ø

Hi! I'm Guillaume. I do stuff relating to startups, software engineering, HCI research, education, game design, amongst other things. www.twitter.com/gardaud www.gardaud.fr Emails at [email protected]

[gardaud-blog]

Better tools = better communities

Reddit is an online community where people share links, and others can vote on them. The most voted links bubble up to the top, and users get a point (named karma) for each vote. High voted links go away after a period of time, so content is always new. (This high level concept has been copied a lot in the recent years.)

The result is a meta-game where people go to the community as much for the news aspect as for the karma collecting. Users with high karma scores are respected, and posting a link that brings a lot of karma is seen as an achievement.

A result of that is that funny pictures, interesting links, or non-news content that has been highly voted on in the past gets resubmitted months later by different users (who frequently claim it as their own), in an attempt to get as many points (since it did well the first time).

Those "reposts" tend to be heavily disliked by the community, especially the more ancient users who want to see perpetually fresh content.

An interesting and recent phenomenon is that some users programmed free web services that, given a link, scour Reddit's archives (using image recognition algorithms, amongst other things) to determine if something is a repost or not. The implicit goal is then to expose the user who reposted it in the comments section, shaming her if she claimed it as her own. (one of the most populars of these tools is karmadecay.com)

At first, those tools were only used by savvy users; but they quickly became popular and now a repost rarely happens without the top voted comment being a public shaming of the poster for claiming the content his own. Some users have even programmed bots that automatically detect those reposts and post a comment to original links.

This is a great example of how a tool not envisioned by the original creators of Reddit enabled the community to better itself. Better tools make for better communities. If Reddit had deployed measures to prevent programmatically crawling its pages or block comment bots, the community would have never been able to do that. If you're building a community and want to achieve such phenomena, your best bet to maximize the odds of that happening is to be friendly to users who want to crawl your pages or automate certain things, and offer easy access to your past data.

[gardaud-blog]

That’s exactly the danger of what too much ambition can do: Narrow the range of acceptable outcomes to the ridiculous, and then make anything less seem like utter failure. It’s irrational, but so are most forms of psychological addiction. You can’t break the spell merely by throwing logic at it.

(from http://37signals.com/svn/posts/3547-ambition-can-be-poison)

[gardaud-blog]

Healthcare and Design, part 2

Just as I posted about design in healthcare last week, this popped up on Reddit:

As always with Reddit, the sources are cruelly lacking, but the 3rd image seems to indicate this is taking place at the AC Camargo Hospital, in São Paulo, Brazil.

Some Google-fu reveals that the project was led by the brazilian firm JWT. More details about this project, including a video, can be found here. (screenshot of the relevant page in case it ever goes down)

[gardaud-blog]

Designing the user experience of healthcare institutions

The vast majority of us don't visit healthcare institutions (hospitals, mental institutions, clinics, etc.) that often. The first thing you realize when you do is how uninviting they are. Stock photographs on the walls, antiseptic and alien looking equipment, drably painted walls– it is rarely a warm friendly experience.

In a study conducted by Children’s Hospital from 2006–2007, the staff found standard, non-decorated, rooms stressed children to the point that many needed to be sedated before they would lie still. In contrast, an underwater-themed room with life-sized mermaids and music reduced anxiety in families and children — and reduced sedation rates — the study found.

(source)

It seems that it would be in every patient's interests to make those places as inviting as possible. What can we do to improve the UX of healthcare institutions?

The only research on the intersection of healthcare and design that I'm familiar of (and I'm certain there's more) is by Doug Dietz, who did the work cited in the quote above.

Without an official budget or staff, Doug put together a team of people who cared deeply about the issue. The result? With some paint, scents, lights and a little imagination, the scan rooms were turned into adventures: in one room, there is an ocean and the scanner is a submarine. In another the scanner becomes a tent in a camping experience.

The ‘Adventure Series’ scanners have been hugely popular with kids, and the Hospitals and Doctors who are GE’s customers. Doug now trains other GE employees–both in the medical division and beyond–to use design thinking and innovation methods in their teams.”

(source)

His TEDx talk about his work can be watched here.

So how could we expand this to go beyond MRI machines for young children? This would be a complicated design task, as I'm sure there are many health safety rules and laws coming into play, but here are a few ideas:

Why should all waiting rooms be the same? Having several "themed" waiting rooms for different categories of patients could be interesting. Young children would be waiting in a room made to feel like being in a spaceship, whereas older people could be waiting in a room with leather chairs and exotic carpets.

Hospital hallways are terribly long, repetitive and uninviting. What about covering hallway walls with entire tiled displays (often seen in airports) playing looping videos of the sea, fields in the wind, etc.?

There's an interesting interplay of presence in such locations: some people are only there to visit a relative or friend, others are patients for mere hours or days, while others may stay there for months or years. What can be done with this?

[gardaud-blog]

Fixing the strained Police-Youth relationship with trading cards

I'm always a sucker for clever, playful solutions to problems pertaining to civics.

A friend of mine just pointed me to this: police officer trading cards, from the Metuchen Police Department (2011). The goal is to give a reason for kids to approach police officers, and foster a connection in the typically strained police officer-youth relationship. There were 30 cards total; the first to collect them all got a $20 gift certificate (but seeing how fun trading cards are in themselves, that's irrelevant, really).

The front of each card shows a photo of an officer. The back of the card contains bio information or a personal or safety message from the officer. The goal of the program is to reach as many children as possible with a message that enhances communication and trust. The program is being sponsored by a number of local businesses.

This initiative was done twice: once in 2001, once in 2011. This is a really, really clever solution to a problem that every single city experiences.

The Metuchen Police trading card program has been a home run for our department in terms of reaching the young people in our community. The officers are approached at the schools in our community, during public functions, and while on patrol. We have made it mandatory that each officer carry their trading cards while on duty.

During a recent holiday parade, the officers were astonished by the number of children requesting the cards. The children in our community have really embraced this program and are striving to complete their sets of police trading cards. This program has put our department in the spotlight with our community in a very positive way.

[gardaud-blog]

"Crunch with us before demoday" : screw you, the founders ARE the janitors

Look at this wonderful post that just graced HN's front page:

Of course, comments are disabled on "YC job ads" posts, which is mostly understandable; but in this case I just had to publicly comment on it.

Read this ad. No matter how cool they make it sound ("scrappy intern"), what they want is basically someone who will do work, for free, that the founders don't want to do (compensation isn't mentioned, and they insist on the "intern" nature of the role while over-selling the "learning experience", so it's safe to assume this is an unpaid position with no equity).

The founders want to play it cool while someone does what they find to be the boring shitty work for them. The position they're describing is "1/3 Executive Assistant, 1/3 Project Manager, 1/3 Product Manager". Do you want to guess how much executive assistants/project managers/product managers are paid in other tech companies? Hint: it's infinitely more than zero.

In many other industries, unpaid internships with no guarantee of a followup job are a common way to get free cheap disposable labor. It's a terrible practice that is good for companies in the short term, but sucks for interns and hurts everyone in the long term.

It's unsurprising that they use a throwaway GMail as their contact email, as any Silicon Valley company should be ashamed for partaking in such practices.

Silicon Valley has avoided the "free intern for coffee and photocopies" practice so far, with companies treating and compensating interns handsomely. Let's keep it that way.

PS: if you meet all the requirements for this ad, contact me (@gardaud). We're hiring for a similar position, except that we will pay you and it's a full time position that will last more than 3 months. Or just apply to pretty much any other company; this one isn't worth your time.