Archive for the ‘Uncategorized’ Category

Choosing a topic for your Bachelor, Master or Ph.D thesis can be hard. I’ve done two Masters and a Ph.D and here is are a few important things I learned on the way and that your professors are unlikely to tell you. The first most important thing to know is that …

It’s not about the topic

People speak a lot about research topics. But the topic is just a very small part of what makes for a good topic. It is really about what that topic allows you to do while you work on it and after you are done. So don’t just accept a topic your professor suggests to you. Try to understand what kind of work that topic implies and see if that’s the kind of work you’d like to do for the next few months (or years). If your professor doesn’t seem supportive, choose another supervisor before it’s too late and you are stuck with having to get up to a job that’s not even a job and which you hate. Doing a thesis is not only about having the title in the end. It can and should be much more.

Thinking about what you would most like to do while you write your thesis – and what you don’t want to do – will help you not to suffer most of the time while you work on it day after day. Even if you like in principle what you need to do for the thesis, you will still sometimes not feel like doing what needs to get done. But having chosen a topic which in principle needs you to do exactly what you want to do helps a lot.

Know what you want.

You would like to meet new interesting people from a specific walk of life ?

Choose a topic which is based on qualitative research and for which you need to do interviews. Say you’d like to meet actors. Develop a topic where they are your experts. Don’t feel imprisoned by your field. Any field in the humanities will allow you to work on what you are really interested in. You are in education science? Work on how actors acquire the ability to retain long texts. You are in economics? Work on collaboration and competition between actors. Don’t forget: it’s not about the topic, but about doing the kind of stuff you want to do.

You want your topic to prepare you for your job?

Develop a topic where you need to meet the people you want to become.

Figure out if the people you want to work for (or with) in the future would be interested in the topic too. To do so you may have to talk to some of them a little bit. If you find it hard to connect to them, your topic might be the door opener – unless, of course, you want to work on contract killers, hustlers, the American president or, maybe, consultants at McKinsey. You want to do communication consulting? Work on how communication consultants position their own brand. You want to work in a gallery? Work on the way galleries try to position their artists. You get the picture. 

You want to get profound knowledge on a certain topic or author ?

Pick a library-based research topic where you have to read what has been written about the subject or by (and on) the author. Whoever or whatever you want to learn about, there will be a topic that forces you to do so. Once you know who or what you burn for, you can also ask your professor to recommend possible topics to you. But if you choose people like Hegel, Marx, Keynes, Freud, Foucault or a subject like free will, power, public healthcare or war be careful to specify a very precise take on the topic. And be aware that if you work on these kinds of figures or topics, you need a high tolerance for frustration and an iron will to be exclusive. However much you work on your thesis, you will know that you have not included everything relevant.

You want to develop your own thing (or theory) ?

Pick a rather exotic topic on which very little has been published. You won’t be able to not read and quote anyone else. But you’ll be able (and forced) to creatively identify sources. There are basically three ways to address such a topic: theory transfer, theory application and bricolage.

Theory application consists in taking general, fundamental and often rather important theories and trying to apply them to the chosen topic. You might try to apply script theory from cognitive science to still images, actor-network theory from sociology to the relationship between hospital staff and doctors or between curators and galleries or apply iconology from art-history to advertisement. Whatever works!

Theory transfer consists in taking a theory which has been developed for one specific case and applying it to another. Noel Carroll wrote about how actors’ personal identities relate to the identities of their characters in movies. You might try to apply and adapt his theory to how, for example, the personal identities of university professors relate to the identities of the institutions where they teach or how nightclubs pick the DJs that perform on Saturday night.

Bricolage means to take bits and pieces from different theories and putting them together. To analyze how a single picture can tell a story, I used Paul Grice’s concept of implicature David Herman’s concept of story scripts and Jean Matter-Mandler’s research on how we spontaneously provide causal links in narrative and many more. To get there, you need to read a lot of theory and play around with it like you would with Lego to find out what fits together. You’ll also often get a chance to make your own pieces, so you can articulate two different theories. For all of these to be successful you need a certain lack of respect. You are building your own building where a part of another persons theory is simply a brick.

You should’t care what the role of the brick you choose had in the building of the other theorist. Don’t feel like you have to perfectly understand her building project, even less explain it in your thesis. It will make you and your readers lose track and patience. Just understand enough to see what place the brick can take in your building.

Know yourself

You easily get bored by a topic?

Maybe do a Bachelor, but don’t do a Ph.D. Well, or choose your topic wisely. Don’t let it have such things as the name of an author, a time period, a geographic location, a specific institution name in it. Let it be “systematic”. My own topic was perfect for me. It was about how to tell a story with one picture. Fortunately, people have tried to tell stories with pictures since pre-history. It allowed me to read narratologists, comics theorist, film theorists, philosophers, art-historians, even child book researchers and archeologists. 

Your are prone to easily get distracted and being all over the place and are scared it will kill your thesis?

I would say choose a subject that has the name of an author, a time period, a geographic location, a specific institution in it. But this is a tricky one. If you are easily distracted, you might also get bored by a topic that has one of these. It might then be better to choose a “systematic” topic – and to take the thesis as a chance to learn to control your distraction.

These things help always: Turn off all push notifications on your phone, stay off facebook, netflix, instagram. Go to the library. Seeing others work helps a lot. Disconnect from the internet if you don’t absolutely need it. But don’t waste away your day smoking cigarettes and going for coffee with your friends. If it’s to hard go to another library. Only take a break when you really cannot focus anymore. I didn’t tell you it would be easy.

You are creative?

Choose a subject where you can develop your own thing.

You are not creative?

Choose a clearly delimited topic. For example one that has the name of an author, a short time period, a geographic location, a specific institution in it.

That’s all for today. Don’t hesitate to criticize, ask questions in the comments or tell me about your own experience.

Oh, and that’s me, academically speaking:


Abu Ghraib

May 15 2008, this was the cover of The Economist. Pictures like this one taken by American soldiers in Abu Ghraib prison had been exposed and were taken to show that America used torture. But there was a whole series of pictures. Why did this one make it to the cover of The Economist – and later become an icon of anti-torture-activism?

Capture d’écran 2015-05-22 à 17.28.27

Why, for example, didn’t this picture make it? Wouldn’t it have been better? After all it wasn’t pixelated…

I think that there are several reasons for this.

First, the first picture is isolated, it is more iconic than a picture with more contextual information (a soldier looking at his photos). Second, the pose, with the arms held higher, is more Christianic, and thus stronger. But I believe the major reason is just the presence of pixels. It makes the picture look more authentic. The better the picture quality, the less the chances that a picture is non-professional, and the higher the risk of manipulation. If there are pixels, at least in 2008, then there is a chance that the picture is authentic. The picture had quite a career, both in politics and in arts. And below you see a collection of works related to this.

Capture d’écran 2015-05-22 à 17.28.35 Capture d’écran 2015-05-22 à 17.28.48 Capture d’écran 2015-05-22 à 17.28.59

Capture d’écran 2015-05-22 à 17.29.18 Capture d’écran 2015-05-22 à 17.29.29

But apart from the success in itself, I’m interested in what happened with the picture. It became more and more simple – and thus, I believe, more general and symbolic. Whereas the photograph with the soldier still connotes one specific moment and prisoner, the abstract picture can evoke every prisoner being tortured. Richard Serra’s drawing (the last on the bottom right above and in the center right in the right image below and the second in the center on the left), is particular insofar as he brings back expressivity to the stripped down symbolic image. In order to account for the differences between these pictures, we can, I believe, distinguish different degrees of specificity, following a chain developed by Scott McCloud in “Understanding Comics. The Invisible Art”. McCloud draws different pictures, in order to show how pictures differ in their degree of generality. His system  (see the three last pictures below) perfectly matches the Abu Ghraib imagery.

Capture d’écran 2015-05-22 à 17.29.47 Capture d’écran 2015-05-22 à 17.29.55 Capture d’écran 2015-05-22 à 17.30.02

Capture d’écran 2015-05-22 à 17.29.38



ImageWorldcloud powered by

Posted: July 9, 2011 in Theorizing, Uncategorized

Why I stopped writing by hand

Posted: April 23, 2011 in Uncategorized

I used to carry a notebook with me when I was at high-school. It used to be important what it looked like. It used to be from Clairefontaine, a traditional French brand. About 5 years ago, I stopped, only occasionally going back to it. Having moved to France, there was no more point in buying Clairefontaine any more. I now bought Muji, because Muji is like no brand. Why do I feel the urge to tell you what brands my notebooks where and are? Because there is some nostalgia involved. And some vanity. And some Superstition. I used to be proud and happy to use a Clairefontaine notebook while I was living in Germany. There was only one shop in my town that sold them.
France had been the country where I had discovered art. Using Clairefontaine, my own writing was affiliated with France and its culture. The famous books by Moleskine are riding on the nostalgia/superstition/affiliation wave very explicitly (I have a hunch there might always be a bit of superstition in nostalgia). And the reason why I never got one – apart from the exaggerated price – is because I want my writing to be mine, personal, actual. It is bad taste to make some things explicit. Informing potential customers that Hemingway used your notebooks is bad taste. Either the books have intrinsic value, or they don’t. Telling me that Hemingway used them is trying to artificially bestow a spirit on them. It’s, basically, for people that don’t have any imagination.
As for me, I eventually stopped writing by hand altogether because I had notebooks. Notebooks full of notes that I never used, never even looked at again. Notes that were virtually there, but practically lost. I have so many ideas, that I couldn’t see myself taking time to copy notes from my notebooks to rework them, make sense of them, develop the thoughts I planted in them.
When I found out that there was a folda

ble real-size keyboard and a Word to Go application, I started using a Palm Handheld, the cheapest Palm on the market at the time, a Palm Zire 21. I still believe that Palm missed a tremendous marketing opportunity with students. A Palm could have been what the Asus EEE PC became, the first netbook on the m

arket. They could have marketed their product to students all over the world, who wanted to have their classnotes in digital form, but found their notebooks to heavy to carry. But crowdsourcing, intelligent or other, and corporate co-creation portals didn’t exist yet – at least Palm didn’t have one – and so I couldn’t tell Palm about my idea.

In any case, the Palm would synchronize with my computer and I wouldn’t loose any of my notes any more. Did I look at the notes I had made on  the go more frequently? No. But I felt that I could. I felt that they weren’t lost in some book. I felt that with a few clicks, what I had written could circulate, be shared with anyone anywhere in the world and be published. Writing something searchable and transmittable made it worth the while to write at all. Any thought written on my Palm could enter circulation by a simple click. My feeling has hardly changed since then. I feel that to simply exist now, writing has to exist in digital form.
The only thing I use paper notebooks for is to make lists, to jot down one sentence ideas – and even these I prefer to mail to myself with my smartphone whenever possible. But reordering your thoughts by reordering the concepts you work with like what you see on the photo I took when writing an article about documents, still isn’t very easy digitally, even with mindmapping applications. But that’s just a practical frontier.

A few days ago, I talked to a friend who told me last year she had had a handwritten correspondence with a man she was in love with and who had a girlfriend. They had written pages and pages, some of her letters were 15 pages long. I was very impressed by this story. I felt like my friend had just revealed me that she was of a different species, I understood that she inhabited a different world than me. Even as gestures, penpals don’t exist in my world any more. My sole use of letters has become administrative. If I’m in love with a girl and she’s not in my town I try to catch up with her on skype – or wait for her to connect to facebook. Does this seem prosaic? It probably is. But I would find sending letters too nostalgic, and, therefore, somehow ridiculous.
I understood that for my friend and her man, it had been a way to show to each other that they inhabited the same world, a different world, beyond the world they shared with their official boy-friend/girl-friend and with most other people, people like me.
The fact that they wrote to each other, and the fact the letters were lost forever when his girlfriend eventually discovered and burned them all, ads to the appeal of their correspondence. Imagine they had written emails to each other. Do you really think the girl would have gotten any satisfaction out of the operation: “search XYZ/select all/delete”? I don’t think so. She probably wouldn’t even have done it.
Literature, Nelson Goodman says, is allographic art, it can be reproduced, it’s not about unicity or about the touch of the brush. Painting is autographic. And so is epistolary writing. There’s a deep valley separating handwritten and computer-written text. But more than the fact that the paper on which a letter is written has actually been touched by the person who wrote it, it’s the fact that a letter is unique, that it can’t be shared easily, which makes it special – and annoys me.

This is somewhat connected to my earlier post about naming and linking. But I want to make a different point about the use of ordinary expressions on the web. Contextuality (the situation in which an expression is being used and who uses it) is an important issue in language in and outside the web. As it has sometimes been noticed, the sense of words on the web is different from their ordinary sense. Their scope, the domain of objects they refer to, is usually more narrow than in ordinary life. In the course of browsing the internet we get used to such reductions: browsing refers not to lazily shopping, but to a certain web-based activity.  Whenever we read “twitter” on the internet, we think of the website rather than this.

“Connection” in ordinary life is a much larger expression than on social networks like hypios, Xing or LinkedIn. In ordinary talk there is “a connection between New York and Paris”, “a connection between the expressions ‘art’ and ‘artificial’ etc.”. On hypios there are only “Connections between Solvers”. The scope of the expression “connection” on hypios is more narrow than in ordinary talk.

But sometimes it’s the opposite: on facebook, “friend” has a larger scope than in ordinary life. Most of my facebook-friends are not what I would call “friends” when speaking to a real-world connection. Many of my friends now use expressions like “one of my facebook friends”. But they might end up getting tired of that. If facebook is as successful as they want it to be, language will evolve. In a few years, the web-sense of “friend” might be the prime one – at least among younger people.

It will have been a long way from Aristotle’s idea that “friends have to live under the same roof” to the hundreds of new friends we connect to every year.

(N.B. An earlier version of this post was published on – the now defunct? –

Everyone who designs websites or theorizes the internet should think about linking at some point. My main subject here is transparency of links.

The phenomenon of expressions with different senses pointing to one reference has never been so important and omnipresent as on the internet today.Each webpage with its unique url can be called a reference. But there are millions of ways in which this reference can be given, and url shorteners like or are tools to generate infinitely many new links, different ways of giving the same reference.

(One can also name a link whatever one wants as in a blog post, when you hover over a link. It can be described any number of ways as when instead of showing a url when you hover the description can say, “link to hypios blog”).

So, while links seem simple enough because we are used to them, linking actually involves us in a complex phenomenon.

The basic issue: How transparent are links?

The basic issue about links is transparency. Links may be more or less transparent. People may try to induce you into thinking that a link points towards the website of, say, your bank and then direct (or redirect) you to a quite different page. Links may try to attract you to a page that you presumably don’t want to go to, like a harmless music video by Rick Astley. And then there’s of course the harmful ones, directing you to pages that damage your computer, or attempt to retrieve confidential information (aka: pishing). Most non-transparent links simply want to get you to buy something or attract your attention. Many of the comments on blogs come with this kind of links (we filter the most obvious ones). It is because transparency is a security issue that and wordpress now include previews of the page a link directs you to. But I’m not going further in the security issue here. What I’m interested in is the connection of transparency and user friendliness.

How transparent should links be?

The fact that a link isn’t clear is not always a sign of bad intentions. It may sometimes be hard to find the most transparent name for a link. On hypios, we used to have a Marketplace tab, which has become the Problems tab: we realized that on a site for problem-solving the former tab (and a tab is a form of link) was much less transparent than the latter.

In fact, initially it was Steve Krug’s Don’t make me think that got me thinking about link-clarity. And I immediately realized that Marketplace wasn’t clear for anyone who had never used hypios before, while Problems would be. Marketplace forces users to think what it could mean while Problems doesn’t.

Where you link from is as important as where you link to

Krug thinks that a link should always have the same name as the page it directs you to.

This is a principle which we don’t always apply in our webdesign, simply because there is more to a link than the page it leads you to: there is also the page where it gets you from.

On twitter, hypios tends to use a twitter specific, usually more catchy, description of our blogposts rather than simply using the post’s name.

When you complete a text with links (as I do it here) and don’t want to interrupt the flow of your writing, you sometimes prefer to link non-transparently, as above when I linked Rick Astley to the wikipedia article concerning the phenomenon of Rickrolling rather than to Astley’s biography. Why? His biography is not interesting for this article, while the phenomenon of Rickrolling is. What this example also shows is that transparency comes in degrees. Linking Rick Astley to the Rickrolling article wasn’t as murky as linking ambiguous link to the music video.

Why the url is not transparent

There are different ways of classifying ambiguities: the most basic way, but surely not the best, is according to the hierarchy website – webpage. If the sense of a link is very ambiguous, it doesn’t transparently sort out which website you’re on your way to, if it is mildly ambiguous, it does not transparently link to the particular page. But I would welcome other classification suggestions.

It might seem like giving the url is the most transparent way of linking. But this is only true in principle (it’s true for computers not for human-users). In many cases, the url is determined at the backend according to a system which has more to do with the site administration than with its usability. This is most prominently the case on youtube. Or did you know what “” was before clicking (go ahead, we promise you won’t get rick roll’d)? Telling you that it is a documentary about how the first problems were solved on hypios would have been more transparent, right?

When the same tab should lead to different pages (users have a history)

Another of Krug’s rules for webdesign is consistency: a link with the same sense (Frege) on your website should always lead to the same page. This implies, for example, that “Connections”, wherever it appears on hypios, should always lead to your list of connections. Well, as it now stands, it doesn’t.

Here’s what happens with all but the Problems tab: they lead to a description of each feature for the people who have not signed up yet while those who have become Solvers can access the function that the tab refers to. In this case, hypios uses a very basic contextual piece of information about a user to determine where the tabs will lead him to. Most of the people who aren’t logged in on hypios are still discovering the website and need to get more information on what hypios will offer them once they decide to sign-up. We believe that rather than confusing our users, this increases hypios’ usabilty.

Further, having the tabs on the site even before you sign up gives you a feel for the environment you will enter once you do join the Solver community.