23 Matching Annotations
  1. Last 7 days
    1. The response begins

      As a reader, I feel like I start entering into the timeline and technical component without really understanding the context or what the real issues of concern are.

      Therefore I would suggest putting in as a second slide some context before you go into 'The response begins"

      I suggest the context cover:

      First, I think you need a new slide something simple that helps the reader to understand the problem that satellite damage assessments solve, their limitations, and why that poses challenges to responders. For example, what used to happen after an earthquake (before we had these type of assessments)? What do these assessments help with (ie. to determine where to deploy SAR, potential people in need humanitarian assistance)? What are the downsides of satellite damage assessments (ie. foreshadow why you can have assessments that come up with such different levels of building damage, or why it might show a building damaged when it's not), And therefore why is it a challenge for humanitarian's that there are so many assessments now?

      2) You need to briefly explain what the Centre's role was in Venezuela (later you reference a 'viewer') and its connection to Copernicus EMS. Otherwise you go straight into Copernicus and the reader doesn't understand why.

    2. For evaluators: keep testing this methodology across different crises, and build the system that makes it routine: a required specification for contributed products (analysed extent, metadata, a common building base) and a standing way for providers to contribute into a shared space, so that combining and evaluating starts the day the first product arrives rather than weeks later.

      I didn't really understand this one, but as long as it's using language that evaluators understand that's fine.

    3. (three of the six reach a Spearman ρ of 0.7 or more against the Copernicus EMS ordering)

      maybe put in the technical brief unless you think this will be widely understood.

    4. and it needs a shared picture of who has looked where to work.

      ? Meaning this is what we provided through the 'viewer' that can put all the assessments together?

    5. During this response, an ad hoc viewer our team stood up within days let responders see who had looked where. It filled part of the gap, not all of it, and nothing had planned for it.

      This feels like it comes out of nowhere and it needs some clarity eg. "who had looked where" - meaning responders to see what satellite data was available? Part of what gap and what does 'nothing had planned for it'.

      Perhaps this needs to be part of the beginning context - because this is the first time we're seeing what role the Centre had this.

    6. The conclusion at the bottom of the graph is good and should be its own box to the side. Also does using the assessments then help to overcome the issue raised in the slide above that ranking by neighborhood only works where it's in an area already known to be damaged? If so, I think it should be clearer that by using the assessments together you can use it not only within a neighborhood to identify with greater confidence what buildings are damaged, but also in areas where you may not have an EMT deployed yet (presumably).

      Postscript :) I see you then go back into the neighborhoods with the next slide.

    7. Ranking orders that zone; it does not find one, and it never names a building.

      This didn't make sense to me, so perhaps it needs to be reworded.

    8. One caveat carries through: this holds inside an area already known to be badly hit.

      This part of the answer is good but then the caveat confused me - meaning we couldn't say the same for another area where EMS weren't on the ground verifying?

    9. The Answer and the blurb at the bottom of the graph "Even on the most generous reading..." Is good. It needs to be its own box and pulled out. The answer is too buried. I want to see the answer and analysis before I look at the graph - it helps the reader to understand it.

    10. It took me a few reads to get this and I wonder if it's just a placement issue. I wonder if this would be better placed in the next section after the questions and before the baseline - meaning the first step is looking at where the assessments can be compared in relation to where Copernicus was operating.

      I would also suggest deleting "manual mapping on very-high resolution imagery, not modeling). I don't think it's necessary to understand the story and so anywhere you can cut technical detail I would.

    11. A stack of assessments of the same disaster: PDFs, shapefiles, dashboards. Different methods, different extents, and answers from 9,635 to 163,225 damaged buildings, a factor of seventeen. Which do you act on?

      This is great, but give a bit more info as to why this is hard for humanitarians.

      The S&R audience probably have a pretty good idea, but for those providing these assessments giving them a clear picture of why this is difficult - the types of decisions that have to be made and why having different assessments makes that harder – will be very helpful and eye opening to them. You want to give this audience the "ah-ha moment" now I get it why this would be hard.

    12. Two further products arrived in the same window but are not evaluated here: HOT's fAIr, an AI-plus-crowd product that publishes no analysed extent, so its misses and false alarms cannot be counted fairly; and DISHA (UN Global Pulse), which is licence-restricted and covers less than half of the shared strip.

      This is quite technical, you could perhaps streamline to "Two further products arrived in the same window but are not evaluated here" and then put the note more detailed explanation in the methodology. The fuller explanation interrupts the flow.

    13. relating detected damage to debris tonnage

      Not sure I understand this? Meaning it detects where there is debris tonnage or the data that comes from it giving an estimate of debris tonnage based on the damage?

    14. At the end here, suggest a short line on what the purpose of this GitHub is - is to explain how to make use of it operationally? This opening slide is like the summary so the reader needs to know up front where it's going.