Like many of you, I started to receive Facebook notices on Saturday telling me my friends in Nepal were “marked safe” on Facebook’s Nepal Earthquake Safety Check feature. Launched on the 3rd anniversary of the Japanese Tōhoku earthquake and tsunami, Safety Check is designed to help Facebook users know who of their friends is accounted for in a natural disaster.
At first glance, Safety Check seems to be a great feature. By alerting others to someone’s post-disaster status, Facebook users can speed up the process of response and recovery. While we often focus on the physical disruption in a disaster, there is an equal or larger emotional disruption. Knowing who is safe, and alleviating worry, is a great benefit for everyone.
However, I have a major issue with Facebook managing the Safety Check feature.
Who decides which disasters warrant a Safety Check?
The Nepal Earthquake is an easy example of when to use Safety Check, but will Facebook activate Safety Check for Chile’s Calbuco volcano? What about another Westgate Mall-type terrorist attack? Is Safety Check only for “natural” disasters? And how “big” is big enough to qualify?
There is a very slippery moral slope in determining what is a disaster, especially from the safe confines of Silicon Valley. I don’t feel comfortable leaving it up to Facebook to decide which disasters are worthy of social media support or not. I would much rather see the Safety Check feature managed or at least influenced by local Red Cross or Red Crescent organizations and government emergency response agencies.
This my opinion – what’s yours?
Do you agree that it should be integrated with local humanitarian response organizations? Should Safety Check have been activated faster than 4 hours after the Nepal earthquake? Did it even help you check on your friends? Jump over to the comments and let us know your thoughts.
Good point Wayan! Thinking this may be the unthought-of consequences of technology. It may have started as a tech for good activity but its side effect is worth to consider. Corporate ruling what is worth to consider a disaster is as negative as a internet/information filter bubble. I think your suggestion is right that local Red Cross or Red Crescent organizations and government emergency response agencies or UNOCHA manage this.
Furthermore this actually means these technologies have proven that it has evolved and become more significant in our lives and in the field of ICT4D and humanitarian aid.
See for example ICT4D 1.0 towards ICT4D 2.0 thinking (Heeks, 2009) http://www.oecd.org/ict/4d/43602651.pdf , although one could argue this is mainly ICT4D 1.0 thinking.
It does seem slightly hubristic that FB launched a service like this without integration with Red Cross, OCHA, or local governments. If you or I were designing it, I would expect that connecting the service to local responders would be an integral part of the process. At the least, we would incorporate Jessica’s suggestion of an access issues reminder.
Wayan, I have a number of issues with this post and its sentiments…
1) Yes. It did help me know that my Nepalese friend was safe *and* I would have loved this feature to have existed during the aftermath of the Westgate attack in Nairobi to help me account for all of my friends’ safety. Instead, I spent hours glued to my facebook scrolling through my feed waiting for status updates to appear. The “dark side” for me would be the fear caused when you don’t see a safety check due to infrastructure damage and lack of access. If anything, FB could add a reminder that in disasters access to the internet can be interrupted and therefore not to panic if all of your loved ones do not immediately check in.
2) As I’ve mentioned to you over many beers, I think one of your greatest assets as a catalyst for the ICT4D community is that you are not afraid to ask the provocative questions and present a controversial viewpoint. As I have also said to you, I feel sometimes you do so in ways that can distort the truth. You raise some good questions here, but nowhere do you provide information about how Facebook makes the decision about launching Safety Check. Surely they have a policy on this that would be good to reference, or at the very least write your questions in a way that ask for transparency rather than vilifying Facebook based on an assumption that it is some arbitrary decision being made out of Silicon Valley. As a widely read blogger with influential opinions, you owe it to your readers and to those of whom you write to do your research — blogging with influence requires some level of journalistic rigor.
3) It’s a learning process. You more than anyone knows that the way we apply technology the first time is seldom right. There were HUGE issues with the deployment of Ushahidi and the 4626 initiative that used SMS and social media to identify and map needs following the 2010 Haiti earthquake in which I took part. There were zero response mechanisms and the mapping relied on over-caffeinated stressed out grad students to prioritize thousands of unverified accounts and place them on a map of an island where most of us had never been…yet we were celebrated for our innovative work. While I wish the world had been more critical at that point and driven faster iteration to an ethical scalable application of technology, to me, Safety Check is part of that evolution…and evolve I am sure it will continue to do. Raise the questions. Push FB to answer them, but also tell the whole story and how this is a tremendous tool even if there may (or may not) be some kinks to work out.
Always your friend, sometimes your critic, Jess.
Jessica, that’s my exact point – we do not know how FB decides what is Safety Check worthy. The service itself is great. I’d like to see it managed transparently.
Glad to see this article as I was just talking about this yesterday with friends.
There are many interesting points about this new FB feature. I have friends in the affected zone and the first place I went on Saturday morning to find information was Facebook. The first place I went to find general on-the-ground realtime news was Twitter. I was told that Snapchat even got into the crisis tracking game this time around. Let’s take a moment to reflect on the role of social media tools being used to track, report and map. This is a far cry from posting selfles or our money coffee routine. This is an amazing evolution.
The most interesting part of this is that FB was able to do something that our aid agencies and governments do not have the ability to do. This is a new time where silos of responsibility and authority are crumbling, where tech companies are filling gaps and where institutions that are failing us may well be on there way out. This is an unbelievable time of change bringing forth new technology tools.
Of course as history as shown, powerful tools can be extremely destructive in the wrong hands. This did not escape me on Saturday and I’m sure this will prompt further conversation.
So good for Facebook to create a useful app for us in a time of major crisis. I can tell you who chooses what is deemed a crisis, we do by how many of us were on FB chatting about the event. I will look forward to hearing the story of how FB rolled this out in the days ahead with a large dose of criticism I’m sure. I hope too, we will all pass along the gratitude we felt when discovering that our loved ones were safe. I hope we all spend time shining the light on what we want to bring forth in the world and give it the airtime it deserves.
Lastly, I’m glad to see the topic of transparency come up in this context because I’ve always wanted more transparency from the ICTworks site as to who and and how editorial content is chosen. It’s always a good idea to take our own advice. Thanks for the opportunity to share my humble opinion.
Jessica,
Now that I have more time to respond, I’d like to thank you for your honest critique. As a fellow critic (in the classic, positive sense of that word), I highly value criticism.
First, let’s separate the idea from the implementation. I LOVE the idea of Safety Check. People already use FB to check in, already look for news from loved ones there, and combining the two into a specific disaster response service is brilliant. I am sure we all see that.
However, who decides what is worthy of Safety Check is a critical issue. The Westgate attack is a great case it in point. So far, the only guidance we have from FB is that Safety Check is for “natural disasters” so sadly, Westgate would not quality, and neither did the Garissa University College attack, which happened after this service was launched and in a high-FB usage country.
So this is my public letter to FB to have them improve their service. Publishing more about what qualifies as Safety Check worthy would be a good start. Next would be integrating with local actors. I am sure you and I could come up with a list of other improvements to what we both can agree is a good start.
Thanks for your thoughts Wayan and lets debate that ideal feature list over our next beer! I’ll give it some thought but my feature list would be pretty short. One thing I like about Safety Check is its simplicity. As you say, it builds on what people already use the platform for and makes it more efficient. Adding needs reporting or more complicated features would undoubtedly open up more complex ethical and operational issues that are well outside FB’s core competencies… and if they didn’t have the partnerships to support it, then we’d all rightfully be up in arms!
Wayan, I agree with your take and the need to decentralize such decisions as much as possible. Often times, we’re not in a good position to understand the situation on the ground. Additionally, there’s the issue of slow onset emergencies, such at the Horn of Africa crisis. In this case, would Facebook have required users to update their Safety Check status on a daily/weekly/monthly basis?
The upside here is that I don’t expect this one service to do more than help propel the discussion. The trend is moving away from the walled garden of Facebook – toward more specialized applications. They’ve started the discussion, however, it’s not hard to imagine a competing RCRC Movement creating it’s own solution down the line.
I don’t see “the dark side” here with FB’s Safety Check feature. As you point out, the counterfactual here is that without the feature it may take greater time for people affected to locate and communicate with each other. And to your point about who decides, its FB’s platform so it’s their decision take a lead in developing a social media tool (on FB)… unless you are arguing for some sort eminent domain activity here?? The only downside I can see so far with the Safety Check feature is that it is one of a few (Google People FInder; the Red Cross’ own missing persons registry) and so people may be missing each other by utilizing different services. So a unified service may increase contact between people but that’s isn’t a big enough reason in my view for Facebook not to play an active role to help increase communication.
John,
I’m arguing that while the idea is a good one, the way FB is implementing it – alone, and without transparency – comes across as another “Silicon Valley Can Save the Day” siloed solution. As you point out, where is the consultation with actual disaster response organizations?
Safety Check has great value. I think it could have so much more.
FaceBook’s foray into online disaster response is an example of a good idea with shortfalls in its executed. Bad ideas badly executed quickly die. Good ideas badly executed have to recover from initial damage.
FaceBook is a service provider that provides free services in exchange for varying degrees of user data mining and user exposure to advertising. Safety Check is offered as a public service but, as has been pointed out, it has not had consultation with, and does not work in collaboration with, the disaster preparedness and response community. This leaves the service with a lot of unanswered questions (some listed in this discussion) and denies it improvement based on the lessons learned from the relevant stakeholder communities.
This hampers capacity building both within the service itself, and across the disaster preparedness and response community. It leaves unaddressed the role of data mining in its execution. It leaves a public service at risk depending on FaceBook’s business strategy and future revenues. A lack of collaboration reduces the ability of others to expand the features of the service, or pick it up should FaceBook stumble. After all, while an individual’s presence on the Internet may live on after death, both the focus and lifespan of Internet companies can be highly variable. The demand for disaster preparedness and relief is permanent.
The article and many of the comments are nothing but speculation and conjecture. I agree with the comment that suggested some research before a post like this.
Sam: Based on what information are you authoritatively stating that they did not consult or collaborate with the disaster response community? As someone *in* the disaster response community, I would like to respectfully state that your statement is factually incorrect.
You are welcome opinions on whether you believe the execution is good, or if it is adequately transparent…great points, and absolutely worthy of debate. But by the same token, opinions should not be portrayed as facts.
@Wayan: There is an official UN definition of a disaster, by the way.
“A serious disruption of the functioning of a community or a society involving widespread human, material, economic or environmental losses and impacts, which exceeds the ability of the affected community or society to cope using its own resources.”
— http://www.unisdr.org/we/inform/terminology
The last part of that is very important, and it may answer some of your questions.
Steve, I am very familiar with the official definition of disaster. What we don’t know is if Facebook is using that definition as well, and at what point do they decide that a disaster is widespread enough to warrant Safety Check. For example, why no Safety Check for Chile’s Calbuco volcano, which is displacing thousands these days too?
@Steve, If you are correct that I am factually incorrect about the extent to which this initiative has consulted with, and been integrated with, the disaster response strategies of the rest of the disaster preparedness community, please direct me to references. It is easier for you to prove that what is there is there, when that is the case, than it is for me to prove what is not there is not there. I would be more than happy to stand corrected on that point, and note that much of the rest of my earlier posting goes beyond that point.
I’ll take a walled garden like Facebook, flaws though it may have, over individuals or agencies deciding the time of greatest need is also the best time to plant a new one.
Case in point: when the Haiti earthquake struck, families, diaspora, and international VTCs spent the time waiting for the national telecommunications infrastructure to come back online by capturing the voices and needs of those trying to connect with loved ones in country. By the time the Google Crisis Response Team launched their own, it was the *seventh* one to come online, and you can imagine after 48 hours the duplicity and misinformation as one source didn’t always confirm the next and which was most current. Facebook and Google represent not simply the maturity of lessons learned and an ever-expanding access to mobiles and Internet but also the desire path of users ad affected populations: self-reporting in a familiar channel with a secondary channel for escalating if (unfortunately) necessary.
Here is where the perspective of humanitarians impedes the needs of families. Is Facebook’s activation protocol transparent? Is it integrated with major emergency agencies established in country or the UN Cluster System? For most, in those first critical hours the correct answer is who gives a shit? Facebook users don’t, families don’t, friends don’t. I didn’t. I wanted to know if my Nepalese friends and colleagues were alive. Social aggregation to further inform WatSan sector distribution priorities comes later this week.
Bottom line is this: As long as Facebook doesn’t now assume it knows best about the distribution of medical supplies, we’re all good. I’d say the “dark side” in all this is not Facebook’s lack of transparency as long as it’s limited to answering the immediate question of “Are you OK?” and then stepping aside. The dark side is the humanitarian community questioning or impeding the value of a service that exists outside our stated rules of engagement.
Agreed that this is serving a need. My open question is still around when we can expect this service.
Is it only when its a big earthquake? If so, what makes it “big enough” to count? When can FB users (and humanitarian response agencies) expect to see this activated? Can users or responders request this feature to be activated?
I love the service. I don’t love being at the whim of FB to see it activated.
Do we need to love the service? Or even really like it? That’s one of the nice things about Facebook being a private company—if its solution doesn’t meet expectations, there’s an opening in the market to improve upon the concept (and the potential to monetize the solution, which provides the incentive). The idea is out there and is being talked about. I wouldn’t be surprised to see similar solutions coming online in the near future.
I love the comments here, and it’s a great discussion. Many above have said opinions better things than I can, so here’s my contribution: why not email this page to someone at Facebook who is familiar with the team? Here’s the previous product manager that came up on a very brief search via LinkedIn/Google https://www.linkedin.com/in/sharonzeng.
Based on her profile, she may not be the current PM, but I bet you she knows who is. AND, I bet that she and the team has probably gone through some of these discussion. Working with partners is time consuming but obviously crucial. It is important to remember that in the minimum viable product obsessed world that FB deals with, discussion like this do happen and tough decisions are made to include or not include features and aspects.
All of that said, FB has a history of ignoring the fringe cases too much, as does the entire technology community. In addition to making this an open letter (which is an awesome, awesome thing), let’s invite the people who make the products in and reach out directly. They may not have the background and understanding that regularly ICTWorks readers have, but they do care about improving their product.
Hey Tucker, that’s a great idea. I just forwarded it over via LinkedIn; we’ll see if anything comes of it.
Thanks Tucker. By the way, I do consider this an “open letter” or as Jessica called it, a “public letter”. Can’t get much more open or public than a blog post with comments enabled.
I would posit that a more public process in deciding what is or isn’t a Safety Check worthy event isn’t a “fringe case” but should be a core component of FB’s interactions in humanitarian situations.
Just because it’s public with comments doesn’t mean that anyone will see it, pay attention to it, think about, and respond (plus I’m sure there are some legal department hurdles), but kudos to you all for reaching out. That’s all you can do, but it’s an extra step beyond just a public letter. Tweet at the individual PM or team as best you can understand. People do listen, we just are all overwhelmed with information every day. Acting on feedback is a different matter.
And no regular reader here would say that transparency on decision criteria for such a public service should be a “fringe-case,” but that doesn’t mean that it wasn’t viewed that way. I don’t know, and it would be helpful to get the response from the team that actually built it.
Also, full disclosure, this idea is really Eric Reiss’ from the Lean Startup. In his case, he uses the tactic as a defensive move to see whether a large company would actually implement a key feature that would kill a startup. It doesn’t mean that they listen or then act on it, but it’s the full circle needed for sites like ICTWorks.
What exact duty does Facebook have to listen to concerns about how it’s implemented? None. Facebook is a shareholder owned company using it’s own money to add a new service. It’s not taxpayer or donor funded in anyway. Facebook can choose freely to make it (or not), listen to ideas/comments (or not) and we can choose freely if we feel like using it (or not).
The Red Cross on the other hand rely on donated money and taxpayer money and declare what happened to the money a trade secret: http://www.bloombergview.com/articles/2014-06-30/red-cross-needs-to-tell-all – now that we should surely have a much bigger problem with.
What is happening on a systems level is that we are surrounded by more and more closed walled garden services: it’s our choice as consumers if we adopt them ( e.g. Skype / Whatsapp ) instead of open interoperable standards that (like the phone networks, emergency broadcasts over phone networks, and SIP internet phones). Who decides what’s in the closed walled service is the one who owns the garden.
And us users of those gardens can advocate for the walls to come down. Yes, even FB has to listen to its users (though, agreed, it rarely does).
Love that Facebook is offering this service, agree it would be helpful to learn more about when and how it works. For example, how is Facebook thinking about potential unanticipated personal safety/security outcomes as a result of its users checking in with Security Check? It may be for this reason that the service is only available in response to natural disasters, and not situations like Westgate Mall or Garissa College University attacks. Still, natural disasters do (frequently) strike in areas where there is either active conflict or underlying tensions that could spark into violence. If it hasn’t already done so, Facebook could begin to mitigate this risk by asking its users to consider whether “checking in” with Security Check may pose any threats to users’ safety and security.
Hi Adele – I’m curious about your last point (“asking its users to consider whether “checking in” with Security Check…”) — how would you propose doing this? One could say that question is already asked implicitly in any one’s decision whether or not to use the feature. Or that it is already a feature of Facebook’s overall Terms of Service. Are you thinking otherwise?
Hi, John! Having never used Security Check I’m in the dark about what the current user experience is like. So here’s my purely theoretical and back of the envelope response. This could be as simple as incorporating some sort of red icon close to whatever “submit” or “check in” button currently exists. That button could be called “Smart Security” or “Security First”. When users hover over it or click on it short text could be displayed along the lines of “Think before you click. Could checking in put you or people close to you at risk?” If it’s not already an option, ideally Security Check would give individuals the option to choose the settings for who can see their check-in. Options (multiple choice) might include: friends, public, only certain individuals, emergency response workers. One can imagine a further step wherein users have the capability to share what their needs are: food, water, shelter, medical care, etc. It’s an exciting world of possibilities that big tech players with global reach like Facebook can help bring to fruition. The challenge, as always, is coordination with relevant local, national, and international actors to make sure that needs shared can be met. This was a major challenge with the Ushahidi deployment in the Haiti quake response, and the formalization of the Digital Humanitarian Network provides one model that shows how the international humanitarian sector can partner with outside actors to work toward making sure that needs identified can be met.
Default setting is friends only. As in I only saw my FB friends who FB thought were in Kathmandu and the nearby areas.
This also brings forth another issue – what is the “affected area” according to FB? They asking OCHA or randomly assigning it?
Yeah, I’ve been looking up statistics and stuff about it and it seems for Nepal to be a bit pointless, like 13% of the country has internet connection (prior to earthquake which disrupted smaller internet companies) and even after that only 17% of those with internet access use social media at least once a month. It leaves us with 2.2 of population (around 600000 which is still a lot but. Because of this I can’t help thinking that Facebook are just capitalising on a disaster, it would be like if coke offered coke to all those injured in a flood with no access to water but made them walk up a hill to get it.
I don’t think it’s right to just be able to mark a safety check, it seems that with your relationship with others it’s a dilemma, you are either close enough to inform them through an email, or Facebook message, or not close enough. Wondering whether someone is or isn’t dead shouldn’t be as simple as a Facebook option. Maybe it’s a time thing everybody only has, say 20 minutes on a computer so they all log on to mark as safe quickly, but then surely it’s quicker to just hand in your name and people to do that for you, which has been the case for years, charts and databases.
Twitter did the same thing in their anniversary video, they showed all the world events they took part in, such as the Hudson air disaster, and the japan tsunami, the tone it gave was annoying and with the constant hashtags of japan in the bottom left of the screen it seemed disrespectful. I can’t shake the feeling Facebook is now marketing disasters as opportunities for togetherness, brought to you by Facebook and that’s wrong to utilise fear so as to use your product.
I think it’s good though the donations that they allow through the website to aid.