- Each proposal is evaluated on the following 5 points on a scale of 0 to 10 with 0 the lowest and 10 the highest (except when otherwise mentioned).
- The criteria are
- Clarity of the proposal (summary field)
- Clarity of the talk details (Outline)
- Bio of authors (reputation, contributions etc.)
- Novelty of the topic (1 for a dead topic, 5 for a novel one).
- Competing talks in the same category (0 if no competing talk, # of competing talks if there are any)
- Overall rating
There is no formula for arriving at the overall rating from the other ratings - it is based on the person who does the evaluation. However the suggestion is to look at items 1, 2 and 3 in that order to arrive at the rating. Novelty of the topic should not be used to arrive at the final rating, but can be used at the end to order the talks which are tied on scores. Topic which are more novel will be ranked higher than topics which are run-of-the-mill. We will be using only the final rating to evaluate the talks.
The final evaluations will be done on a scale of 25, so we will simply scale the scores in 10 to 25.
Talks which are not complete in their submission will be rejected. A sumbission can be incomplete in the following respects.
- No proper author name, instead only a nickname is provided (For example "johndoe" instead of "John W Doe").
- Missing outline or summary.
- Missing author bio.
For those talks which are not rejected, but are still incomplete in their submissions, we will ask the author to do a re-submission after editing the required fields. A talk can be slotted for re-submission if,
- It doesn't satisfy the rejection criteria above.
- It satisfies one or more of the criteria below.
- The connection of the topic with Python is not clear.
- The outline is not proper. An outline has to contain at least 2 or 3 points
- on the detailed content the author is going to present.
- The talk outline is too long so as to not fit the time slot.
- The talk seems to be more suited for a tutorial track than a regular talk slot.
- The talk is submitted in the wrong category.
- The talk looks like it could be a marketing pitch on a product/company by the author(s).
- (Talks on true open source projects are exempt from this).
(The above are only the most important criteria for not accepting a talk, there could be more which are subjective.)
If a talk is slotted for re-submission and not re-submitted within the deadline (which will be announced), it is automatically rejected.
There will be four people who will be judging a talk, each making a rating out of 25. The total rating of the talk will be the sum of all ratings of each reviewer so as to get the final rating out of 100.
We will be selecting up to 60 talks initially. All talks will be ordered in decreasing order of overall rating out of 100 and the first 60 talks will be selected. Novelty factor and other factors (# of talks submitted for the same topic, # of talks submitted by the same author etc) will be used to select among talks having the same score.