Blog & White

L'envie m'a prise de construire une machine à battre les cartes en LEGO. Dont acte, voici la bête. Le principe est très simple : des roues en caoutchouc entrainent la carte du dessous d'un paquet et l'envoie à travers une fente. Ce mécanisme est reproduit deux fois, et les cartes de deux demi-paquets sont envoyées au hasard vers un emplacement central.

Détail du mécanisme d'entraînement. La grande roue dentée sert de manivelle. On voir aussi une roue en caoutchouc.

Une des difficultés a été d'empêcher les cartes de se placer de travers lorsqu'elles sont entraînées. Les barres verticales les maintiennent en position et sont ajustables à la largeur du jeu de carte.

Le mecanisme en action : on voit une carte passer à travers la fente. Pour plus de clarité, seul le coté droit contient des cartes.

C'était sympa à bricoler, mais un vrai casse-tête à faire fonctionner correctement. D'ailleurs, ça ne fonctionne pas correctement, trop souvent une carte passe à travers la fente d'en face et les cartes se bloquent ; ou encore une carte se pose de travers et empêche les suivantes de s'empiler correctement.

“Homes already have TVs and PCs so it is logical for the devices to unite.”

This sentence is quoted in the 22.06 – 30.08.2007 issue of Six Degrees (Finland's English Magazine) and attributed to Esko Sario, program manager for Microsoft Finland.

Well, Mister Program Manager, I have another one for you:

“Homes already have ovens and refrigerators so it is logical for the devices to unite.”

I'm impatiently waiting for the first prototypes. I hope they will build digital ones, too.

Last Christmas, I got a LED-based lamp that you can clip onto a book for reading in dark places, e.g., in bed without lighting the whole room. It's very handy, lightweight and it lights both pages of a book very well. It would be perfect if it didn't require Lithium batteries (CR2032, button-shaped batteries, 20 mm diameter and 3.2 mm thick). These batteries are expensive if bought in small amounts, and they need to be replaced often compared to e.g., AAA batteries (at least in theory).

So I decided to fit AAA batteries onto it. The original lamp's design is very simple: two 3V batteries in series, one white LED, a switch and that's it. No fancy electroncics, no black magick: you can't make it any simpler. So replacing the Lithium batteries with alkalines is very simple in theory: fit a holder for 3 AAA alkaline batteries, add a 33 ohms resitor in series, connect the wires to the contacts, and Bob's your uncle. In practice, it was a bit more difficult, but not much.

3-AAA Battery Holder

The electronics shop I went to didn't have any 3-AAA holders, only 1 piece or 4 pieces holders. So I went for three times one and I glued them together with cyanoacrylate glue, after sanding the surfaces to ensure that they were perfectly flat.

Then, I needed to connect the three batteries in series. This has been done with the legs of the resistor.

I added two wires from the remaining connectors and routed them through the ready-made holes in the holders. To one of the wires, I soldered the resistor. It rests in the angle of the battery holder: since the holder is square, and the battery is round, there is some space left along the sides, enough for a wire or even a resistor. I used insulation tape to prevent contact between the soldered resistor and the battery.

Lid Replacement

The original lid for the battery compartment is nicely round and everything, but this makes it impractical to glue anything onto. I therefore decided to make my own replacement for this part, which would be flat (and probably ugly). I got a piece of Plexiglas (PMMA, or acrylic glass) from a friend, and cut a piece of more or less the right shape. The fitting was actually quite difficult, because the battery compartment has several screws and bits that are protruding slightly. It required a lot of sanding before getting a good enough fit. Plexiglas can be bent slightly, but if the back of the support wouldn't be flat, the battery holder could not be glued anymore with cyanoacrylate (which requires perfectly matching surfaces for bonding). Thank $DEITY for powertools!

This is the inside of the battery compartment. It has one screw on the left, two protruding rings holding the original CR2032 batteries, and two more screws on the right. Moreover, it has a slot in the far left where the lid's tail bit goes. The next picture shows this tail bit, on the far left of the support piece.

The tail bit is made of a piece of the same plexiglas sheet as the support. I glued it in the right place with cyanoacrylate, and then I sanded it until it was thin, short and narrow enough to fit into the slot. The two next pictures show more of the sanding that was needed in order for the lid to fit on top of the screws and other bits. The first picture shows the tail bit again on the left, as well as a hole for the screw located next to the slot. The second picture shows the hole through which the fixation screw goes, as well as the places where the two other screws go. On both pictures, you can see the central hole where the wires go through to the contacts in the lamp's original battery compartment.

The next two pictures show the whole device I built, including the contacts, from two more angles. You can see the resistor wrapped in red insulation tape.

Contacts

My first idea was to use two 10 eurocents coins as contacts, since they have almost exactly the same diameter (19.75 mm) as the original batteries (20 mm) while being a bit thinner, which would have allowed to add the wires or some kind of foam adding pressure and maintaining the electrical contact. It was a bad idea: soldering onto the coins is difficult, because the coin acts like a heat sink for the soldering iron and the tin, instead of reaching the right temperature that would allow to make a proper solder. Well, it's not that bad: the device is 0.20 EUR cheaper this way.

Instead, I used aluminium folio, cardboard and double-sided tape to build round contacts that fit into the battery holes, as shown on the three next pictures.

Integration

Finally, here's the device with its batteries, mounted onto the lamp:

It looks quite “hacker-techno-raw” (or whatever you want to call it), but it works. It's of course heavier than the original lamp, but the clamp is strong enough to hold even on a paperback. The only drawback is that the angles are still quite sharp, and the battery pack hurts the skin when it lies on your lap.

Electricity Consumption

The 3 AAA batteries give 4.6 V without a load (i.e., when the lamp is off) and 4.5 when it's on. The voltage across the 33 R resistor is 0.7 V, which gives a 21 mA current and 3.8 V across the LED, which look like the nominal values one can find on the Web. The LED therefore consumes 80 mW and 15 mW are wasted in the resistor and the batteries' internal resistance (that's about 16%).

Parmi les quelques blogs sur le thé chinois que je lis régulièrement, l'un d'entre eux (pas le plus modéré dans ses propos, il est vrai) a publié voici quelques semaines de ça un article élogieux sur mon magasin de thé préferré, Langage du thé situé rue des Charpentiers à Strasbourg. En guise d'introduction, l'auteur a listé les magasins strasbourgeois les plus significatifs dans le domaine sans les nommer, en donnant son opinion à leur sujet. L'un des magasins a été décrit comme suit :

« un grossiste importateur-revendeur de thés "bio-équitable" et son jardin zen paumé au fin fond de la campagne (pour ceux qui se contentent d'un joli décor plutôt que de la piètre qualité des thés proposés; le tout étant dirigé par un couple d'anciens XXXXXXX XXXXXXXXXXX XXXXXXXX-XXXXXXXX écolos-XXXXXXX-XXXXXXXXXXXXXXXXX qui XXXXXX XX XXXXX XXXX : bref, XX XXXX XXXXX; sans commentaire !!) »

(source: le cache de Google. Les mots qui pourraient être considérés comme diffamatoires ont été censurés, puisque même une citation à titre d'illustration d'un propos diffamatoire est considéré comme diffamatoire).

J'ai appris hier que les responsables de ce magasin ont menacé l'auteur de cette critique d'un procès s'il ne retirait pas immédiatement son article. Ne voulant pas risquer un procès (qui coûterait cher en temps, en l'argent et en santé mentale), il a obtempéré, retirant tout l'article.

Je ne sais pas quel a été le motif de leur menace, mais en dehors de la diffamation ou de l'injure publique, je ne vois rien qui puisse avoir une base légale légitime, et certainement pas la critique de la qualité de leurs produits. Mais surtout, si (ça reste à vérifier) ces gens là ont exigé le retrait de l'article dans son intégralité, ce qui a pour effet de faire disparaître en particulier la critique de leurs produits et l'éloge de leur concurrent, je ne peux que soupçonner qu'ils ont cherché à faire taire un de leurs détracteurs, une technique qu'on peut qualifier de terrorisme juridique (au sens propre, « forcer en faisant peur en usant du système légal »), très à la mode ces dernières années, surtout en abusant des lois sur le copyright ou le droit d'auteur.

Cette histoire m'a aussi permis de me pencher sur la définition légale de la diffamation : juridiquement, c'est une cochonnerie qui s'oppose exactement à la liberté d'expression lorsque des personnes ou des corps sont impliqués. N'importe quelle critique impliquant directement ou indirectement des personnes peut être attaquée pour diffamation. L'injure publique, définie juste après la diffamation, est encore plus difficile à cerner, puisqu'un terme utilisé vis-à-vis d'une personne peut être considéré comme injurieux, alors qu'une autre peut en retirer de la fierté. Par exemple, je me définis moi-même comme un écolo-gaucho-altermondialiste alors qu'un capitaliste néoconservateur trouvera le terme injurieux. De même, toujours à titre d'exemple, des anciens hippies poussiéreux soixante-huitards se souviendrait avec nostalgie de cette époque, alors que d'autres se sentiraient insultés.

Quant à moi, je ne mettrai plus les pieds chez eux, et surtout, je le ferai savoir autour de moi.

Et hop, un an de blog, 270 message, et même quelques commentaires. Et aussi presque 42000 spams.

© BoardGameGeek.com

When thinking about Dan Rosewater's expansion cards for Caylus, I wondered how to decide if a card should bring extra prestige points or rather royal favours?

Reviewing the official cards in Caylus, I came to the following conclusion:

• You get prestige when you do something that increases your rank in the social hierarchy, when other people in city notice you more, maybe respect you more. In other words, you get prestige when you make yourself look good in the eyes of the others. This can be achieved by expanding the city and the castle of course, but also by making donations to the church, or wearing expensive cloth (made by the tailor) or jewellery (made by the jeweller, if you use this extension card).
• You get a royal favour when you do something that makes you look good in the eyes of the King, such as promoting education (e.g., building the university), arts and culture (building the theater, the monument), providing entertainment(fighting in the joust field), or flattering the King (building the statue – because I'm pretty sure the statue is a statue of the King).

Then about Dan's cards:

• The Guild of Thieves: I love the idea, but I think that the card should allow players to trade prestige points in exchange for deniers or resources (how many of them remains yet to be decided). This would be consistent with what I wrote above, i.e., when you do something bad such as using the services of a thief, you loose some amount of prestige. Moreover, those should not be stolen from other players, but taken from the “bank”, i.e., stolen from someone in general. Resource/Denier exchange between players is not a part of the game mechanics, and maybe it should remain so. Finally, it should maybe cost more than one or two prestige points to build, I think. And it also could give its owner a denier e.g., as a tax taken by the guild from its members, when used by another player.
• The Hospital: I love the idea too, but I think that it should not provide a royal favour to the players, but rather be used for trading any resource or deniers for prestige points. In other words, when you make a donation to the hospital, you get to look good in the eyes of the people in the city. Maybe the hospital would be worth getting a favour when built, though.
• The Royal Luxury Taylor: this one I hate, because there is already a taylor, and getting prestige for nothing is not in the spirit of the game.
• The Trader: I somehow like it, but again not really. There are already trader-like buildings in the game, such as at the market, the peddler, the alchemist and the bank. Otherwise, it acts as a resource providing building mixed with the comptoir, providing both resources and deniers, which is something new. I have to think more about it.

© BoardGameGeek.com

The resources in Caylus are not equally easy to get when using production facilities such as farms, sawmills, quarries etc. We can then find what are their respective values based on their scarcity and by extension, we can derive the values of the deniers and of the pretige point.

Values of Resources

If we consider that all the production facilities have been built, we can count how many resources are produced by each of them (when an alternative is possible, such as in the wooden farms, we consider only half of the number of resources; resources gained by the owner of stone buildings are not taken into account) and calculate values of the resources relative to each other. Here are the results:

• Gold: 100.0 (reference value)
• Food: 18.3
• Wood: 18.3
• Stone: 20.2
• Cloth: 25.3

Food and wood have the same value, stone is worth 10% more than the two previous ones, cloth is worth 38% more than food and wood, and 25% more than stone, and gold is worth the most, as expected.

Deniers

So, how much are deniers worth? It depends a lot where you spend them or where from you get them.

At the bank, one denier is worth 50.0 or 40.0, depending on the amount you change.

At the neutral market, its worth 4.6, 5.1, 6.3 or 25.0, depending on the kind of resource you trade, respectively food/wood, stone, cloth or gold. At the wooden market, these values are respectively 3.1, 3.4, 4.2 or 16.7.

At the neutral peddler, it's worth 9.2, 10.1, or 12.7 for food/wood, stone or cloth, respectively. At the wooden peddler, the values are respectively 18.3, 20.2 or 25.3.

Prestige Points

The taylor changes cloth into prestige points, with a rate of one cloth turned into two points. This sets the prestige point's value to 12.7.

The jeweller sets the value of the prestige point to 20.0 or 22.2, depending on how much gold is spent at once.

The church sets the prestige point to 2.1-2.5 (with a denier worth 3.1).

Building wooden buildings sets the prestige point to 18.3 (farms, sawmill, quarry), 9.2 (mason, market/peddler built with food), 9.6 (market/peddler built with stone) or 10.9 (notary, market/peddler with cloth).

Building stone buildings sets the prestige point to 12.8 (farm, park, workshop), 6.4 (architect, taylor, bank, alchemist) or 5.7-15.1 (church, depending on how much prestige the favour can be turned into).

Residences are harder to evaluate, since they cost a denier in addition to the cloth, but using the lowest rate of the denier (3.1), it can be set to 14.2.

The points obtained from the construction of prestige buildings are worth as follows: 15.5 (library), 16.0 (cathedral), 14.8-20.0 (monument), 16.3 (hotel), 13.4-18.2 (theater), 13.7-18.6 (university), 14.7 (weaver), 15.5 (granary), 11.7-20 (statue). Some of these buildings also give favours, which can be turned into one to five prestige points, which explains the variable values for the monument, the theatre, the university and the statue.

Finally, building at the castle sets the prestige point's value to 7.1-18.9, depending on which part of the castle is being built and wether you get a favour that can be changed into prestige points because you were the first builder.

Optimization

Now, how do you get the most out of it? This is not the same as rich, since we have seen above that the value of deniers is very much variable depending on who you deal with.

One possibility would be to sell at the market food/wood for deniers, and then buy cloth at the peddler's or even gold at the bank. This allows to change e.g., one food/wood for three cloth or two golds. It's much more interesting than going to the alchemist's (one food/wood for half a gold), except if you can't spend two rounds on the transactions.

Another possibility is to sell wood/food at the wooden market, and then buy prestige points at the church, at a very low cost. This is probably the cheapest way to gain prestige points, but it's quite slow.

With building construction, the best choices are the architect, the taylor, the bank and the alchemist, since they bring prestige points at the lowest cost.

At the castle, the best possibility is to be the first to build a piece of the walls and turn the favour into five prestige points.

Warning

These values are true only when all the production buildings have been built. If some of them are missing, the value of the resources that they are not producing is getting higher and this changes all the calculations. If I have time, I'll make a script that recalculates all these quickly.

Here are the fonts used in game artwork that I have needed for making my own game components. For both games below the fonts were not freely available, but in the case of Caylus a good replacement was available, and for Carcassonne I used MyFonts.com's sample text rendition to get words or pieces of words, and then I used potrace to trace a vector contour and save the words in EPS. This gives the possibility to change the size of the words without any of the aliasing problems that occur when resizing a bitmap. Assembling the words or pieces of words to make titles was annoying, but that's better than nothing.

Caylus

The offcial one is Engraver's Old English.

Replacement fonts, available for free:

• Cloister Black.
• Old London.
• English Towne (this is the one I used, it's very similar to Engraver's Old English except the lowercase “y”).

Carcassonne

The font is very similar, although not identical to Lindsay, and has no known free equivalent. Lindsay is thinner, its legs are a bit longer and has no serif on e.g., the left leg of its “n”.

The physics involved in flying a water rocket is not very complex if you have studied a little bit of mechanics and fluid mechanics. It is summarized in a PDF file written by Peter Nielsen, Associate Professor in Civil Engineering of the University of Queensland, Australia. I wrote a Perl script that implements the formulas and outputs data plotable with Gnuplot.

I just finished fixing a bug in the script which was promising wonderful performance of my rocket (like over 200 m altitude and a maximum speed of over 350 km/h), but which were never reached by our rocket. The water rocket that we built last week-end is in theory able to reach 38 m after 2.7 s of flight. Its top speed is 35.5 m/s (126 km/h) after 0.33 s of acceleration, and the acceleration increases from 95 to 169 m/s² within the same time.

After that, it decelerates during a bit over two seconds, and then starts to drop back to the ground, and lands about 5.8 s after take off (omitting the effect of the parachute).

Tuning the parameters of the script, I found the following optimal values: the weight of the rocket should be 140 g, the filling ratio of the water tank should be 0.38 with a pressure of 7 bar, and 0.45 with a pressure of 5 bar. Because of former erroneous calculations, I thought that the optimal filling ratio was about 0.60, which lead to much poorer performance, due to the overload. This second graph shows the theoretical height, speed and acceleration of the rocket during its last flight. The water empties in about 0.32 s, which corresponds to what has been seen on the video clip, and the maximum height is 36.8 m, which seems reasonable (although it has probably never been reached because the parachute opened far too early).

The drag coefficient is one parameter that is very difficult to estimate, and for now its value is 0.9 for the calculations, which is barely better than the one of a brick (about 1.1). Its real value may be less. Or more. Who knows…

Also, the leaks in the pressuring system make it difficult to estimate what was the actual pressure in the rocket at take off. And water was poured without real accuracy, so the above filling ratio may as well be 0.8 or 0.5.

Reijo and Juha have been busy last week building another rocket using old CDs as fins and also building a new launching pad out of metal tubes. Here are the pictures.

Sikaflex has again been used to glue the fins to the bottle.

This launchpad is made of metal tubes and doesn't leak anymore. Moreover, it is orientable and mounted on a stand (on the contrary to the old one which was attached to a pole).

Mozillalla (ja ehkä myös Firefoxilla) Elisan/Kolumbuksen webmail ei toimi hyvin, koska viestin otsake ei näy. Tässä on helppo korjaus:

userContent.css-tiedostoon, lisätään

div#mail-header, .content-functions, div#mail-steps, div#contact-cards, div#contacts-header, div#view, form.search, div#mail-header p, .splitted, form.search p, .action-info, p.go-oma-elisa, .prompt-controls, form#login p, .new-features { height:auto ! important; }

userContent.css:n paikka rippuu siitä, mitä käyttöjärjestelmä on käytössä. Tämä sivu selittää mistä se löytyy Firefoxilla (englanniksi).

Muutama linkki:

• Vaestörekisteri
• Ajoneuvokeskus
• Suomen Suoramarkkinoiintiliitto
• Tietosuojavaltuutetun toimisto

Some months ago, I discovered water rockets while reading Make magazine's 5th issue. The idea is very simple: take a plastic soda bottle, fill it partly with water, add pressured air, and release. The practical difficulties are mainly the release mechanism and the recovery mechanism (i.e., releasing the parachute not too early and not too late). So two weeks ago I started to experiment with my brother-in-law, starting with the release mechanism and a simple bottle, as well as a standalone parachte. This (long) week-end, we built a full rocket, inclunding a nose holding the parachute and fins. It was painted red, because the transparent bottle was hard to spot in the sky, and even harder to find after landing in high grass.

Lauchpad Construction

The lauchpad is made of one PVC tube that goes into the rocket, plus one wider tube around the first one that is part of the release mechanism and holds the rocket while on the tube. Three screws placed on the grey tube at 120° from each other hold the white tube in the middle of the grey one.

An O-ring is placed on the thinner tube so that the neck of the bottle will go just around it.

Release mechanism

The release mechanism relies on the wide ring just below the bottle's neck. Once on place on the tube, a metal pin will go around the neck and above the ring, preventing the bottle from taking off. The closeups will show the ring, just visible through the slot in the grey tube.

A rope connected to the pin allows to pull it from a safe distance. The rope is attached to the pole so that the pin doesn't jump to the face of the person pulling it (it hurts).

Currently, the release mechanism is not very reliable, it gets jammed from time to time and the rocket doesn't take off vertically. Also, the seal around the launch tube is not tight enough, there is some amount of leaking.

The other end of the thinner tube is connected to the air compressor. Getting a tight enough connection here was not easy, but after several attempts we found one combination of rubber hoses and collars that works. More or less. It sometimes pops off.

Rocket Construction

Building the rocket was quite straightforward. One thing to remember: epoxy will hold the fins onto the rocket quite well, but it will break upon impact on the ground (understand: crash landing when the parachute doesn't open). Polyurethane glue (actually some kind of PU filling material) on the other hand will hold them in place and be flexible enough so that the bond won't break on impact. The aspect is pretty ugly, but it works. It just needs to dry overnight. Building a support for the bottle and the fins helps keeping them in place when drying. It's made of a cork screwed into a piece of wood of the proper height, and attached to a plank into which slots have beed sawn in order to hold the fins.

Parachute

The parachute is folded into the nose cone during the flight, and the nose is supposed to pop off when the rocket reaches its apogee. The problem was that more often than not, the cone didn't open, and the parachute didn't open at all. Finding the balance between a nose which is too tight on the rocket and a nose that falls off with a simple gust of wind before take off is very difficult. According to this website, I think the rocket should take off as vertically as possible, and the nose should hold only because of the force of the acceleration, and fall off as soon as the rocket is not in stable flight anymore. But the holes in the cone for decreaseing the pressure inside it is something to try, too.

Flight

This is the last flight of the week end, the only one that was more or less a success with the red rocket. The video was shot at 30 fps, therefore one frame below lasts 33 ms. One can see that the nose separates quite early from the body, probably because of the flight not being straight up.

The original video clip is also available.

Bad Landings

These bad landings where quite soft ones, because the parachute did open and slowed the rocket down.

There have been other bad landings where the nose of the rocket didn't fall off, and where the rocket fell down like a stone straight onto its nose. It's been bumped in many times, but it was still able to fly afterwards! The test bottle once crashed and had many bumps and creases, but it was possible to inflate it again, and the only tiny hole it had could be fixed with duct tape (blessed be the inventor of duct tape!)

Food pairing is fun, because it looks absurd at first sight. Then you think about it, and it still seems improbable. Then you taste it, and… sometimes it's good, sometimes it's not ;)

This second edition of They Go Really Well Together started by Martin at khymos.org is now hosted by Tara at Should You Eat That?. This time, the goal is to associate banana with parsley. I did it with an obvious parsley enhanced banana milk shake.

Ingredients

• 2 ripe bananas
• 100g vanilla ice cream
• about 5 dL milk
• 5 cL fresh parsley

Preparation

Peel and slice the bananas into a blender. Add the ice cream cut into small parcels (otherwise you might clog the blender) and the milk. Mix until smooth.

In a small and fast blender, chop the parsley. Add about 2 dL of milk shake, and mix until smooth.

Comment

It's foamy and creamy like any banana milk shake, but definitely much more green! The first attempt smelled and tasted obviously of banana and vanilla, with a hint of something else you can't recognize. The parsley then bursts out in the aftertaste.

In the second attempt I put a little less parsley (the leftovers of the bouquet, actually), and whereas it looked the same and tasted fine, the aftertaste was absent. I suppose numerous experiment and a precision scale (for weighing the parsley) would be necessary to be able to reproduce the recipe.

I dug again into my hacked blosxom to find more ugliness, such as opening files for all the stories, including the ones that are not going to be displayed. This divided the time spent in blosxom::generate by 3.

Surprisingly, replacing the look-behind assertions in interpolate_fancy::__ANON__'s regexps didn't made as big a difference as in textrite::rite (1.11 ms against 0.813 ms), probably because there wasn't that many of them and weren't called that often

Profiling results are now like this:

Total Elapsed Time = 1.236275 Seconds
  User+System Time = 1.066275 Seconds
Exclusive Times
 %Time ExclSec CumulS  #Calls sec/call Csec/c Name
  10.2   0.109 0.127     134   0.0008 0.0010 interpolate_fancy::__ANON__
  7.50   0.080 0.178      13   0.0061 0.0137 blosxom::BEGIN
  5.06   0.054 0.428       1   0.0540 0.4283 blosxom::generate
  4.41   0.047 0.098     368   0.0001 0.0003 entries_index::__ANON__
  3.75   0.040 0.040       7   0.0057 0.0057 CGI::_compile
  3.75   0.040 0.054       9   0.0044 0.0061 CGI::import
  3.75   0.040 0.040      40   0.0010 0.0010 textrite::rite
  3.66   0.039 0.039     690   0.0001 0.0001 File::Basename::fileparse
  2.81   0.030 0.060       8   0.0037 0.0074 Net::SMTP::BEGIN
  2.81   0.030 -0.000     62   0.0005      -  Exporter::import
  2.81   0.030 0.030      40   0.0007 0.0007 magiclink::story
  1.88   0.020 0.020       5   0.0040 0.0040 autotrack::BEGIN
  1.88   0.020 0.030       7   0.0029 0.0042 IO::File::BEGIN
  1.88   0.020 0.089       8   0.0025 0.0111 writeback::BEGIN
  1.88   0.020 0.020      41   0.0005 0.0005 writeback::real_path

The User+System Time is lower, but for some reason the Total Elapsed Time is about the same.

Profiling results for the new blog engine (main page):

Total Elapsed Time = 1.245757 Seconds
  User+System Time = 1.235757 Seconds
Exclusive Times
 %Time ExclSec CumulS  #Calls sec/call Csec/c Name
  14.9   0.185 0.586       1   0.1851 0.5857 blosxom::generate
  12.0   0.149 0.187     134   0.0011 0.0014 interpolate_fancy::__ANON__
  6.47   0.080 0.158      14   0.0057 0.0113 blosxom::BEGIN
  5.66   0.070 0.070      40   0.0017 0.0017 textrite::rite
  4.53   0.056 0.095     367   0.0002 0.0003 entries_index::__ANON__
  4.05   0.050 0.059       8   0.0062 0.0074 Net::SMTP::BEGIN
  3.24   0.040 0.108       8   0.0050 0.0135 writeback::BEGIN
  2.43   0.030 0.030       7   0.0043 0.0043 CGI::_compile
  2.43   0.030 0.043       9   0.0033 0.0048 CGI::import
  2.27   0.028 0.037     317   0.0001 0.0001 interpolate_fancy::_resolve_nested
  1.62   0.020 0.020       1   0.0200 0.0200 archives::filter
  1.62   0.020 0.019       4   0.0050 0.0048 entries_index::BEGIN
  1.62   0.020 0.030       5   0.0040 0.0060 autotrack::BEGIN
  1.62   0.020 0.020      27   0.0007 0.0007 vars::import
  1.62   0.020 0.010      62   0.0003 0.0002 Exporter::import

The problem in textrite::rite came from the use of look-behind assertions in Perl regexp. I've changed all of them so that look-behind is not needed anymore. The same problem occurs in interpolate_fancy::__ANON__, but I'm not going to look into it yet.

The other problem was that textrite::rite was called for all the stories, including the ones that were not displayed (which was most of them). I changed the internal structure of the main script so that the processing of the story happens after discarding the non-displayed ones.

Un litre d'essence produit 2,4 kg de CO₂, donc ma voiture qui a consommé en moyenne 6,8 L/100 km ces deux dernières années produit environ 16,3 kg de CO₂ pour 100 km.

Et un humain alors ? Si on considère qu'une inspiration-expiration échange 0,5 L d'air, que l'air inspiré contient 21% d'O₂, que l'air expiré en contient 16% et qu'on respire environ 20 fois par minute, on trouve qu'un homme produit 1,3 kg de CO₂ par jour. Sachant qu'un homme peut marcher 40 km/j, il faut 2,5 jours pour parcourir 100 km, d'où une production de 3,3 kg de CO₂/100 km. Sur cette distance, la voiture peut transporter 4 personnes sans produire considérablement plus de CO₂, donc 4 personnes marchant 100 km produiront 13,2 kg de CO₂, ce qui finalement n'est pas tellement moins (19%) que ma voiture.

La différence, évidemment, c'est que nos quatre marcheurs ne consomment pas de l'énergie fossile mais de la biomasse, donc le CO₂ qu'ils produisent a déjà été extrait de l'atmosphère.

Profiling results for the blog engine (main page):

Total Elapsed Time =     3.74 Seconds
         User Time =     2.99 Seconds
Exclusive Times
 %Time ExclSec CumulS  #Calls sec/call Csec/c Name
  26.0   0.780 0.780     253   0.0031 0.0031 textrite::rite
  18.0   0.540 0.800     876   0.0006 0.0009 interpolate_fancy::__ANON__
  9.03   0.270 2.540       1   0.2700 2.5400 blosxom::generate
  7.02   0.210 0.260    1911   0.0001 0.0001 interpolate_fancy::_resolve_nested
  3.68   0.110 0.110    3663   0.0000 0.0000 IO::File::open
  2.34   0.070 0.160      11   0.0064 0.0145 blosxom::BEGIN
  2.34   0.070 0.070   11867   0.0000 0.0000 UNIVERSAL::can
  2.34   0.070 0.140     741   0.0001 0.0002 lang::__ANON__
  2.34   0.070 0.290     253   0.0003 0.0011 writeback::story
  1.67   0.050 0.070     253   0.0002 0.0003 magiclink::story
  1.34   0.040 0.070       9   0.0044 0.0078 CGI::import
  1.34   0.040 0.040     254   0.0002 0.0002 writeback::real_path
  1.34   0.040 0.040     253   0.0002 0.0002 translate::story
  1.00   0.030 0.030    1052   0.0000 0.0000 UNIVERSAL::isa
  1.00   0.030 0.080       8   0.0037 0.0100 writeback::BEGIN

That's what you get when you abuse of Perl regexp. Maybe caching of the HTMLized stories would help a bit?

I installed yesterday Firefox 2.0.0 on my desktop computer, which is a silent, low consumption computer based on a Via C3 processor (1GHz, 256 MB RAM). Guess what? When scrolling down a long page (the main page of this blog), Firefox was slower than Mozilla 1.7.5 which I have used for years. Mozilla was scrolling down more or less smoothly, using up to 70% of the CPU time, whereas Firefox was not smooth at all, using 100% of the CPU most of the time. Firefox maybe an advanced browser, but definitely not a lightweight one whatever some people may claim. I think I'll stick to plain old Mozilla for still some time.

Français

• Maman lapin peut pi pas faire dodo = Maman Lapin ne peut pas faire dodo

Suomi

• minä pelästyttää = minä pelkään
• jälkiluoka läätelö (meillä ei syödä jäätelöä koko aijan!)
• sitä tulee minulla villasukat
• tämä menee vasempaan ja tämä menee oikeanpaan
• Emma tutkii ruispalan ulkopintaa: “Tässä taatsessa on tämmöseitä jauhetta”

J'ai un gros rhume depuis jeudi dernier, et depuis quelques jours je n'ai plus d'odorat (ça s'appelle anosmie, un mot à retenir pour briller dans les salons où l'on cause). Le résultat, c'est que plus rien n'a de goût : le jus d'orange ressemble à de l'eau acide et vaguement sucrée, les tartines sont fades, le saucisson est simplement salé. J'ai même accepté de manger de la soupe ce soir, c'est dire ! Jusqu'à hier, j'aurais pu penser que la perte d'odorat était due au nez bouché, mais depuis aujourd'hui je peux respirer à peu près librement par le nez, et l'anosmie persiste. Même la rétro-olfaction ne fonctionne pas, sauf à me moucher très fort, et encore les odeurs perçues sont faibles (je suppose que la pression de l'air sur la paroi du nez a à voir avec l'odorat).

I found a nerd test on Tinou's blog, so of course, I had to take it. Looks like I'm quite a nerd…

Damn, they had a geek test on the same page, so of course, I had to take it too.

Wez Furlong suggests to use RBL to reject blog spam.

I submitted my collection of spam to Spamhaus's Zen service, but only 40% of my spammer's IP addresses were recognized. Checking for the existence of the URLs in these spams takes much more time, since most of them don't exist anymore and you waste time waiting for DNS to timeout. Testing with the last 40 ones didn't return any match, maybe they are too recent to be already in the RBL.

Stéphane, auteur du blog Teamasters, a récemment acquis une théière en argent massif. Au vu de nombreuses personnes dubitatives quant à l'intérêt réel d'une théière de ce type, il a étudié le refroidissement de l'eau que l'on verse dans la théière, comparant argent, terre cuite et porcelaine. J'ai tenté aujourd'hui de reproduire l'expérience. Comme je n'ai pas de théière en argent (je ne suis pas encore fanatique à ce point), j'ai utilisé un gobelet en argent massif, un gaiwan blanc en porcelaine et un récipient gradué en inox. Les mesures sont faites sans couvercle.

Conditions initiales

• température de la pièce (et donc des récipents) : 22 °C
• température de l'eau au sortir de la bouilloire : 98 °C
• volume d'eau : 1 dL

Résultats

Pour chaque récipient, on donne le matériau dans lequel il est fabriqué, la masse du-dit récipient, la chaleur massique du matériau, puis la température de l'eau au bout de 10 secondes, une minute et cinq minutes.

Discussion

Le thermomètre utilisé refroidit assez vite, mais se réchauffe plus lentement. Obtenir une mesure précise à 10 secondes est plutôt hasardeux s'il est plus froid que l'eau que l'on verse. J'ai donc décidé de le réchauffer dans la bouilloire (ce qui permet aussi de prendre la températude de l'eau « bouillante »), de le transférer rapidement dans le récipient, puis de verser l'eau par dessus. Par ailleurs, dix secondes est un laps de temps très proche de celui nécessaire au versement de l'eau, donc l'instant zéro est plutôt mal défini (sans compter que je n'ai que deux mains, donc tenir la bouilloire, le thermomètre et lancer le chronomètre en même temps est difficile).

On voit cependant dans les chiffres que l'acier inoxydable semble être un meilleur matériau que l'argent, mais la diffrérence importante de masse entre les deux récipients rend la comparaison assez difficile, étant donné qu'il faut aussi estimer la vitesse du transfert de chaleur entre l'eau et les différents matériaux, qui dépend des matériaux et de la surface d'échange entre eux. De plus, les récipients ayant des tailles différentes, il y a pour chacun une partie qui n'est pas en contact avec l'eau, mais qui va aussi absorber de la chaleur par conduction thermique, la vitesse d'absorption dépendant alors entre autres de l'épaisseur de la paroi du récipient.

Cela dit, si la chaleur massique du matériau est effectivemnt la clé du problème du refroidissement de l'eau dans la théière, alors une théière en or massif serait encore mieux adaptée, puisque sa chaleur massique est environ moitié de celle de l'argent. Elle serait en revanche extrêmement sensible aux chocs…

Sources

Chaleur massique: http://www.alyon.org/InfosTechniques/chimie/chaleur_massique_diverses_substances.html, pedagogie.ac-toulouse.fr/biotech-sante-envir/STHE_Equipements_partie_1.pdf page 14.

In the kitchen next to my office, there is a giant flatscreen TV. Below it, there is the TV's cardboard box, throwing to my unconsenting eyes cool new technology acronyms such as 40 LCD, HD Ready, HDMI, Dolby Surround and BBE Digital.

I wonder if other people are as offended as I am by the advertisement of consumer-alienation technologies as above. Then I decided to design my own set of cool acronyms to add to these devices. It could be something like Enhanced EPC (Electrical Power Consumption), AO (Already Obsolete) Ready, UYA (Up Your Ass) Connector, RCR (Restricted Consumer Rights) Technology, PPP (Pay! Pay! Pay!) Enabled. I'm pretty sure I could sell stuff bearing these.

Acronyms are so cool.

© Ultimate Ears

Après moultes péripéties avec le vendeur en ligne, j'ai fini par recevoir mes nouveaux écouteurs. Contrairement aux précédents (Panasonic, livrés avec mon vieux lecteur de CD portable) qui se plaçaient dans le pavillon de l'oreille, ceux-ci se placent dans le conduit auditif, bloquant effectivement les sons extérieurs. En revanche, tous les sons du corps se font remarquer : les cervicales qui grincent quand je tourne la tête, un grattement de l'ongle sur le nez, la respiration, les battemements du c½ur… Le port des écouteurs n'est pas à proprement parler douloureux, mais il est bien plus intrusif que les précédents et quelque peu gênant. Peut-être que cela disparaitra avec l'usage ?

Du point de vue de la qualité sonore, les Super.fi 3 Studio sont sans commune mesure avec les anciens écouteurs. Bien que certains leur reprochent un manque de basse, ils ont un rendu des basses pourtant bien meilleur et semblent reproduire les sons en général de manière plus précise. Je soupçonne cette qualité accrue de provenir tant de la qualité du transducteur que de la position dans l'oreille et l'isolation supplémentaire qu'elle apporte. Un inconvénient cependant : positionner correctement l'écouteur dans l'oreille s'avère difficile, surtout lorsqu'on a comme moi des conduits auditifs tordus, et le positionnement a une influence sur le rendu sonore.

Enfin, le contenu du paquet ne correspond pas tout à fait à ce qui est annoncé sur le site Web de Ultimate Ears et dans la notice : j'ai eu en cadeau bonus un atténuateur (à utiliser dans l'avion, à en croire le petit logo dessiné dessus) et une paire d'adaptateurs en silicone supplémentaire, à double rebord (de toutes façons trop gros pour mes oreilles). Cadeau du marchand ou erreur du fabriquant ? Je ne vais pas me plaindre :)

Thanks to Niko, my Leffakone project is finally complete. He built the timer I designed already two and a half years ago, that enables Leffakone to boot automatically when it is time to record a TV show.

The device is based on a Microchip 16F84A PIC and is connected to the Wake On LAN connector of the computer, as well as to the parallel port. The WOL connector provides +5V to the device even when the computer is powered down. The parallel port is used for sending the 24-bit timeout value to the timer, one bit at a time (one line is data, another line is a “data available” signal). The timer then ignores all input for three minutes, giving enough time to shutdown the computer (when the computer shuts down, some garbage is sent to the parallel port, which would be considered as input if it were not ignoring it). After that delay, it counts down to zero, and then sends a signal to the WOL connector that boots the computer. It's as simple as that.

The device has two LEDs, one indicating that power is on, and one status LED. The status LED is:

• On when the device is in standby, waiting for input.
• Off when the device has received partial input and waiting for more.
• Blinking at 0.25 Hz when ignoring all inputs.
• Blinking at 0.5 Hz when counting down.

When the timer is in the process of receiving input, and the procedure does not complete, the internal watchdog timer makes it reboot after about two seconds, putting the device back to standby mode.

The timer is supposed to accept timeout values in seconds, but the programming contraints in the device are such that the actual time unit is worth about 1.02 seconds.

For years I've been especially following three TV series: Murder, She Wrote, MacGyver, and Lois & Clarke: The New Adventures of Superman. All three of them came to an end within a couple of weeks. The last episode of Murder, She Wrote aired on February 28th, the last Lois & Clarke on March 3rd and MacGyver March 17th. I feel empty, like I'd have lost several friends at once, especially since I had been watching Murder, She Wrote assiduously for over three years.

Thankfully, the wonderful world of television has brought relief to my pain through the remarkable concept of rerun: Jessica Fletcher still shows up regularly in my set from season eight, as do Lois, Clark and MacGyver, since the two shows are broadcast again from the start! This allows me incidentally to see the first six or seven episodes of Lois & Clarke which I missed last time, and to re-record the episodes of MacGyver which are incomplete in my collection.

© Tara Expeditions

« Le voilier scientifique polaire Tara a commencé en septembre 2006, une dérive arctique qui durera deux ans. La coque arrondie et plate de Tara lui permet en effet de résister aux pressions extrêmes exercées par la banquise et de se laisser porter par elle.

L'expédition, dirigée par Etienne Bourgois, s'inscrit dans le cadre de l'Année Polaire Internationale (API) 2007-2008. Elle est un partenaire majeur du programme européen scientifique DAMOCLES. Programme d'envergure qui réunit plus de 45 laboratoires développant un système d'observations et de prévisions à long terme de l'Océan Glacial Arctique. Ceci afin dévaluer et prédire les risques et les impacts des changements climatiques sur notre environnement.

Au carrefour de la science, de la technologie, de la pédagogie et de la communication, Tara Arctic est une grande aventure humaine qui a la volonté de sensibiliser les citoyens de la planète sur l'importance des équilibres écologiques. »

(source: Tara Expeditions)

Sur France Inter, ils n'arrêtent pas de parler de l'API, alors forcément j'ai fini par en entendre parler aussi. Et puis, au détour d'une interview dans l'émission du 3 mars de CO2, mon amour, j'ai appris que je suis encore plus près du pôle que je le croyais (86°14'41" aujourd'hui).

2006-07-25 11:24:57 -> 2007-03-11 14:03:25

Browsers

• MSIE: 4074 (62.7%)
• Firefox: 1855 (28.5%)
• Safari: 246 (3.8%)
• Mozilla: 154 (2.3%)
• Opera: 143 (2.2%)
• Konqueror: 20 (0.3%)
• Lynx: 8 (0.1%)

OS

• Windows: 5769 (88.8%)
• Mac OS: 382 (5.9%)
• Linux, *BSD: 260 (4.0%)
• ???: 89 (1.3%)

L'aspirateur nasal est un appareil qui permet de moucher un enfant qui n'a pas encore appris à souffler par le nez. Les appareils les plus simples sont constitués de deux tuyaux connectés à un réservoir. On place un tuyau dans le nez de l'enfant et on aspire avec la bouche par l'autre. Pour éviter les remontées de microbes vers la bouche sous forme d'aérosol, le tuyau « bouche » est muni d'un filtre. Ce filtre absorble les liquides et finit par se boucher, en particulier si un nettoyage malheureux sous le robinet le noie… Ne reculant devant aucun sacrifice, et plutôt que d'acheter un nouvel aspirateur, j'ai supprimé le filtre, ce qui augmentait le pouvoir aspirant de l'appareil, mais laissait tous les microbes venir dans ma bouche. Voila ce qui arrive quand on veut recycler du matériel prévu pour un usage unique (enfin je suppose que c'est à usage unique).

Ceci m'a conduit à concevoir un aspirateur nasal à filtre amovible, nettoyable et remplaçable, pour un coût minime. Le montage schématisé ci-contre utilise un second aspirateur qui jouera le rôle de filtre sous forme de flacon de lavage. Le liquide peut être de l'eau, qui diluant l'aérosol diminuera la probabilité qu'un microbe remonte vers la bouche. On peut imaginer utiliser un antiseptique à la place de l'eau, tel du vinaigre, de l'alcool médical ou la gnôle infecte qui traine depuis des années au fond du placard.

Donc un jour où j'aurai pas la flemme, j'irai à la pharmacie acheter un deuxième aspirateur nasal et bricoler le biniou ci-dessus.

La recette d'hummus trouvée sur le web. Je l'ai faite plusieurs fois, sans tahini (parce j'aime pas trop ça).

Aujourd'hui, c'était Mardi-Gras, alors j'ai fait des churros au bureau (histoire de changer un peu des éternels laskiaispulla finlandais ou des beignets de carnaval français). La recette, je l'ai trouvée dans ce blog. D'après Niko qui a mangé des churros en Espagne il n'y a pas longtemps, le chocolat chaud (dont la recette est à la fin de l'article) devrait être plus épais pour mieux coller au churro. Il faudrait ajouter davantage de fécule ?

Et aussi, faire de la friture dans la cuisine du bureau, c'est le bazar parce qu'il n'y a jamais les bons outils pour nettoyer (déjà que j'ai déplacé la moitié de ma cuisine pour avoir tout le matériel nécessaire, j'ai dû prendre le panier à linge pour tout transporter). Évidemment, si un churro expériemental de calibre 20mm n'avait pas explosé en déversant plein d'huile bouillante sur la plaque de cuisson, le plan de travail et le sol, il y aurait eu moins de choses à nettoyer.

Une chose intéressante à noter, c'est que l'huile après cuisson (que j'ai reversée dans les bouteilles vides pour la jeter à la poubelle) occupait presque le même volume qu'avant. J'en ai déduit que mes churros devaient contenir entre 2 et 5% de matière grasse. C'est pas si lourd que ça finalement ! Mais c'est le chocolat chaud qui m'a tué : 7 heures plus tard, j'ai toujours le ventre plein.

J'ai remplacé les trois ampoules de 40 W (410 lm) du salon par des ampoules basse consommation de 11 W (700 lm) de marque Megaman Liliput Plus. La couleur est un peu différente, mais on s'y habituera. En revanche, vu qu'une ampoule fluocompacte de 11 W équivaut à une ampoule classique de 60 W, on y a gagné en luminosité : 71% de lumière en plus, pour 73% de consommation en moins. J'ai choisi la 11 W plutôt que la 9 W (équivalente à une ampoule classique de 40 W) parce qu'elle était moins longue (bien que plus large) et dépasse moins du lustre.

Évidemment, chaque ampoule basse consommation coûte 14 EUR, mais elles ont une durée de vie annoncée de 15 000 heures (à comparer aux ampoules classiques qui ont une durée de vie de 1 000 heures, c'est à dire environ un an). Au total, mon ampoule me coûte .93 EUR par an, ce qui est légèrement moins que le prix d'une ampoule classique. De plus, à 0,10 EUR du kWh (ils ont augmenté le prix de l'électricité de 20% d'un coup cette année les vaches), mon ampoule basse consommation me coûte 1,10 EUR par an, contre 4 EUR par an pour l'ampoule classique. C'est finalement peu par rapport au budget électricité annuel.

© eu.art.com

I watched an episode of Lois & Clark: The New Adventures of Superman yesterday evening, were they managed to track back a radio signal (sic) to a point located at 47.55 degrees of latitude and 38.69 degrees of longitude. Thanks to Google Maps, I know now that this place, which according to the series is the headquarters of the Daily Planet in Metropolis, is actually located in Russia, north of the Black Sea, not far from the Ukrainian border, in a region full of fields. Superman is therefore a Mujik.

However, since they didn't mention if the latitude in North or South, nor if the longitude is East or West, I checked the three other possibilities, which all are in the middle of the sea. If Superman is not a Russian farmer, then he's completely soaked.

Un peu de froid c'est bien, mais là, franchement, c'est abusé. Pour ceux qui ont la flemme de cliquer sur l'image pour la voir en grand, le thermomètre indique -36 °C. La photo a été prise ce matin à 8h de la fenêtre de la chambre à coucher. C'est bien d'avoir un thermomètre visible au saut du lit, ça annonce tout de suite la couleur. Mais pas de panique, ça doit remonter dans la journée, avec un pic à -25 °C.

La voiture a à peine réussi à démarrer (après avoir bêtement calé, la première fois c'était plus facile), on a l'impression que l'embrayage est lubrifié avec du goudron et le siège est un cube de glace.

© Posti Oyj

Palokan posti muutti Euromarketiin muutama kuukausi sitten. Nyt ei enää ole postitoimistoa, vain postiluukku Euromarketin infotiskin vieressä. Siellä ei ole enää postineitejä vaan Euromarketin työntekijöitä, jotka palvelivat myös infotiskillä. Hyvä juttu on, että aukioloajat ovat 7-21. Huonot jutut ovat, että…

Ensimmäisellä kerralla halusin lähettää painavan kirjen Ranskaan, mutta työntkijä ei tiennyt missä laatikossa postimerkit olivat. Toisella kerralla, hän unohti kertoa postinkantajalle että minulle oli saapunut paketti, ja minulla oli pakko mennä sinne ja valittaa. Kolmannella kerralla, minulla oli ilmoituskortti, mutta työntekijä ei löytänyt pakettiani, koska hän oli liimannut siihen väärän viivakoodimerkin ja laittanut pakettin väärään paikkaan.

Mitä tapahtuu seuraavalla kerralla?