About my blog

I am multithreaded, and sometimes the threads get tangled.

tagline


a Studiolab production.

FIY

food design course at TUDelft
I blog in Italian
I have a LinkedIn profile
my lab is Studiolab!
I love Jasper van Kuijk's product usability weblog

the wind I am enjoying


Disclaimer

De meningen ge-uit door medewerkers en studenten van de TU Delft en de commentaren die zijn gegeven reflecteren niet perse de mening(en) van de TU Delft. De TU Delft is dan ook niet verantwoordelijk voor de inhoud van hetgeen op de TU Delft weblogs zichtbaar is. Wel vindt de TU Delft het belangrijk - en ook waarde toevoegend - dat medewerkers en studenten op deze, door de TU Delft gefaciliteerde, omgeving hun mening kunnen geven.

Walter Aprile's pochedespès

my professional blog at TUDelft

Intel Design Expo 2009: 2 ITD projects go to San Francisco

 

We have taken 2 ITD projects, TATE and KeyPing, to the Intel Design Expo, an event associated with the Intel Developer Forum 2009. The event takes place in San Francisco, but you would not be able to tell, since the Intel Design Expo took place in the bowels of the Marriott Hotel: a vaste expanse of carpet where we, and other design schools, installed our cool stuff. The show was organized and run by Joy Mountford, whose long and fruitful career in designing interfaces and interaction is too long to summarize here.

 

The event has been covered in the Dutch Cowboys blog. There is an Intel news release.  And an Intel video about the whole event:

YouTube Preview Image

My Ars Electronica remarks

Thanks to the Interactive Environments minor, I have spent a remarkably intense weekend in Linz, visiting the Ars Electronica event.

From Ars Electronica 2009 2009-09-10-Linz

 

Find some photos of things I liked on my Picasa space.

Lampan92 gets closer to reality

In a previous post, I had spouted endless technotrivia on how to connect an Arduino to a FoxBoard Linux SBC. As one German former student of mine, you could ask "but vy?". This is the why: building the world biggest networked computer controlled IKEA hack lamp. Soon to be installed in the central hall of the Industrial Design building at TUDelft.

The lamp is an idea of Daniel Saakes, brought into physical reality by Aadjan van der Helm, Aditya Pawar, Sterre van der Helm and yours truly, with the usual fundamental collaboration of Rob Luxen. This lamp is made up of 92 individual IKEA lamps, wired as 12 groups (corresponding to the faces of a dodecahedron), plus two more groups for additional lamps placed on the vertices and one group that contains one, solitary, bulb on the inside of the lamp. You can consider it a baroque evolution of the original Big lamps from IKEA Lampan lamps idea.

Here are some photos of the work in progress

This is the lamp at an advanced stage of construction. Notice that the caps are not yet on the individual lamps, and there is overall lumpiness. Not yet sweetly polyhedral.

after plenty of work, here it is, looking much better. Thomasz is inspecting the innards, perhaps a little too close for comfort.

This is the lamp, powered up, in its temporary hanging location in the basement of industrial design.

and a closeup shows the gnarly internal structure, and part of the wiring.

The lamp’s internal FoxBoard queries a service, written as a python server in the Google App Engine environment. The server produces a configuration message that contains the new state of the lamp (that is, which groups are on and which are off). The FoxBoard then writes out the message to an Arduino over USB: the Arduino, in turn, squirts it out to two shift register circuits that present their outputs to optoisolators. The optoisolators separate the low voltage side of the system from the 220 V side, and drive the lamps.

The system is built around the fundamental limitations of compact fluorescent lamps: they are not easy to dim, and they don’t like being switched on and off frequently. Hence, the configuration will be pulled infrequently, perhaps once every ten minutes.

The Google App Engine server has been built on top of my (currently in progress) SERPE service prototyping environment.

August brings five lazy links to interesting stuff.

In August everybody is lazy, so I will just publish three five links

  1. A fantastic periodic table of information visualization. Techniques organized to show their relationship, with examples for each one.
  2. You cannot innovate like Apple. Apple is the canonical example of all that is good in the design of interactive things. But it looks like their process is not something you can grab and apply somewhere else. Not unless you manage to clone Steve Jobs and a few other key people. More about it here.
  3. Ivan Krstić‘s blog. Cool guy, worked at OLPC, now is in Apple.
  4. Just so you don’t forget how nerdy I can be, playing around with ChucK looks damn good right now. Regardless of the fact that I have managed to pack my schedule full from now to Christmas. Anyway, it is a Strongly-timed, Concurrent, and On-the-fly Audio Programming Language, and if that does not get you excited, I don’t know what will
  5. Oh, and I was reminded -by the resourceful Saakes- of Hypershot, "the first digital camera for your data", very fancy rendering, apparently not too difficult to use. Maybe I will give it a spin inside Computer Visualization.

There you go. Enjoy your summer sun, if you are in the right latitudes and you happen to actually like unfiltered solar radiation. I am also including a gratuitous picture of happy users

 

this is what happens at TUDelft industrial design when there is a fire alarm: people rescue their laptops, stream out of the building, hook up to the wireless and continue work. You could argue that we don’t really need a building, just a lot of grassy knolls with some power sockets here and there. 

 

 

making the FoxBoard talk to Arduino

Suppose you have a sketch loaded on your Arduino, and suppose that you need some serial communication. You try it out on your Windows box, and all is happiness. Then you attach the cable to the Foxboard (a single board computer running Linux) and nothing works anymore. Why?

Because apparently the default settings of the serial port are wrong for your purposes.

To set the serial port your way, you have to use the wonderful stty command, something that our hairy ancestors used to do when setting up terminals in the Lascaux caves. First of all, run the dmesg command. You will see lots of stuff, and at the end you should see something like this


crisv10_irq dbg: ctr_status_irq, controller status: host_mode started
crisv10_irq dbg: ctr_status_irq, controller status: host_mode started running
usb 1-1: new full speed USB device using hc-crisv10 and address 4
ftdi_sio 1-1:1.0: FTDI USB Serial Device converter detected
drivers/usb/serial/ftdi_sio.c: Detected FT232BM
usb 1-1: FTDI USB Serial Device converter now attached to ttyUSB0

This means that the USB serial converter in the Arduino is recognized, the driver has been loaded, and the Arduino is mapped by ttyUSB0. I have marked in bold the most important part. It mans that the Arduino’s serial port is now available as the /dev/ttyUSB0 file, for reading and writing.

To see the current configuration of the port type stty -F /dev/ttyUSB0

[root@vdhelm1 /dev]118# stty -F /dev/ttyUSB0
speed 9600 baud;
-brkint -imaxbel

This is a quite typical configuration, good for controlling your UNIX minicomputer from a VT100 terminal, and if we were on a time machine to Dr. Who land it would be probably the perfect configuration. For our purposes, though, we need to dumb it down a bit.

I have noticed that to make things work "like on Windows" I have to kill the local echo. This is done with the command

stty -F /dev/ttyUSB0 -echo

and then things work pretty nicely. Send stuff to the Arduino like this:

echo "whatever" >/dev/ttyUSB0

and see on screen what the Arduino is sending back like this:

cat </dev/ttyUSB0

extra info: if things work strangely, or don’t work at all, one reason can be that your application on the Foxboard side is sending out the right stuff, but the terminal is not sending it to the Arduino because it is in "canonical" mode. Canonical mode means, among other things, that information is sent one line at a time – which really made a lot of sense for controlling printers. But it may be that you need your data to reach the Arduino as soon as you send it. Additionally, binary protocol data cannot really be divided in "lines". What you want is a device that will just shoot stuff out on the wire as soon as you send it. To get just that, you need to put your device in "raw" mode.

[root@vdhelm1 /mnt/flash/root]116# stty raw -F /dev/ttyUSB0
[root@vdhelm1 /mnt/flash/root]116# stty -F /dev/ttyUSB0
speed 9600 baud;
min = 1; time = 0;
-brkint -icrnl -imaxbel
-opost
-isig -icanon

the parts in bold are the ones you type. Without delving too much in the output of stty, notice that "raw" is actually a macro for a bunch of configuration options. The "-" means NOT, so these are all character processing options having been switched off. If you want to find out why this whole wacky tty mess exists, and why it makes a sort of sense (historical, at any rate), there is a great page about The TTY demistified.

caveat: if, by mistake, you set your current tty to raw mode, things will become very funny indeed. Time to get another tty 🙂

the cube sphere egg fallacy (#5)

My colleague Thomas likes to quote a professor of his as being very happy whenever his students designed an interactive object in the form of a glowing cube, because this meant that the cube had been done and that the real design could start. I guess that it is always a strong temptation, having to do in part with science fiction imagery, such as the Borg cube

 

the Borg, being an entirely rational race, make cubic spaceships because they are not concerned with aerodynamics. Since these cubes will never enter an atmosphere, they can be cuboid and covered with fiddly little appendages. When they feel like doing something wild, they make a spherical spaceship.

On the other hand, if a human designs a cubic object, particularly if it is supposed to be grabbed by someone, it is a safe bet that he is a taking an easy way out and neglecting thought. Just try to remember how many perfectly cubical objects you know. Not many, right?

  • dice, cubic so that they can fall on each side with the same probability
  • Rubik’s Cube, cubic for gameplay reason
  • the Power Mac Cube, that was not really a cube
  • the Borg cubes (already discussed, and you are not designing for the Borg anyway)
  • the Weighted Companion Cube in the videogame Portal, designed to weird you out with its cubic yet loving inexpressiveness
  • silly MP3 players from 2 years ago
  • frankly stupid cube shaped clocks that go out of their way to make reading the time difficult

…one runs out of presentable examples pretty fast. And that is because a cube is not such a good form to hold in your hand: first of all it is pointy, and secondly it makes sense only if all its faces are equally significant and work well by being exactly the same size. It is not strange that there are very few cubes around us, while there are millions of objects that are bricks, tablets, slabs, and thin plates: their asimmetry has a reason, while the cube’s peculiar and extreme simmetry is just … a form. Real world objects have a good end and a bad end, a business end and a handle end, a top and a bottom, a front and a back. Their sides specialize, to a degree, just like animals – notice that most higher animals are symmetric only on one plane.

Cubes, on the other hand, have no favorite resting position. Each side is equally good. 

 

 

 

These are MP3 players and sharing devices, whose designer I will not mention, and they are really wrong. They offer no use cues: are you supposed to shake them, squeeze them, press them or put them one close to the other? Or eat them with a fine Chianti? Why do they glow, other than they can be made to glow? The bland cube, in its simmetry, is rivaled only by the blander sphere (or egg). 

 

These (horribly boring by now) glowing LED spheres make some sort of sense, because they are hand sized, meant to be grabbed and I bet that you have to squeeze them to switch them on and off. And this is about the only meaningful interaction you can have with a featureless sphere, other than throwing it at people and small animals. The moment you want to add more interactions, though, you will suddenly realize that the sphere is the wrong shape.

The sphere’s best buddy is the egg. Nicely aerodynamic and archetipal, it is a very good shape for containing a baby chicken and making kitchen timers. Oh, and chocolate eggs. And debatable USB minipicture frames distributed by Brando. Here on the side you can see probably the ugliest MP3 player in the universe, that combines the senseless egg shape with commands on the headphone wire (bad) and an incredible fake marble skin. That’s what I call a clear winner, and a good point to stop rambling.

 

 

If you liked this post, you might also like the previous fallacy.

 

 

Philips Lumiblade, milking the development kit

At the Salone del Mobile I visited the rather good Philips Lumiblade stand in the Superstudio area (via Tortona). I left them my contact data, and today they sent me a piece of email and a link, inviting me to obtain the Lumiblade experience kit. The Lumiblade is an OLED light source. Monocromatic, cool to the touch, energy efficient, and still relatively new.

 

(image grabbed from the Philips email without permission)

 

 

I followed the link, and I found out that the bare minimum  you have to spend to see the technology on your desk is 

70 Euro for the power source

72 Euro for one little white square (30 x 30 mm)

In case you wanted to experiment with colors, you could get 3 small rectangles in red, green and blue for a cool 500 Euro. Of course, plus VAT and postage. Can I give this to students to play with? Not really. Besides, Philips has not yet convinced me that their technology does something that electroluminescent (EL) material can not do… 

I understand that OLED is a new technology. But the buyers of these kits are not luxury-chasing consumers: they are the designers and architects that must snag the luxury-chasing consumers – and the schools and universities that train the architects and designers. The analogy here is with the software development kits (SDK) that are necessary to make software for new products. If you make the SDK expensive, the software companies will delay adoption of your new hardware: nobody wants to sink large sums of money into risky new technology. Wise companies that want to push a new technology make SDKs free or almost free (the iPhone SDK, for example, is free). Unwise companies that want to make a quick buck make SDKs very expensive. 

Is there ever a case for an expensive DK (or an expensive experience kit, as is the case for the Philips OLED)?

  • It makes sense to be expensive if your revenue comes from selling the DK much more than from selling the technology, perhaps because there are many other companies out there doing manufacturing
  • It makes sense to be expensive if you are selling such incredibly hot, unique, shit that nobody else has it, and so you can basically squeeze the customers for the product, the developers for the DK and any other stakeholder as much as you want. There are not many businesses like this.

I wonder, is Philips in either of these positions with regards to the OLED light sources?

the magic fallacy (#4)

In the vein of the abstract fallacy, today I point out the magic fallacy. The most typical case is when a project is described in terms similar to these:

the user opens the box, a green light inside it starts pulsating and the user feels a magical feeling

How exactly does this happen? Of course the user does not know why e.g. opening a box and seeing something gives him a magical feeling, but the designer should know. To put it more formally, a designer is not allowed to suppose psychological changes in the user without indicating the psychological means that produce these changes. Of course it is possible to surprise users, to delight them and even to make them feel like magic, but how do you do that? The project description should indicate what psychological mechanism happens. Is it a visual reference to something well-known? Is it a perceptual trick? Is it culture specific?

If you state that Canon’s installation at Salone del Mobile, the enormous NEOREAL piece installed in Triennale, creates a magic feeling through large scale complex geometries in a dimmed room

 

and the illusion of being in an underwater world where you can interact with glowing jellyfishes

 

 

then I believe you. There is no theoretical reason, it is just that, in my model of human behavior and reaction, large scale immersive spaces where a touch of sensorial deprivation is combined with big, sharp biological images in soothing blue DO indeed produce a magical, far-out, slightly hallucinatory feeling. Just like a desser composed of a kiwi, a tangerine and some peanuts produces a vaguely unsatisfying bizarre feeling at the end of lunch. 

 

To fall back on the well known, let ne quote the usual Proust: we are not just told "I ate a little sweet dipped in tea, and I felt incredibly happy." The narrator lets us know exactly what connects the happiness to the food (and also the process by which he finds out the connection). His question

D’où avait pu me venir cette puissante joie? Je sentais q’elle était
liée au goût du thé et du gâteau, mais qu’elle le dépassait infiniment,
ne devait pas être de même nature. D’où venait-elle? Que
signifiait-elle? Où l’appréhender?

in the English translation by C. K. Scott Moncrieff:

Whence could it have come to me, this all-powerful joy? I was conscious
that it was connected with the taste of tea and cake, but that it
infinitely transcended those savours, could not, indeed, be of the same
nature as theirs. Whence did it come? What did it signify? How could I
seize upon and define it?

should also be the designer’s question. I do not deny that a green light pulsating in a box can be a source of magical feelings, but whence the connection? What does this feeling mean? How can we grab it and define it?


 

(the Canon NEOREAL photographs are taken from the project site: the kiwi picture is unfortunately mine)

 

 

the abstract fallacy (#3)

Whenever students tell me …and this phenomenon will be visualized by abstract graphics… or basically anything that includes the word abstract, I know that there is trouble on the horizon. Why? Well, this is abstract

 

The Matrix screenshot

 

 

this is abstract

 

and this too!

 

this, btw, is a variation on the Sierpinski triangle fractal, made by a guy that calls himself FireLizzard with a fantastically interesting piece of software: Context Free.

But to go back to why saying abstract is bad, I should be more specific: it is bad only when you say it to create a smokescreen over missing visuals or even serious difficulties in information visualization or interaction design. Abstract graphics, particularly if they are meant to convey some information (as opposed to being screensavers or cute wallpapers) are damn hard, I daresay because they sit at the intersection of graphic design and information design, with the extra difficulty that they cannot contain representations of real world objects or icons that refer to them (abstract, you see…), so if you want to represent the state of health of an old person (for example) you can’t just have an old-guy icon bent double because of backache or happily strolling around. You have to come up with something else, and I bet that it is going to be pulsating and glowing like a damn Macintosh power LED. But I digress.

To sum it up,  abstract is OK as long as you realize that it is a placeholder, and that making those abstract graphics work well could easily be the most difficult part of the whole design.

 

 

in other news: Max Bruinsma is a highly articulate guy with very interesting things to say on design

ITD Jukebox project goes to Salone – 1

Just a quick note from the center of insanity that is our little lab here in Milano. The prototype is being finished, the software is being tweaked and the Dremel is getting a nice little workout. The Dutch students are finding out exactly how many espressos you can drink before you pitch face forward into a pizza. Meet the prototype as Jordi spins it rapidly for stress testing. Or was it just to look cool?

 

The odd-looking eggy-pill like things in the hands of Sjoerd are the song tokens: each one stands for a specific piece of music, according to a very smart and impossible to understand color code. Color coding rules! Unfortunately. 

 

 

A detail of the top plate. While you are laser cutting, why not take some extra time and do a fancy bit of laser engraving? 

 

 

This is what things look like late in the TUDelft-at-IdLab workspace at night with 24 hours to delivery time. Perplexed people considering magnets and fancy spring-based braking devices. Of course right now everything is MUCH better and all technical issues have been solved by Dieter who, in the next photo, looks just so happy.

 

 

 

It is a bit blurry, but I just wanted to share it with the world. Suicide by glue gun. A very TU way to go. Don’t try this at home, unless you are wearing your fancy giant white IKEA protection helmet!

Now we have to rush to Triennale, to install our fancy graphics on a 2.5 x 2.5 m display wall: or at least this is what they have told us. Considering the Salone way of doing things, I wouldn’t be surprised to find out that our space has turned into a 10 x 10 m ellipse, or the inside of a turnip, or the spires of a Gothic cathedral.

 

© 2011 TU Delft