airavata-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Emre Brookes <e...@biochem.uthscsa.edu>
Subject Re: [DISCUSS] Accept GenApp into Apache Airavata
Date Thu, 26 Mar 2015 15:11:11 GMT
Chris, Suresh,

I am not quite settled yet.  I was leaning towards direct Airavata 
integration,
but Chris's points have me thinking that incubator status might be
a good place to get everything resolved beforehand.

I am going to have to take some time to think about this a bit more.

Thanks,
Emre


Suresh Marru wrote:
> Hi Chris,
>
> I appreciate you taking time in such a detail, very insightful. I am ok either way and
will defer to Emre.
>
> Suresh
>
>> On Mar 26, 2015, at 2:39 AM, Mattmann, Chris A (3980) <chris.a.mattmann@jpl.nasa.gov>
wrote:
>>
>> Hi Suresh,
>>
>> I would honestly advise against this for a multitude of
>> reasons. Note we recently went through a similar thought
>> on OODT/Wings with the prevailing sentiment from me and
>> a few others being suggesting Wings either goes through
>> Incubation at the ASF or remain at Github until there is
>> an actual connection (direct) between Wings and OODT such
>> that they are complimentary products and “bound” together
>> (aka you can’t release one without the other).
>>
>> Here are a few reasons:
>>
>> 1. Binding the products together on a committee requires
>> that the committee (PMC) have merit in each other’s products.
>> I don’t see that starting off at least.
>>
>> 2. Having mutual products together also potentially binds
>> their release cycle - sure we can release as a committee
>> “independent products”, but there is then scrutiny and
>> sometimes “forced” instead of “natural” binding glue
>> developed between the software products if it wasn’t there
>> already.
>>
>> 3. IP clearance; brand; trademarks etc are things that
>> the PMC can do, but that things like the Incubator is set
>> up to help (or even direct to TLP options that are now
>> available [see Zest]).
>>
>> There are many more reasons that “umbrella” projects didn’t
>> work out at the ASF and are generally discouraged. I wouldn’t
>> recommend turning Airavata into one.
>>
>> Instead, I would recommend the following:
>>
>> R1. GenApp through the Incubator
>> R2. Mentors include the Airavata community PMC members that
>> are ASF or IPMC members (Suresh, Marlon, etc.)
>> R3. GenApp consider a few Airavata PMC/committers in its
>> initial PPMC makeup to develop synergy between the groups,
>> and to see if there are answers to 1-3 and more to be worked
>> out during Incubation.
>>
>> If the result of R1-R3 yields a desire to “graduate into
>> Airavata” the answers to the questions 1-3 above will have
>> been worked out and it will be a much easier answer then.
>>
>> Cheers,
>> Chris
>>
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> Chris Mattmann, Ph.D.
>> Chief Architect
>> Instrument Software and Science Data Systems Section (398)
>> NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
>> Office: 168-519, Mailstop: 168-527
>> Email: chris.a.mattmann@nasa.gov
>> WWW:  http://sunset.usc.edu/~mattmann/
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> Adjunct Associate Professor, Computer Science Department
>> University of Southern California, Los Angeles, CA 90089 USA
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>
>>
>>
>>
>>
>>
>> -----Original Message-----
>> From: <Marru>, Suresh <smarru@iu.edu>
>> Reply-To: "dev@airavata.apache.org" <dev@airavata.apache.org>
>> Date: Wednesday, March 25, 2015 at 11:14 AM
>> To: Emre Brookes <emre@biochem.uthscsa.edu>
>> Cc: Airavata Dev <dev@airavata.apache.org>
>> Subject: [DISCUSS] Accept GenApp into Apache Airavata
>>
>>> Hi All,
>>>
>>> Since there is not much discussion on this thread, I will make it more
>>> explicit on one option (which is Emre’s preference) and solicit further
>>> discussion.
>>>
>>> How about Emre can bring in GenApp into Airavata, integrate it further
>>> into Airavata and bootstrap the community here. Some of Airavata
>>> Community might get interested in GenApp. Once there is a quorum and feel
>>> if GenApp goals are diverging from Airavata (or if the community becomes
>>> distinct), then GenApp could go through incubator and get on a path
>>> towards a Apache Top Level Project (TLP).
>>>
>>> Either way, Emre has been guiding GSoC students, on Airavata Integration,
>>> I suggest they join the dev list and write their proposals on this list.
>>>
>>> Thoughts, concerns?
>>>
>>> Cheers,
>>> Suresh
>>>
>>> P.S Chris, I am taking liberty to send your private list reply to this,
>>> please excuse.
>>>
>>> On Mar 22, 2015, at 3:45 PM, Mattmann, Chris A (3980)
>>> <chris.a.mattmann@jpl.nasa.gov> wrote:
>>>
>>> Hi everyone! Welcome Emre and if there are any questions about
>>> Incubation I’d be happy to help answer them. Suresh great work
>>> bringing the conversation to the lists.
>>>
>>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>> Chris Mattmann, Ph.D.
>>> Chief Architect
>>> Instrument Software and Science Data Systems Section (398)
>>> NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
>>> Office: 168-519, Mailstop: 168-527
>>> Email: chris.a.mattmann@nasa.gov
>>> WWW:  http://sunset.usc.edu/~mattmann/
>>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>> Adjunct Associate Professor, Computer Science Department
>>> University of Southern California, Los Angeles, CA 90089 USA
>>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>>
>>>
>>>> On Mar 19, 2015, at 11:46 AM, Marru, Suresh <smarru@iu.edu> wrote:
>>>>
>>>> Hi Emre,
>>>>
>>>> I firstly applaud the CCP-SAS team in having GenApp as an Open Source
>>>> project with an interest to grow a community. Normally, Apache has not
>>>> been encouraging “umbrella” projects where lot of sub-projects exist.
So
>>>> the suggest mechanism, will be to go through the Incubator Process and
>>>> work towards a GenApp top level project. But, here is a debatable
>>>> situation and there are multiple ways we can view this.
>>>>
>>>> * GenApp can be considered a downstream project of Airavata in which
>>>> case it can be argued for a stand alone project.
>>>> * GenApp is consuming Airavata API’s and helping users build gateways
>>>> based on their applications (the lab generated code you refer below). In
>>>> this perspective, GenApp could rightly belong into Airavata itself and
>>>> have its own product releases. This is very similar to how Airavata
>>>> currently releases XBaya and very soon a PHP Gateway.
>>>>
>>>> Ofcourse, there is a pragmatic way to approach this and have you get
>>>> started within Airavata and if we realize its significant to stand on
>>>> its own (and you might have generated some developer/community interest
>>>> by then), this can then spin off into an incubator project and
>>>> eventually into a TLP.
>>>>
>>>> I am cc’ing Airavata Community for input. I would have suggested apache
>>>> incubator general list, but that may be early to start with. We should
>>>> have expert advice here (there are a lot of apache members on airavata
>>>> pmc and we have Chris Mattmann who is member of the current board).
>>>>
>>>> Cheers,
>>>> Suresh
>>>>
>>>>> On Mar 19, 2015, at 9:23 AM, Emre Brookes <emre@biochem.uthscsa.edu>
>>>>> wrote:
>>>>>
>>>>> Dear Suresh Marru,
>>>>>
>>>>> I am writing this letter pursuant to your consideration of the
>>>>> inclusion of
>>>>> "GenApp" as a sub-project of Apache Airavata.
>>>>>
>>>>> The GenApp framework is a new open framework generating code on a set
>>>>> of scientific modules that is easily extensible to new environments.
>>>>> For example, one can take a set of module definitions and generate a
>>>>> complete HTML5/PHP science gateway and a Qt4/GUI application on the
>>>>> identical set of modules.  If a new technology comes along, the
>>>>> framework can easily be extended to new "target languages" by
>>>>> including appropriate code fragments without effecting the underlying
>>>>> modules. One motivation for the development was based upon observation
>>>>> of scientific lab generated code, which frequently is underfunded and
>>>>> developed by overburdened researchers.  Many times useful code and
>>>>> routines are lost with the retirement or redirected interest of the
>>>>> scientists.  One goal for this framework is to insure good scientific
>>>>> software can be preserved in an ever evolving software landscape
>>>>> without the expense of a full time CS staff.   A GSoC 2014 project
>>>>> integrated GenApp with Apache Airavata in the HTML5/PHP, Qt3/GUI
>>>>> and Qt4/GUI "target languages".
>>>>>
>>>>> The GenApp framework was developed for and is currently in use by
>>>>> CCP-SAS http://ccp-sas.org to make accessible codes involved
>>>>> in the the scientific analysis of small angle scattering experiments.
>>>>>
>>>>> After reviewing the philosophy of Apache
>>>>> http://www.apache.org/foundation/how-it-works.html
>>>>> we feel in concert and believe the GenApp code and Apache will benefit
>>>>> by its inclusion.  We understand that there will be additional work
>>>>> and overhead involved in managing such a project, but that their are
>>>>> many benefits.  As one of our projects goals is to insure the longevity
>>>>> of scientific lab developed software, building a community is
>>>>> essential.
>>>>> Apache membership will provide an established organization for users
>>>>> and developers to collaborate. As we are in a rapid growth phase,
>>>>> with multiple scientific labs interested in bringing their codes to the
>>>>> framework, and a formal roll out to the scientific community of
>>>>> a GenApp generated science gateway at the end of May, we feel the
>>>>> time is now to apply the Apache model.
>>>>>
>>>>> Please let me know if you have any questions or wish further details.
>>>>>
>>>>> Best regards,
>>>>> Emre
>>>>>
>>>>>
>>>>> Emre Brookes
>>>>> Assistant Professor
>>>>> Department of Biochemistry,
>>>>> U. Texas Health Science Center @ San Antonio
>>>>> emre@biochem.uthscsa.edu
>>>>>


Mime
View raw message