front page

[jywarren]

We need to introduce the project on the front page, and display some data. How much and what kind of data can we show? @shapironick - to what degree do we want to share aggregate graphs or narratives with the broader, non-signed-up public? @novakn - can we mitigate what effect this might have on answers from people who later sign up to contribute?

[jywarren]

Ok, so I took a very rough look at content for a front page. I stayed away from graphs for the time being but think it'd be a very important part of showing what the site is about, along with some sample narratives. And the ability to begin browsing these graphs/stories without logging in, i think?

[shapironick]

This looks amazing! Thanks Jeff! I think we can definitely hold off on putting any graphs up front until phase two.

I like the join the conversation approach as many people might not be clear on what it means to join a website. so maybe it should be more a focus on joining the conversation rather than joining WWB, as some may feel like its a political endorsement to 'join' us. We can explain that we need them to register so we get the facts straight and so they can have more control over what they share.

But perhaps we can put a picture or two up there?

or

Photos by Akasha Rabut http://www.akasharabut.com/

[novakn]

I love the picture idea, and agree that it might help ground people.

Regarding showing aggregate graphs on the public page (or really, to anyone who hasn't completed the full survey yet), I think it would be best to stick to questions that are more concrete:

• age,
• how long you've lived in a house,
• make/manufacture.

since people's answers to those are less likely to be changed if they see information about other respondents.

We should display as few as possible:

• symptom questions,
• questions about what people think is causing their symptoms,

since those are the ones that a scientific audience might consider most vulnerable to bias.

On Fri, Jun 20, 2014 at 10:58 AM, shapironick notifications@github.com wrote:

This looks amazing! Thanks Jeff! I think we can definitely hold off on putting any graphs up front until phase two.

I like the join the conversation approach as many people might not be clear on what it means to join a website. so maybe it should be more a focus on joining the conversation rather than joining WWB, as some may feel like its a political endorsement to 'join' us. We can explain that we need them to register so we get the facts straight and so they can have more control over what they share.

But perhaps we can put a picture or two up there?

[image: 80980006] https://cloud.githubusercontent.com/assets/7503343/3341686/c4ce8c86-f88a-11e3-8216-8918c116a825.JPG or [image: k] https://cloud.githubusercontent.com/assets/7503343/3341694/d5a0bd0e-f88a-11e3-9a33-19ef0b444434.jpg

— Reply to this email directly or view it on GitHub https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-46687733 .

[jywarren]

I really think there has to be some discussion of symptoms publicly viewable, since a major point of the project is to inform the public discussion about these issues. If there's no compelling data to people who haven't contributed, then the purpose of the site is narrower -- just to enable affected people to communicate with each other, not to share their stories and data with the world. Less advocacy and more organizing/mutual-support. Thoughts?

Love the photos, thanks. Are they yours? How do we credit them and can they be Creative Commons?

[novakn]

Hey Jeff, I absolutely understand the need to reconcile these different purposes and can see how it would be valuable to have information about people's symptoms more widely visible. (I should note, collecting data that is considered robust to scientific audiences is another way we will be able to advocate, albeit in the longer term). (Secondarily, and my apologies if this makes things even more frustrating, if we are planning to do follow-up surveys with people over time, I even have concerns about showing people within the site aggregate symptom data after they have taken the survey).

One way to deal with this would be to pick a few "public symptoms" that we choose to display publicly, with the understanding that this data will become less useful for the scientific audiences. The tricky thing with this is that many of the symptoms that would be most prevalent and interesting/compelling to a public audience (fatigue, headaches, respiratory symptoms) are also central to the scientific questions we've committed to answer.

I don't mean to be a pain, I just feel like it is my job to point these things out. I could obviously work with data that was collected after people had seen aggregate results, but it would limit the data's perceived validity in more conventional scientific settings.

On Fri, Jun 20, 2014 at 4:16 PM, Jeffrey Warren notifications@github.com wrote:

I really think there has to be some discussion of symptoms publicly viewable, since a major point of the project is to inform the public discussion about these issues. If there's no compelling data to people who haven't contributed, then the purpose of the site is narrower -- just to enable affected people to communicate with each other, not to share their stories and data with the world. Less advocacy and more organizing/mutual-support. Thoughts?

Love the photos, thanks. Are they yours? How do we credit them and can they be Creative Commons?

— Reply to this email directly or view it on GitHub https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-46721832 .

[shapironick]

This is a really important exchange.

I think we should definitely have some mention of symptoms on the front page. One of the things that Nicole mentioned a while back, and what really stuck with me, was her mention of a preponderance of studies which found individuals to be more actuate in assessing their health than doctors (or something very similar). While pretty obvious sounding, I found that to be very empowering. Perhaps we could have something like: "On average, formaldehyde levels are much higher in manufactured homes than site-built homes but scientists still are not certain of how such exposures affect inhabitants. Studies have shown people everyday people have a better understanding of their health than doctors do, share your symptoms and teach scientists about the impacts of formaldehyde." (on a somewhat related riff I think what would be really amazing down the line is to develop a way for users to study their doctors and report on how their doctors don't take into account environmental factors. a very nice reversal of the scientific gaze)

I'm not sure how much people will be motivated by seeing a specific number of how many people have whatever symptom as these are not likely to be data-driven quantified self types. But some mention of health effects is important. Also, this is a situation in which many people are likely to find the site by searching for "mobile home sickness" "sick house" "indoor air mobile home" or "formaldehyde symptoms" so selection bias will be at play but also people may have googled 'formaldehyde symptoms' before so they are likely not 'blank slates' when it comes to formaldehyde exposure. I just googled the symptoms and google told me in a little box that long term exposure can result in several symptoms: "respiratory symptoms, and eye, nose, and throat irritation" @novakn https://github.com/novakn is there some way that if we put those symptoms on the front page that we could somehow level the playing field between those that have done research on formaldehyde and those that haven't? But maybe I'm coming at this from the wrong way?

We could maybe have a call to talk these things out, if y'all want?

The photos are my friend Akasha Rabut's We should credit her and I'll ask her about creative commons. She's probs down for creative commons but will ask her over the next couple days. http://www.akasharabut.com/

On Fri, Jun 20, 2014 at 10:05 PM, novakn notifications@github.com wrote:

Hey Jeff, I absolutely understand the need to reconcile these different purposes and can see how it would be valuable to have information about people's symptoms more widely visible. (I should note, collecting data that is considered robust to scientific audiences is another way we will be able to advocate, albeit in the longer term). (Secondarily, and my apologies if this makes things even more frustrating, if we are planning to do follow-up surveys with people over time, I even have concerns about showing people within the site aggregate symptom data after they have taken the survey).

One way to deal with this would be to pick a few "public symptoms" that we choose to display publicly, with the understanding that this data will become less useful for the scientific audiences. The tricky thing with this is that many of the symptoms that would be most prevalent and interesting/compelling to a public audience (fatigue, headaches, respiratory symptoms) are also central to the scientific questions we've committed to answer.

I don't mean to be a pain, I just feel like it is my job to point these things out. I could obviously work with data that was collected after people had seen aggregate results, but it would limit the data's perceived validity in more conventional scientific settings.

On Fri, Jun 20, 2014 at 4:16 PM, Jeffrey Warren notifications@github.com

wrote:

I really think there has to be some discussion of symptoms publicly viewable, since a major point of the project is to inform the public discussion about these issues. If there's no compelling data to people who haven't contributed, then the purpose of the site is narrower -- just to enable affected people to communicate with each other, not to share their stories and data with the world. Less advocacy and more organizing/mutual-support. Thoughts?

Love the photos, thanks. Are they yours? How do we credit them and can they be Creative Commons?

— Reply to this email directly or view it on GitHub < https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-46721832>

.

— Reply to this email directly or view it on GitHub https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-46726537 .

[novakn]

Hey guys,

I agree that this could be a good point for us to have a more big-picture chat about science/advocacy and the way we want to navigate them together. I'd love to have a call and am pretty flexible this week, so let me know what works best!

On Nick's point, I think what I had mentioned was literature finding that self-rated health can be a better predictor of mortality than physician assessment or other known risk factors. I put a few papers on this in our Dropbox, Nick, and the citations for ones we might want to cite on the website are included below:

DeSalvo KB, Muntner P. Discordance between physician and patient self-rated health and all-cause mortality. Ochsner J 2011;11(3):232-240.

Jylha M. What is self-rated health and why does it predict mortality? Toward a unified conceptual model. Soc Sci Med 2009;69:307-316.

From the abstract of the Jylha paper: "A unique source of information is provided by the bodily sensations that are directly available only to the individual him- or herself. According to recent findings in human biology, these sensations may reflect important physiological dysregulations, such as inflammatory processes." I can look into this more if you want.

I think those papers could back up the text Nick suggested above, with a few tiny tweaks:

"On average, formaldehyde levels are much higher in manufactured homes than site-built homes but scientists are not certain how such exposures affect inhabitants. Studies have shown that everyday people have a better understanding of their health than doctors do, so share your symptoms and teach scientists about the impacts of formaldehyde."

I am finishing up responding to Nick's questions on the questionnaire--I will send you guys the new one later today.

[shapironick]

I think the survey looks ready to go!

Nicole's point is basically, we are doing so many novel things here in terms of data collection--that are critiques-as-practice--that we risk being being totally scientifically disqualified if we mention any symptoms up front. So if putting symptoms up front draws more people but also undercuts everyone's data then I think its kinda a hard sell. Not trying to close down the conversation here, just connect the different registers that we are talking on.

Akasha has given us the green light to use those photos with attribution and the standard public lab CC licensing. I also have TONS more taken by both her and me across the country. I chose these because they felt the most generalizable (ie not just tiny tiny FEMA trailers and instances of profound poverty).

((Related to the infragram project @jywarren these photos http://www.richardmosse.com/works/infra/#1 appropriate infrared film that was originally developed to detect military camouflage and are pretty stunning))

[jywarren]

getting back to this important thread shortly, sorry for delay, and quick update on front page image:

[novakn]

beautiful!

On Thu, Jul 17, 2014 at 11:36 AM, Jeffrey Warren notifications@github.com wrote:

getting back to this important thread shortly, sorry for delay, and quick update on front page image:

[image: screen shot 2014-07-17 at 11 36 08 am] https://cloud.githubusercontent.com/assets/24359/3615404/1f05154c-0dc8-11e4-8366-6b1fc320b324.png

— Reply to this email directly or view it on GitHub https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-49323729 .

[shapironick]

This is fantastic!

On Thu, Jul 17, 2014 at 6:50 PM, novakn notifications@github.com wrote:

beautiful!

On Thu, Jul 17, 2014 at 11:36 AM, Jeffrey Warren notifications@github.com

wrote:

getting back to this important thread shortly, sorry for delay, and quick update on front page image:

[image: screen shot 2014-07-17 at 11 36 08 am] < https://cloud.githubusercontent.com/assets/24359/3615404/1f05154c-0dc8-11e4-8366-6b1fc320b324.png>

— Reply to this email directly or view it on GitHub < https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-49323729>

.

— Reply to this email directly or view it on GitHub https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-49333345 .

[jywarren]

OK, i'm sorry i've taken so long to respond to this -- partially catching up after my trip but also digesting these ideas a bit.

I think the crux of the issue (for which up-front display of data is only one consideration) is what this site is intended to accomplish. I think we have different ideas, but they're not necessarily at odds if we do it right. The site could variously be:

1. a means for epidemiologists to collect data about respiratory health
2. a means for epidemiologists to collaborate with people with respiratory issues
3. a means for people affected by respiratory issues to communicate/collaborate with each other, and/or organize

The goals could variously be:

1. to advocate for change related to manufactured housing
2. to help people living in manufactured homes learn about air quality risks and find resources
3. to develop new research to better scientific understanding of respiratory health issues
4. to improve public understanding of respiratory risks in manufactured housing

There are more combinations, of course, including the one Nick suggested of people studying epidemiologists. But it's going to be important to be very clear about which of the above are the primary functions/goals of the site. If it seems like we're developing an informational or collaborative site, but then a lot of the data goes to a research team and not much is made public, that could seem disengenuous, and damage our credibility with people using the site. A hybrid model which could work might be to carefully balance the site's usefulness to people participating with its usefulness to formal researchers and research.

Do people primarily feel that they are sending information to epidemiologists, or sharing it with one another? Do they feel comfortable with either or both of those possibilities? One way to potentially break this down a bit is to say that:

A. Sharing aggregate data is valuable to researchers and research and minimally risks privacy (vs. specific data). B. Sharing specific narratives & resources and communicating/collaborating with others via the site may be more useful to people not engaged in formal research.

We could make it possible to communicate/collaborate/share only with other contributors, which places greater burden on formal researchers to communicate and build collaborations in order to be granted access to personal data? Then there is a bit of a trustful walled garden within which everyone is sharing health data equally, giving and getting -- focusing on B, above. But I can see how this would be difficult to reconcile with A -- how do we successfully strike that balance?

[mmnoo]

1. Regarding the competing ideas about open data vs. data in a walled garden looked after by experts, could we delay openness so that the data is good data. For example, if the epidemiological research project will be pulling the data on Jan 1, 2015 to commence their data analysis, up to that date we take the walled garden approach. After that date, we could release the data aggregated and/or presented in a interactive section of the website.
   • This also gives us an acceptable excuse to email our survey takers and potentially reignite interest in people who will have likely moved on to a ton of other flashy content on the internet and already have long forgotten us. "Find out how others responded to the "x" survey you took on indoor air quality!" - we could have a checkbox on sign up "check if you want to be notified when the survey results go public"
   • This scenario could be improved by some communication to users about reducing bias, and whatever else, being the reason that the data isnt available yet.
2. With a narratives page having a bunch of interactive graphs, we are assuming that a significant proportion of our audience will have the 'geek orientation' and love poking around charts.
   • Do we need to validate that assumption? Maybe we can have an empty "coming soon" page that is called something like "explore survey results", and use something like Google Analytics to measure how many clicks the page gets in proportion to the rest of the site. We can then use that click data to estimate interest, which will inform if we merely put up a link to download our aggregated data (potentially at a later phase as suggested in the first point) or if we should invest time in creating a section where people can click around the data. (Jeff, what's your experience with Public Lab and the level of geeky interest in these things? What proportion of your audience navigates through to the data and charts?)

[shapironick]

Hi All, I'm on holiday this and last week so my apologies if this is a discombobulated response. Just trying to keep the conversation going in a timely manner.

I see the project as being about evidence-based advocacy and organizing. The science is a means of both seeing what the potential health issues of mobile home owners are and then capturing evidence of those harms in a way that will enable advocacy. Our methods will be educational, help those affected advocate for change, and help the exposed mitigate their exposures (eventually). Beyond case specific aspects, our epidemiological methods could be extrapolated to a broad number of spatially distributed environmental issues that have proven difficult to apprehend on both a social and a technical level. I think there is also an important designerly role here in figuring out more-than-data ways for these issues to be apprehended as history has made it very clear that science alone is not enough for regulatory or social change in relation to toxics. I think these are questions for the second stage, but we can start making space for them, opening up their possibilities, in the first stage. I see A (in Jeff’s comments above) to be our primary priority now but B is important for communicating, collaborating, and networking which will be our primary priority after we have collected enough information to give B the evidentiary fodder needed to strike up meaningful contestations of prevailing regulatory conventions. I don’t see A as more important than B, but I do see it as an important first step.

So I think the primary functions of the site will change with time, like Melissa suggests. As I see it the goals are 1) Involve the users in problem formation and evidence capture and 2) organize and advocate. I think the stronger the first step the stronger the second step. So it might feel like it's not as participatory as it could be at the beginning but this will be for reasons that will lend credence to the voices of users in the second stage. The two stages won’t be separate, but primary emphases will shift. It might be difficult to have a time-based deadline (like switching on a specific date) but we could begin shifting priorities once we have say 200 complete surveys.

Jeff, in terms of walled garden approach do you mean no data would be shared privately to the research team, and no stories shared publicly? So there would only be private but private would mean anybody who could cook up a VIN number that wasn’t already registered? If so, i think that's super interesting but I don’t think having access to a VIN or HUD number is enough of a wall for me to feel like we are creating a safe space (IRB’s may also feel like that might be a bit of a false security). If we did want to go that route what additional barriers to entry would we want and how would they change both our pool of users and the type of information shares and the community that can be sewn?

I do see this project as science-led for strategic reasons, but science that critiques standard modes of science, and science that uses its authority to mobilize people effectively. Also the science that we will be using will be participatory (citizen contributions, and later DIY monitoring, DIY studies of doctors that disqualify environmental issues etc). Mostly just repeating these things so I can wrap my head around it.

I think Melissa's question about how geeky our demographic will be is spot on. I think the graphs will be at least ephemerally interesting but connecting people with similar symptoms, exposure levels, jurisdictional boundaries (state), or trailer types, would be great for both organizing and psychologically knowing real people who share their plight.

On Mon, Jul 21, 2014 at 11:48 PM, Melissa notifications@github.com wrote:

1.

Regarding the competing ideas about open data vs. data in a walled garden looked after by experts, could we delay openness so that the data is good data. For example, if the epidemiological research project will be pulling the data on Jan 1, 2015 to commence their data analysis, up to that date we take the walled garden approach. After that date, we could release the data aggregated and/or presented in a interactive section of the website.

• This also gives us an acceptable excuse to email our survey takers and potentially reignite interest in people who will have likely moved on to a ton of other flashy content on the internet and already have long forgotten us. "Find out how others responded to the "x" survey you took on indoor air quality!" - we could have a checkbox on sign up "check if you want to be notified when the survey results go public"

  • This scenario could be improved by some communication to users about reducing bias, and whatever else, being the reason that the data isnt available yet. 2.

  With a narratives page having a bunch of interactive graphs, we are assuming that a significant proportion of our audience will have the 'geek orientation' and love poking around charts.

• Do we need to validate that assumption though? Maybe we can have an empty "coming soon" page that is called something like "explore survey results", and use something like Google Analytics to measure how many clicks the page gets in proportion to the rest of the site. We can then use that click data to estimate interest, which will inform if we merely put up a link to download our aggregated data (potentially at a later phase as suggested in the first point) or if we should invest time in creating a section where people can click around the data. (Jeff, what's your experience with Public Lab and the level of geeky interest in these things? What proportion of your audience navigates through to the data and charts?)

— Reply to this email directly or view it on GitHub https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-49670717 .

[shapironick]

Hi All,

Just re-posting some of my comments from #31, threading them up.

• The front picture works really well, but I wonder if it would be possible to have more of the home visible, so it can be a little more immediately recognizable. Even if the crest of the roof can’t fit. It would be nice to be able to see the window. o The test along the asphalt works well but maybe the research question could be along the top in the trees and the join WWB could be along the bottom? • Maybe the first thing should be to identify the broad user group and the top question above the photo could be “Do you live in a mobile home?” And we could combine that with a statement of the problem, so the top question could be “Do you live in a mobile home? Did you know that studies have shown that manufactured housing to have elevated levels of formaldehyde, compared to site-built homes?” That would then lead into the photo and the question more smoothly. o Also the first sentence above the photo and the first sentence in the text box are slightly repetitive. The broadness of the second version I think is helpful on the first page and maybe “Sharing your story towards a better public understanding of indoor air quality health issues” is more of a motto that one needs to first understand who the “you” is and why air quality is an important issue for them. I think it’s a good motto and could maybe go in the footer. • Should we first identify ourselves as simply as possible? So perhaps instead of identifying ourselves as Public Lab and various universities we could first say that “Where We Breath is asking” and we can put PL and universities into the footer, if we need to have that on the front page. At that point in their reading of the website I’m not sure that they will have an understanding of what WWB is and they will maybe be reticent to join what they don’t understand. • Maybe we could rephrase the research question to say “Where We Breathe is website where manufactured home residents can work with our research team towards answering the question: How does formaldehyde exposure affect the residents of manufactured homes?” I think the “how” is perhaps more engaging than the yes/no way that we are currently posing the question. • Use this site to share information about indoor air quality health issues and read others' stories.” • Should we mention the tools we bring to the problem? We cover the user group, and the problem but not our tools, which might be the last part of the triad.

[jywarren]

Just briefly, let's hold off on the lead image tweaks, I'd like to address them myself in a bit. Thanks, Nick.

On Mon, Jul 28, 2014 at 1:53 PM, shapironick notifications@github.com wrote:

Hi All,

Just re-posting some of my comments from #31 https://github.com/publiclab/wherewebreathe/issues/31, threading them up.

• The front picture works really well, but I wonder if it would be possible to have more of the home visible, so it can be a little more immediately recognizable. Even if the crest of the roof can’t fit. It would be nice to be able to see the window. o The test along the asphalt works well but maybe the research question could be along the top in the trees and the join WWB could be along the bottom? • Maybe the first thing should be to identify the broad user group and the top question above the photo could be “Do you live in a mobile home?” And we could combine that with a statement of the problem, so the top question could be “Do you live in a mobile home? Did you know that studies have shown that manufactured housing to have elevated levels of formaldehyde, compared to site-built homes?” That would then lead into the photo and the question more smoothly. o Also the first sentence above the photo and the first sentence in the text box are slightly repetitive. The broadness of the second version I think is helpful on the first page and maybe “Sharing your story towards a better public understanding of indoor air quality health issues” is more of a motto that one needs to first understand who the “you” is and why air quality is an important issue for them. I think it’s a good motto and could maybe go in the footer. • Should we first identify ourselves as simply as possible? So perhaps instead of identifying ourselves as Public Lab and various universities we could first say that “Where We Breath is asking” and we can put PL and universities into the footer, if we need to have that on the front page. At that point in their reading of the website I’m not sure that they will have an understanding of what WWB is and they will maybe be reticent to join what they don’t understand. • Maybe we could rephrase the research question to say “Where We Breathe is website where manufactured home residents can work with our research team towards answering the question: How does formaldehyde exposure affect the residents of manufactured homes?” I think the “how” is perhaps more engaging than the yes/no way that we are currently posing the question. • Use this site to share information about indoor air quality health issues and read others' stories.” • Should we mention the tools we bring to the problem? We cover the user group, and the problem but not our tools, which might be the last part of the triad.

— Reply to this email directly or view it on GitHub https://github.com/publiclab/wherewebreathe/issues/15#issuecomment-50373443 .

[shapironick]

Just picking this thread up.

after messing around for a few hours with various photos its very clear that the way the photo is set up in the current home page works really well compared to other option (i'm appending another option at the end but I don't think it works as well--don't know where the text would go).

Some changes that should probably be made:

--Lead line above the photo to state the problem: "Do you live in a manufactured home? Studies show that manufactured housing have elevated levels of formaldehyde| Learn More>>" --Research question and identity clarification: "Where We Breathe is website where manufactured home residents can work with our research team towards answering the question: How does formaldehyde exposure affect the residents of manufactured homes?" --Followed by same "Join Where We Breathe" button --In text bubble below photo "Use this site to share information about indoor air quality health issues and read others' stories." And then more elaboration on privacy etc once that is finalized. --In footer: "Sharing your story towards a better public understanding of indoor air quality health issues.

This is a Project of Public Lab; this website is open source software."

Here's another photo option that I think doesn't make room for the text

[jywarren]

Good comments, thanks -- yeah, the current photo is kind of perfectly laid out WRT dark/light and being behind the text. We could have a second image or alternate, and change where the text is displayed, but the current one works on both mobile and desktop, so it's kind of a perfect image. It'd be nice to show more vertically of the house itself, and I can play with that a bit.

The photo you've just posted is super also. They're all great. I think we'll be able to use them in other places on the site. Or we could eventually try out a slideshow to circulate them through, though I'm not sure it's worth the extra design effort.