You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@systemml.apache.org by Madison Myers <ma...@gmail.com> on 2016/10/06 17:33:01 UTC
Getting new users to adopt SystemML
Hi all,
Thought it may be a good idea to implement some measures on getting new
users to adopt SystemML. At the moment, I feel there is a big learning
curve and that we could make it easier through clearer tutorials and
examples, a better user experience on the website (continuing the process
of making it better), more articles that explain SystemML at a high level
and how to use it/why it's important. I also think we should get more
questions and answers on Stack Overflow so that overcoming issues is a bit
easier for new users. Longer-term I'd also love to see a course taught
online such as on Coursera as well as in university classrooms so that
people in the field adopt SystemML more early-on.
Also Deron and Mike had some interesting suggestions about visibility. They
suggested having projects using SystemML on personal Github accounts as
well as mentioning SystemML as a skill on LinkedIn.
Would love to know if others agree and/or have suggestions.
Thanks!
--
*Madison J. Myers*
*UC Berkeley, Master of Information & Data Science '17*
*King's College London, MA Political Science '14*
*New York University, BA Political Science '12*
-
LinkedIn <http://linkedin.com/in/madisonjmyers>
Re: Getting new users to adopt SystemML
Posted by du...@gmail.com.
+1 this is all great!
Yeah let's definitely get some JIRA issues and PRs started to incorporate new updates.
Also, I'd like to suggest that we construct some beginner examples and have everyone on here (& any others!) try them out. Anything that is difficult or confusing should be documented and analyzed for a possible improvement to the project.
Also, as for having projects and use-cases in personal GitHub accounts, one thing that's been done is to start in a personal account and then additionally propose adding it to the main project once it is relatively stable (perhaps as an example case). The nice thing about this is having more exposure, especially for these types of (side) projects that use SystemML in a neat way.
--
Mike Dusenberry
GitHub: github.com/dusenberrymw
LinkedIn: linkedin.com/in/mikedusenberry
Sent from my iPhone.
> On Oct 18, 2016, at 3:42 PM, Jeremy Anderson <je...@objectadjective.com> wrote:
>
> +1 all
>
> ...........................
>
> Jeremy Anderson
> https://twitter.com/ObjectAdjective
> http://www.linkedin.com/in/objectadjective
>
> On 18 October 2016 at 11:25, Renee Mascarinas <re...@gmail.com>
> wrote:
>
>> +1 Thanks Madison!
>>
>> I agree with everything. As a designer, it's been really hard to
>> understand the value of SystemML without case studies. There's so much
>> claims out there, but without real use cases, it's difficult to understand
>> and promote when there aren't tangible proof from existing users.
>>
>> For visibility, I think it's a great idea to add SystemML as a skill on
>> LinkedIn and posting questions and answers on Stack Overflow. From our
>> research findings, the latter is a must for troubleshooting for data
>> scientist, second to Google.
>>
>> Renee
>>
>> On Thu, Oct 6, 2016 at 7:17 PM, Jeremy Anderson <
>> jeremy@objectadjective.com>
>> wrote:
>>
>>> +1 Madison
>>>
>>> I started started doing some user research with a few other designers.
>>> We're looking for patterns and pain points around adoption and use of
>>> SystemML. We'd love to share our findings so far to continue this
>>> discussion and get thoughts and feedback from everyone. I think clear
>>> tutorials and site architecture will help address some of the concerns
>>> we're hearing and could improve adoption. I love the idea of generating
>>> more articles and content on Stack Overflow. All of this would be really
>>> helpful in increasing awareness and building community.
>>>
>>> We're synthesizing our research so far and we'll share it here. Thanks
>> for
>>> starting this thread!
>>>
>>> Jeremy
>>>
>>> ...........................
>>>
>>> Jeremy Anderson
>>> https://twitter.com/ObjectAdjective
>>> http://www.linkedin.com/in/objectadjective
>>>
>>>
>>> On 6 October 2016 at 10:33, Madison Myers <ma...@gmail.com>
>> wrote:
>>>
>>>> Hi all,
>>>>
>>>> Thought it may be a good idea to implement some measures on getting new
>>>> users to adopt SystemML. At the moment, I feel there is a big learning
>>>> curve and that we could make it easier through clearer tutorials and
>>>> examples, a better user experience on the website (continuing the
>> process
>>>> of making it better), more articles that explain SystemML at a high
>> level
>>>> and how to use it/why it's important. I also think we should get more
>>>> questions and answers on Stack Overflow so that overcoming issues is a
>>> bit
>>>> easier for new users. Longer-term I'd also love to see a course taught
>>>> online such as on Coursera as well as in university classrooms so that
>>>> people in the field adopt SystemML more early-on.
>>>>
>>>> Also Deron and Mike had some interesting suggestions about visibility.
>>> They
>>>> suggested having projects using SystemML on personal Github accounts as
>>>> well as mentioning SystemML as a skill on LinkedIn.
>>>>
>>>> Would love to know if others agree and/or have suggestions.
>>>>
>>>> Thanks!
>>>>
>>>> --
>>>> *Madison J. Myers*
>>>> *UC Berkeley, Master of Information & Data Science '17*
>>>>
>>>> *King's College London, MA Political Science '14*
>>>> *New York University, BA Political Science '12*
>>>>
>>>> -
>>>> LinkedIn <http://linkedin.com/in/madisonjmyers>
>>>>
>>>
>>
Re: Getting new users to adopt SystemML
Posted by Jeremy Anderson <je...@objectadjective.com>.
+1 all
...........................
Jeremy Anderson
https://twitter.com/ObjectAdjective
http://www.linkedin.com/in/objectadjective
On 18 October 2016 at 11:25, Renee Mascarinas <re...@gmail.com>
wrote:
> +1 Thanks Madison!
>
> I agree with everything. As a designer, it's been really hard to
> understand the value of SystemML without case studies. There's so much
> claims out there, but without real use cases, it's difficult to understand
> and promote when there aren't tangible proof from existing users.
>
> For visibility, I think it's a great idea to add SystemML as a skill on
> LinkedIn and posting questions and answers on Stack Overflow. From our
> research findings, the latter is a must for troubleshooting for data
> scientist, second to Google.
>
> Renee
>
> On Thu, Oct 6, 2016 at 7:17 PM, Jeremy Anderson <
> jeremy@objectadjective.com>
> wrote:
>
> > +1 Madison
> >
> > I started started doing some user research with a few other designers.
> > We're looking for patterns and pain points around adoption and use of
> > SystemML. We'd love to share our findings so far to continue this
> > discussion and get thoughts and feedback from everyone. I think clear
> > tutorials and site architecture will help address some of the concerns
> > we're hearing and could improve adoption. I love the idea of generating
> > more articles and content on Stack Overflow. All of this would be really
> > helpful in increasing awareness and building community.
> >
> > We're synthesizing our research so far and we'll share it here. Thanks
> for
> > starting this thread!
> >
> > Jeremy
> >
> > ...........................
> >
> > Jeremy Anderson
> > https://twitter.com/ObjectAdjective
> > http://www.linkedin.com/in/objectadjective
> >
> >
> > On 6 October 2016 at 10:33, Madison Myers <ma...@gmail.com>
> wrote:
> >
> > > Hi all,
> > >
> > > Thought it may be a good idea to implement some measures on getting new
> > > users to adopt SystemML. At the moment, I feel there is a big learning
> > > curve and that we could make it easier through clearer tutorials and
> > > examples, a better user experience on the website (continuing the
> process
> > > of making it better), more articles that explain SystemML at a high
> level
> > > and how to use it/why it's important. I also think we should get more
> > > questions and answers on Stack Overflow so that overcoming issues is a
> > bit
> > > easier for new users. Longer-term I'd also love to see a course taught
> > > online such as on Coursera as well as in university classrooms so that
> > > people in the field adopt SystemML more early-on.
> > >
> > > Also Deron and Mike had some interesting suggestions about visibility.
> > They
> > > suggested having projects using SystemML on personal Github accounts as
> > > well as mentioning SystemML as a skill on LinkedIn.
> > >
> > > Would love to know if others agree and/or have suggestions.
> > >
> > > Thanks!
> > >
> > > --
> > > *Madison J. Myers*
> > > *UC Berkeley, Master of Information & Data Science '17*
> > >
> > > *King's College London, MA Political Science '14*
> > > *New York University, BA Political Science '12*
> > >
> > > -
> > > LinkedIn <http://linkedin.com/in/madisonjmyers>
> > >
> >
>
Re: Getting new users to adopt SystemML
Posted by Renee Mascarinas <re...@gmail.com>.
+1 Thanks Madison!
I agree with everything. As a designer, it's been really hard to
understand the value of SystemML without case studies. There's so much
claims out there, but without real use cases, it's difficult to understand
and promote when there aren't tangible proof from existing users.
For visibility, I think it's a great idea to add SystemML as a skill on
LinkedIn and posting questions and answers on Stack Overflow. From our
research findings, the latter is a must for troubleshooting for data
scientist, second to Google.
Renee
On Thu, Oct 6, 2016 at 7:17 PM, Jeremy Anderson <je...@objectadjective.com>
wrote:
> +1 Madison
>
> I started started doing some user research with a few other designers.
> We're looking for patterns and pain points around adoption and use of
> SystemML. We'd love to share our findings so far to continue this
> discussion and get thoughts and feedback from everyone. I think clear
> tutorials and site architecture will help address some of the concerns
> we're hearing and could improve adoption. I love the idea of generating
> more articles and content on Stack Overflow. All of this would be really
> helpful in increasing awareness and building community.
>
> We're synthesizing our research so far and we'll share it here. Thanks for
> starting this thread!
>
> Jeremy
>
> ...........................
>
> Jeremy Anderson
> https://twitter.com/ObjectAdjective
> http://www.linkedin.com/in/objectadjective
>
>
> On 6 October 2016 at 10:33, Madison Myers <ma...@gmail.com> wrote:
>
> > Hi all,
> >
> > Thought it may be a good idea to implement some measures on getting new
> > users to adopt SystemML. At the moment, I feel there is a big learning
> > curve and that we could make it easier through clearer tutorials and
> > examples, a better user experience on the website (continuing the process
> > of making it better), more articles that explain SystemML at a high level
> > and how to use it/why it's important. I also think we should get more
> > questions and answers on Stack Overflow so that overcoming issues is a
> bit
> > easier for new users. Longer-term I'd also love to see a course taught
> > online such as on Coursera as well as in university classrooms so that
> > people in the field adopt SystemML more early-on.
> >
> > Also Deron and Mike had some interesting suggestions about visibility.
> They
> > suggested having projects using SystemML on personal Github accounts as
> > well as mentioning SystemML as a skill on LinkedIn.
> >
> > Would love to know if others agree and/or have suggestions.
> >
> > Thanks!
> >
> > --
> > *Madison J. Myers*
> > *UC Berkeley, Master of Information & Data Science '17*
> >
> > *King's College London, MA Political Science '14*
> > *New York University, BA Political Science '12*
> >
> > -
> > LinkedIn <http://linkedin.com/in/madisonjmyers>
> >
>
Re: Getting new users to adopt SystemML
Posted by Jeremy Anderson <je...@objectadjective.com>.
+1 Madison
I started started doing some user research with a few other designers.
We're looking for patterns and pain points around adoption and use of
SystemML. We'd love to share our findings so far to continue this
discussion and get thoughts and feedback from everyone. I think clear
tutorials and site architecture will help address some of the concerns
we're hearing and could improve adoption. I love the idea of generating
more articles and content on Stack Overflow. All of this would be really
helpful in increasing awareness and building community.
We're synthesizing our research so far and we'll share it here. Thanks for
starting this thread!
Jeremy
...........................
Jeremy Anderson
https://twitter.com/ObjectAdjective
http://www.linkedin.com/in/objectadjective
On 6 October 2016 at 10:33, Madison Myers <ma...@gmail.com> wrote:
> Hi all,
>
> Thought it may be a good idea to implement some measures on getting new
> users to adopt SystemML. At the moment, I feel there is a big learning
> curve and that we could make it easier through clearer tutorials and
> examples, a better user experience on the website (continuing the process
> of making it better), more articles that explain SystemML at a high level
> and how to use it/why it's important. I also think we should get more
> questions and answers on Stack Overflow so that overcoming issues is a bit
> easier for new users. Longer-term I'd also love to see a course taught
> online such as on Coursera as well as in university classrooms so that
> people in the field adopt SystemML more early-on.
>
> Also Deron and Mike had some interesting suggestions about visibility. They
> suggested having projects using SystemML on personal Github accounts as
> well as mentioning SystemML as a skill on LinkedIn.
>
> Would love to know if others agree and/or have suggestions.
>
> Thanks!
>
> --
> *Madison J. Myers*
> *UC Berkeley, Master of Information & Data Science '17*
>
> *King's College London, MA Political Science '14*
> *New York University, BA Political Science '12*
>
> -
> LinkedIn <http://linkedin.com/in/madisonjmyers>
>
Re: Getting new users to adopt SystemML
Posted by Luciano Resende <lu...@gmail.com>.
On Fri, Oct 7, 2016 at 5:43 PM, Jeremy Anderson <je...@objectadjective.com>
wrote:
> > Although this is unavoidable, if the projects are closely associated with
> > SystemML it might be better to have it collocated in the project, as it
> > might make it simpler for users to discover and consistently use.
>
> Thanks Luciano. I'm starting to think this may be the way we want to go,
> creating a research directory in the main apache repo. Our original
> thinking was we wanted to create one centralized repo for all data related
> ux research. I do love this idea, but our research to date is very specific
> to SystemML. We are hoping to link research findings to pull requests. Over
> time, this would provide context for why certain features/changes were made
> and make it easier for future development. If anyone has thoughts on how to
> best handle this, I'd love to hear. :)
>
+1 to incorporate this as one page in the systemml website or doc
--
Luciano Resende
http://twitter.com/lresende1975
http://lresende.blogspot.com/
Re: Getting new users to adopt SystemML
Posted by Jeremy Anderson <je...@objectadjective.com>.
> Although this is unavoidable, if the projects are closely associated with
> SystemML it might be better to have it collocated in the project, as it
> might make it simpler for users to discover and consistently use.
Thanks Luciano. I'm starting to think this may be the way we want to go,
creating a research directory in the main apache repo. Our original
thinking was we wanted to create one centralized repo for all data related
ux research. I do love this idea, but our research to date is very specific
to SystemML. We are hoping to link research findings to pull requests. Over
time, this would provide context for why certain features/changes were made
and make it easier for future development. If anyone has thoughts on how to
best handle this, I'd love to hear. :)
...........................
Jeremy Anderson
https://twitter.com/ObjectAdjective
http://www.linkedin.com/in/objectadjective
On 7 October 2016 at 12:38, Madison Myers <ma...@gmail.com> wrote:
> Also +1 Jeremy!
>
> On Fri, Oct 7, 2016 at 12:36 PM, Madison Myers <ma...@gmail.com>
> wrote:
>
> > Although this is unavoidable, if the projects are closely associated with
> > SystemML it might be better to have it collocated in the project, as it
> > might make it simpler for users to discover and consistently use.
> >
> > +1
> >
> > On Fri, Oct 7, 2016 at 7:37 AM, Luciano Resende <lu...@gmail.com>
> > wrote:
> >
> >> On Thu, Oct 6, 2016 at 10:33 AM, Madison Myers <madisonjmyers@gmail.com
> >
> >> wrote:
> >>
> >> > Hi all,
> >> >
> >> > Thought it may be a good idea to implement some measures on getting
> new
> >> > users to adopt SystemML. At the moment, I feel there is a big learning
> >> > curve and that we could make it easier through clearer tutorials and
> >> > examples, a better user experience on the website (continuing the
> >> process
> >> > of making it better), more articles that explain SystemML at a high
> >> level
> >> > and how to use it/why it's important. I also think we should get more
> >> > questions and answers on Stack Overflow so that overcoming issues is a
> >> bit
> >> > easier for new users. Longer-term I'd also love to see a course taught
> >> > online such as on Coursera as well as in university classrooms so that
> >> > people in the field adopt SystemML more early-on.
> >> >
> >> >
> >> +1
> >>
> >>
> >> > Also Deron and Mike had some interesting suggestions about visibility.
> >> They
> >> > suggested having projects using SystemML on personal Github accounts
> as
> >> > well as mentioning SystemML as a skill on LinkedIn.
> >> >
> >> >
> >> Although this is unavoidable, if the projects are closely associated
> with
> >> SystemML it might be better to have it collocated in the project, as it
> >> might make it simpler for users to discover and consistently use.
> >>
> >>
> >> > Would love to know if others agree and/or have suggestions.
> >> >
> >> > Thanks!
> >> >
> >> > --
> >> > *Madison J. Myers*
> >> > *UC Berkeley, Master of Information & Data Science '17*
> >> >
> >> > *King's College London, MA Political Science '14*
> >> > *New York University, BA Political Science '12*
> >> >
> >> > -
> >> > LinkedIn <http://linkedin.com/in/madisonjmyers>
> >> >
> >>
> >>
> >>
> >> --
> >> Luciano Resende
> >> http://twitter.com/lresende1975
> >> http://lresende.blogspot.com/
> >>
> >
> >
> >
> > --
> > *Madison J. Myers*
> > *UC Berkeley, Master of Information & Data Science '17*
> >
> > *King's College London, MA Political Science '14*
> > *New York University, BA Political Science '12*
> >
> > -
> > LinkedIn <http://linkedin.com/in/madisonjmyers>
> >
> >
>
>
> --
> *Madison J. Myers*
> *UC Berkeley, Master of Information & Data Science '17*
>
> *King's College London, MA Political Science '14*
> *New York University, BA Political Science '12*
>
> -
> LinkedIn <http://linkedin.com/in/madisonjmyers>
>
Re: Getting new users to adopt SystemML
Posted by Madison Myers <ma...@gmail.com>.
Also +1 Jeremy!
On Fri, Oct 7, 2016 at 12:36 PM, Madison Myers <ma...@gmail.com>
wrote:
> Although this is unavoidable, if the projects are closely associated with
> SystemML it might be better to have it collocated in the project, as it
> might make it simpler for users to discover and consistently use.
>
> +1
>
> On Fri, Oct 7, 2016 at 7:37 AM, Luciano Resende <lu...@gmail.com>
> wrote:
>
>> On Thu, Oct 6, 2016 at 10:33 AM, Madison Myers <ma...@gmail.com>
>> wrote:
>>
>> > Hi all,
>> >
>> > Thought it may be a good idea to implement some measures on getting new
>> > users to adopt SystemML. At the moment, I feel there is a big learning
>> > curve and that we could make it easier through clearer tutorials and
>> > examples, a better user experience on the website (continuing the
>> process
>> > of making it better), more articles that explain SystemML at a high
>> level
>> > and how to use it/why it's important. I also think we should get more
>> > questions and answers on Stack Overflow so that overcoming issues is a
>> bit
>> > easier for new users. Longer-term I'd also love to see a course taught
>> > online such as on Coursera as well as in university classrooms so that
>> > people in the field adopt SystemML more early-on.
>> >
>> >
>> +1
>>
>>
>> > Also Deron and Mike had some interesting suggestions about visibility.
>> They
>> > suggested having projects using SystemML on personal Github accounts as
>> > well as mentioning SystemML as a skill on LinkedIn.
>> >
>> >
>> Although this is unavoidable, if the projects are closely associated with
>> SystemML it might be better to have it collocated in the project, as it
>> might make it simpler for users to discover and consistently use.
>>
>>
>> > Would love to know if others agree and/or have suggestions.
>> >
>> > Thanks!
>> >
>> > --
>> > *Madison J. Myers*
>> > *UC Berkeley, Master of Information & Data Science '17*
>> >
>> > *King's College London, MA Political Science '14*
>> > *New York University, BA Political Science '12*
>> >
>> > -
>> > LinkedIn <http://linkedin.com/in/madisonjmyers>
>> >
>>
>>
>>
>> --
>> Luciano Resende
>> http://twitter.com/lresende1975
>> http://lresende.blogspot.com/
>>
>
>
>
> --
> *Madison J. Myers*
> *UC Berkeley, Master of Information & Data Science '17*
>
> *King's College London, MA Political Science '14*
> *New York University, BA Political Science '12*
>
> -
> LinkedIn <http://linkedin.com/in/madisonjmyers>
>
>
--
*Madison J. Myers*
*UC Berkeley, Master of Information & Data Science '17*
*King's College London, MA Political Science '14*
*New York University, BA Political Science '12*
-
LinkedIn <http://linkedin.com/in/madisonjmyers>
Re: Getting new users to adopt SystemML
Posted by Madison Myers <ma...@gmail.com>.
Although this is unavoidable, if the projects are closely associated with
SystemML it might be better to have it collocated in the project, as it
might make it simpler for users to discover and consistently use.
+1
On Fri, Oct 7, 2016 at 7:37 AM, Luciano Resende <lu...@gmail.com>
wrote:
> On Thu, Oct 6, 2016 at 10:33 AM, Madison Myers <ma...@gmail.com>
> wrote:
>
> > Hi all,
> >
> > Thought it may be a good idea to implement some measures on getting new
> > users to adopt SystemML. At the moment, I feel there is a big learning
> > curve and that we could make it easier through clearer tutorials and
> > examples, a better user experience on the website (continuing the process
> > of making it better), more articles that explain SystemML at a high level
> > and how to use it/why it's important. I also think we should get more
> > questions and answers on Stack Overflow so that overcoming issues is a
> bit
> > easier for new users. Longer-term I'd also love to see a course taught
> > online such as on Coursera as well as in university classrooms so that
> > people in the field adopt SystemML more early-on.
> >
> >
> +1
>
>
> > Also Deron and Mike had some interesting suggestions about visibility.
> They
> > suggested having projects using SystemML on personal Github accounts as
> > well as mentioning SystemML as a skill on LinkedIn.
> >
> >
> Although this is unavoidable, if the projects are closely associated with
> SystemML it might be better to have it collocated in the project, as it
> might make it simpler for users to discover and consistently use.
>
>
> > Would love to know if others agree and/or have suggestions.
> >
> > Thanks!
> >
> > --
> > *Madison J. Myers*
> > *UC Berkeley, Master of Information & Data Science '17*
> >
> > *King's College London, MA Political Science '14*
> > *New York University, BA Political Science '12*
> >
> > -
> > LinkedIn <http://linkedin.com/in/madisonjmyers>
> >
>
>
>
> --
> Luciano Resende
> http://twitter.com/lresende1975
> http://lresende.blogspot.com/
>
--
*Madison J. Myers*
*UC Berkeley, Master of Information & Data Science '17*
*King's College London, MA Political Science '14*
*New York University, BA Political Science '12*
-
LinkedIn <http://linkedin.com/in/madisonjmyers>
Re: Getting new users to adopt SystemML
Posted by Luciano Resende <lu...@gmail.com>.
On Thu, Oct 6, 2016 at 10:33 AM, Madison Myers <ma...@gmail.com>
wrote:
> Hi all,
>
> Thought it may be a good idea to implement some measures on getting new
> users to adopt SystemML. At the moment, I feel there is a big learning
> curve and that we could make it easier through clearer tutorials and
> examples, a better user experience on the website (continuing the process
> of making it better), more articles that explain SystemML at a high level
> and how to use it/why it's important. I also think we should get more
> questions and answers on Stack Overflow so that overcoming issues is a bit
> easier for new users. Longer-term I'd also love to see a course taught
> online such as on Coursera as well as in university classrooms so that
> people in the field adopt SystemML more early-on.
>
>
+1
> Also Deron and Mike had some interesting suggestions about visibility. They
> suggested having projects using SystemML on personal Github accounts as
> well as mentioning SystemML as a skill on LinkedIn.
>
>
Although this is unavoidable, if the projects are closely associated with
SystemML it might be better to have it collocated in the project, as it
might make it simpler for users to discover and consistently use.
> Would love to know if others agree and/or have suggestions.
>
> Thanks!
>
> --
> *Madison J. Myers*
> *UC Berkeley, Master of Information & Data Science '17*
>
> *King's College London, MA Political Science '14*
> *New York University, BA Political Science '12*
>
> -
> LinkedIn <http://linkedin.com/in/madisonjmyers>
>
--
Luciano Resende
http://twitter.com/lresende1975
http://lresende.blogspot.com/