09-13-2015 06:00 AM - edited 09-13-2015 06:04 AM
Could you please clarify what constitutes an "original work" per the contest rules? Would an IVisual-implementation of an existing D3 visualization be considered original?
I ask in light of the recent bullet chart entry (http://community.powerbi.com/t5/Best-Visual-Contest/Bullet-Chart-by-SQLBI/cns-p/6774). While the previous entries have been unique D3 from all appearances, the bullet chart is a credited IVisual adaptation of the example found on the D3 website (http://bl.ocks.org/mbostock/4061961). The attribution to the work of others is in the license for that viz on Github, so it meets the MIT criteria and could (and should!) be included in Power BI. I think that it would be a well-received and great addition to the software. From a contest perspective though, how would something like this be judged? There is significant work needed to adapt an existing viz to the IVisual interface, but is the originality in the viz itself or in how it has been adapted to IVisual?
In short, are the examples found in the D3 gallery (https://github.com/mbostock/d3/wiki/Gallery) open for adaptation as the bullet chart has been, and would adaptations like this be considered "original" and able to win this contest?
09-14-2015 01:09 PM
Just my 2 cents about that: I would love to see the entire D3 gallery implemented in Power BI, so the contest could be a nice boost for that. I think that the evaluation of these components will certainly consider the lower "originality" of such components, giving more credit to brand new components.
In a open source world, I would not count the number of lines inherited by other components but the overall usability and usefulness of the final result. In terms of original work, I would consider the bullet chart as a simple implementation of Stephen Few design, and this is more important than the amount of code inherited by D3 gallery. A brand new idea would be certainly more valuable (for the contest evaluation).
Anyway, this is my humble opinion.
A clarification from organizers will be certainly useful to avoid disappointement by the contributors.
09-14-2015 02:36 PM
Marco, thanks for weighing in. In my opinion, I would also like to see numerous other examples from the D3 gallery added to Power BI, whether contest-eligible or not. Part of the reason that I had hoped to seek more clarity on how the judges will perceive "originality" is in the hope that more adaptation and contributions may be possible. I think that it would drive more interest if entrants knew early on that originality was based on the *entire development process* for a visual and not simply a closely scrutinized look at every aspect of the D3 code alone.
For anyone familiar with the custom visuals framework, it involves much more than D3 development. There is no way someone could simply copy and paste from the gallery and have a functional visualization. Developers could easily invest more time determining a visual's Power BI-specific capabilities, data binding, etc. than in the D3 code itself. Hence the desired need for clarification on what constitutes "original". Barring all-out plagiarism, something "inspired by" an existing D3 viz would be original in my opinion due to the significant work needed for adaptation to Power BI.
In any case, I do not know what else Daniele and others still have planned, but I would hate to see someone invest a lot of time on something only to find out a viz could be disqualified at the end of the contest due to vague wording in the Official Rules. This requirement is not even stated on the main contest page. Hopefully some clarification can be provided soon.
09-15-2015 08:35 AM
Sorry for the delay in driving clarity here.
Let's start with philosophy: the reason we contributed the Power BI visualization stack (and the IVisual interface spec) to the open source community was to make it easy to extend Power BI with new visualizations. One of the great things about OSS is that it allows one to build atop the great work of others, adding value versus reinventing. We embrace OSS wherever possible, and obviously contribute back as well.
As noted above, one doesn't simply take D3 and use it. There is work to be done to conform it to the IVisual interface, one can add additional features and capabilities, etc. But there is no judging "ding" associated with starting from the good work of others and extending it for use in Power BI (assuming you have the right to use that other work in this way). That's how our industry works now.
If all we got in this contest were a lot of "wrapped" D3 visuals that worked in Power BI, that would be a huge win for the community - we'd all have a lot more options for use in our Power BI reports and dashboards (as soon as we allow custom visuals to be generally used in Power BI...coming very soon).
Thanks for participating in the contest and I'm super excited to see the collective creativity of the community unleashed.
09-15-2015 09:15 AM
Following up on what James said, the entry is a valid entry for the contest. We’re excited to see participants build on top of other Open Source projects when building their entries. While we’ve not done a deep technical scrub of the entry yet, it seems to implement IVisual, and integrates with Power BI. We will judge all entries based on creativity, applicability and technical excellence for contest prizes. Obviously there are multiple paths to get the Grand Prize . Regarding “original work” – this means that the code the entrant submits for the contest that they claim is written by them, actually is. We expect anyone who leverages existing open source work to correctly attribute to the original authors and preserve any required licenses for the pre-existing open source code.