkj2QX1SVc7SDgfylwWY2SQoNQfqrAjVEg
>>>>> Y32pDGsX42c2MO4GomJlcIMKtuk4FB5vInVGDFezLxdVoVPby6wwoh7BN6+poVAy
>>>>> ICnn0+bbjrQEfrM7yGyQDSjkfCnO2yWqds7pxxkwrWnFtGrUtsHFwM7mzalF99UL
>>>>> u72KBJn2HgIZTMZVTpIm+sZYtWCCxriANw4QqsMOCiFouepM6ez+j+TlTH83OJt3
>>>>> DOc2HGKZWk4zfYqFEw62N3MxWpsYFsXT/ekCgYS0GvjSuVUqn2I6Nyy0NtrQnSPF
>>>>> ZLjMp55uaMt9M05mhVVF
>>>>> =4S2Q
>>>>> -----END PGP SIGNATURE-----
>>>>>
>>>>>
>>>>
>>>
>>
>

Mime
View raw message