Chrome extension vs. a standalone app

[msachi]

Following some conversations in the pub last night, I think that we should at least attempt to implement our app as a Chrome extension.

As Matteo has said in his pitch, there are many resources (CBT-based and others) that help people with low self-esteem. The main idea of our app is to focus on the immediacy of the experience of low self-esteem caused by browsing social media, and on real-time intervention. If we make it a standalone app we lose our main 'selling point', and I think it is unrealistic to expect young people to be proactively going into an external app to help them deal with their experience of social media.

I think technically it shouldn't be to difficult to do this, but I will research it today to get a sense of what's possible. This is an example guide that looks quite straight-forward: https://developer.chrome.com/extensions/getstarted

The only functional disadvantage that I see is that making it as a Chrome extension would probably limit it to a desktop app. This may not be ideal given that most adolescents probably spend more time on their mobiles than desktops (thoughts @MatteoTurco?)

[MatteoTurco]

This is a difficult one. I also think young people access social media through mobile phones more readily (though this is just a hunch - I'll have a look if I can find the relevant numbers), therefore it would make more sense to me to develop a tool that is accessible from that platform, as this is essentially what we mean by immediacy.  Also, a couple of questions re: the extensions option: a) do Chrome extensions operate in Chrome only? And if so, wouldn't that limit our range quite a bit?  b) given our proposed used of pop-ups, would the following be an issue:    "Can extensions open browser/page action popups without user interaction?No, popups can only be opened if the user clicks on the corresponding page or browser action. An extension cannot open it's popup programmatically. " Perhaps neither is an issue but I just wanted to check with you. In any case, I feel that focusing on developing what is technically feasible, rather than what has better scope of action, could be a useful guideline at this point.

[msachi]

Thanks for these thoughts, that's very helpful, and made me think about the issue more carefully.

I agree that mobile will be the preferred platform for adolescents (mobile internet use recently 'took over' desktop internet use, so I'd say it might be something like 60:40 for general population and more for adolescents).

However, I think we missed something: when people (and especially young people) use Facebook, Instagram etc. on their phones, they do so through the native apps (Facebook app, Messenger), not via their mobile web browser. I think it would be vary hard for us to interact with the native mobile apps in any useful way (Facebook knows why...), and so if we take the 'real time' route we're limited to the desktop browser. Secondly, I wonder - if we're focusing on the kind of activity on social media that make people feel low, it's probably endless scrolling through the news feed, rather than messaging, that is the culprit. And I suspect most of the endless scrolling is done on desktops.

To answer your questions: 1) Yes, it's Chrome only, but it's the most popular browser (74% users at the moment, and still rising) and this number is likely to be higher with young people who are tech savvy. I don't think that would be a major issue. 2) Very good point! It seems like pop-ups are indeed not allowed. That is a bigger problem I think - it means that the user needs to proactively open the extension to interact with it, and that has implications for what we want to achieve.

We'll discuss more tomorrow morning, but I think we have two options, both with their limitations: 1) Mobile web app that is not 'real time'. Users could be asked to do CBT exercises after their exposure to social media, or the app could ask them to get some exposure and work from there. 2) Desktop Chrome extension that works in 'real time' but that the user still needs to proactively open and engage with.

[iteles]

A note, if you do end up going with a Chrome extension, some good resources to get you started here: https://github.com/dwyl/learn-chrome-extensions

[marisid]

@MatteoTurco thanks for your input and thoughtful questions!

As @msachi pointed out, browser statistics show that Chrome is the most widely used browser by far (74%) and its popularity grows even more with time. Also, 33% of Chrome users had an extension installed in their browser (source). I couldn't find a more recent source regarding the use of extensions but if we are starting from the desktop version, building a browser extension on Chrome seems a reasonable approach for both feasibility and popularity reasons.

Actually, this development strategy has already been used for the reword project to tackle the cyberbullying problem. I'm quoting from their site:

We're currently piloting reword on Google Chrome only, but our immediate next steps are to roll it out to Firefox and Safari, so stay tuned! We also have plans to see this extended onto mobile devices.

The fact that we cannot use pop-ups, as originally envisaged, is a problem we need to address as it will definitely affect immediacy - so we should integrate with the user journey /stories.

We'll have a think about all the above and discuss later today. The good thing is that we have a great list of resources regarding building Chrome extensions - thanks a lot @iteles :smile:

[marisid]

@msachi and I had a brainstorming session this morning where we did a lot of research and asked for technical advice here at Founders & Coders. We think the best way forward at the moment is creating a Chrome extension for the following reasons:

• It allows us to interact with Facebook (or whichever social media account) on the browser. In contrast, on mobile, there is very limited interaction with the Facebook app (actually iOS allows no interaction across apps for security reasons).
• It allows us to access content on the Facebook page and maybe use it as input in the CBT process, i.e. turning the app to a personalised and customisable tool.
• It allows to embed content on the Facebook page (e.g. timer and SocialUp button) and so, there is no need for pop-up notifications.
• If we drop immediacy and build a mobile app (whether native or progressive web app), chances are that we need to build separate apps for Android and iOS, that would send notifications to the screen. The problem is that it would be quite difficult to connect these notifications with the concurrent use of Facebook, i.e. make the apps talk to each other in real time. In which case, there is no real-time connection between social media and CBT technique, which would be the main novelty of our app.

@MatteoTurco please let us know if you have any questions or concerns regarding the above.