No Food for Thought

Are you migrating to a new operating system (Windows, or quite possibly any) and failing to setup your Gmail account in Thunderbird (version 68.1.2, and surely many more)? Getting a small popup which says "Authentication failure while connecting to server imap.gmail.com.", in the bottom right corner of the window, with only an 'x' to dismiss the message? Despite Thunderbird restarts? And even OS restarts?
If you're like me, you're in luck, the solution is very simple : just connect to Google. That's right - open a browser, go to Google, connect. And then just retry connecting in Thunderbird (or restart it).
And that is how intuitive it is to use a top 2 MUA with a top 2 mail service. An MUA from an organization who's done tens of millions of USD-s in business cooperating with that service's provider. Congratulations, Google
Welcome to 2019

P.S. For those less lucky, see the "Less secure app access" section in https://myaccount.google.com/security
I figured this one out thanks to this long thread (get ready to click "more" several times to reach a useful reply).

Temperature is the topic of much debate. Even inside temperature is a topic of disagreement.

But this disagreement has just gotten a lot less light, and should get much more heated. Indeed, inside air temperature no longer matters only for its ecological impact, but also for its direct economical impact, according to new study Battle for the thermostat: Gender and the effect of temperature on cognitive performance, as this TIME article reports.

Here's hoping that more research on this issue will come soon.

My good old and short dynamic domain name ido.ath.cx which I used to point to my home PC stopped working after DynDNS stopped being free. Instead of looking for another free provider which could also stop being free at some point, I decided to use a subdomain of my domain philippecloutier.com, which I bought from GoDaddy. Unfortunately, GoDaddy does not offer dynamic DNS, so on my friend Xavier's suggestion, I signed up for free to DNSEXIT and changed my nameservers for DNSEXIT's.

I then simply needed to setup my router to update the IP. OpenWrt allows that using ddns-scripts. DNSEXIT's website is a little amateur, but configuration was not difficult. I was surprised to notice at the end that DNSEXIT asked for a link to its website since I use free DNS. I do not wish to make an electronic payment just for dynamic DNS, but I am not entirely comfortable using such a service for nothing, so... here is your link:

The other surprise was to notice that OpenWrt's DNS update method for DNSEXIT appears to send the password unencrypted.
Meh. Not a panacea, but good enough for me.

I bought this U-lock last year to upgrade from a flexible cable lock. I expected the lock would be heavier, but I didn't expect so many disadvantages.

First, the mount bracket is really poor. I mounted it on the vertical bar, and it's at least the second time I need to reinstall it, since it's hard to tighten enough. The provided hex key is crap, even though Kryptonite claims it will do fine. The short segment is so short that it won't be possible to use the long segment as handle. One needs a real 3 mm key to install this properly. I even attached a rubber band to the frame to increase friction. Even though I tightened it more than once as hard as possible, I lost hope that it will even hold for good. Vibrations will eventually require a new intervention. Kryptonite warns you shouldn't tighten too much, because some frames are too fragile! Think you're lucky enough since your frame is not carbon? Still, tightening too much will cause the bracket to become distorted and the spline to get ten times harder to insert or remove. They advise to check tightness daily or weekly! Who has the time for that? If you install this vertically, expect to waste at the very least 1 man-hour on the mount over its lifetime.
Moreover, the spline provides flexibility, but it increases the space which needs to be reserved for the lock. In the end, with my bike (standard size for an adult male), I can't install this lock in the inside of the frame without losing the possibility of installing a bottle mount.

Then there's the lock's keys. The keys work fine, but they have a thick plastic "handle". This is not a problem if you have keys already, but if you have no other keys, the key means you have one more thing to carry in your pockets. It's hard to leave your house forgetting the door key, but if you just have a lock key, it's very easy to forget it, and obviously often quite a problem when it happens. It would be really simple to avoid that by keeping the key in your wallet, except since the key's maximal thickness is about 5 mm, that probably won't be an option. I believe Kryptonite should ship 1 thick key and the other key should be thin.

Finally, there's the cable. I chose this model because the cable reassured me that I wouldn't lose the flexibility of cable locks. The cable adds much flexibility, but its usability is bad. In real life, you won't use that cable often. But when you're rolling, where do you put it? Kryptonite has no answer to that. If you put it on the lock, it will slow you each time you need to lock your bike. But if you don't, good luck finding another place. What I ended up doing is tying it around the handlebar, but tying it there properly requires about 1 minute each time. It's not easy to avoid a conflict with reflectors and other stuff on the handlebar. So when you don't carry it tied to the U, most of the time you end up with an unlocked cable on a locked bike. I was lucky enough that no one stole it. Yet...

Ah, and that is probably not specific to this lock, but manipulating this is dirty. I wash the lock at least monthly, and I still check my hands after every time I lock it. Also not really specific to this lock, but you need space in your frame's triangle to fit such a U-lock. I was about to order a second bracket so I could use this lock with my Garneau Cityzen Sub-0 when I realized I would need a much smaller lock to fit such an open frame (and such a small lock would be even harder to use).

The only reason I don't recommend another lock is I've never bought any other U-lock. And unfortunately, I remember spending several hours, reading several reviews, before determining which model I should buy, so... good luck!

Il y a quelques semaines, j'ai eu le privilège d'assister à une promesse électorale plus grande que nature. Une promesse de la taille d'un rhinocéros en chef.

En effet, j'assistais au discours inaugural du chef du Parti Rhinocéros du Canada, Sébastien CoRhino Corriveau, dans sa circonscription de Québec, lorsqu'il promit que pour chaque promesse faite par un parti concurrent, le parti rhinocéros allait renchérir en promettant le double. Une promesse qui saura sans aucun doute attirer la clientèle la plus sélective.
Bravo aux rhinos, et que les résultats de la campagne soient à la hauteur de leur innovation!

A couple of weeks ago I was disturbed by a colleague while debugging an amateur 15-year old document parser, equipped with a badly buggy debugger. I was very displeased, but I realized I couldn't blame my colleague; I rather realized that now that I work in an open office, I need a way to indicate to colleagues when they shouldn't disturb me.

So I searched for devices which would allow me to turn on some red light when I need extreme focus. I quickly found Plenom's kuando Busylights. The hardware seemed great, and the price was right. I was about to buy when I realized the lights had a single year of warranty. Which made me question durability. I can afford shopping and setting up a device once, but I can't do that every third year.

I saw that Plenom offers a manual color control application for Microsoft Windows and "Mac / OSX". But I realized there was nothing for GNU/Linux. In addition, the source for the application wasn't provided. At the bottom of the download page was a reference to an interface specification:

At that point, I was disappointed to see that Plenom didn't offer any code nor support for GNU/Linux, but thought that with its SDK and interface specification, Plenom was close to an acceptable level, and figured that if Plenom was OK with it, I could patch this small flaw by publishing the specification on this website. So I sent the following message to Plenom:

Plenom courteously sent the following reply:

(The mail included documentation, but I cannot provide it here.)

I was disappointed that Plenom didn't offer redistributing, but found Plenom's concern justifiable, and its reply very courteous, so I tried finding a compromise with the following reply:

To my surprise, the next thing I heard from Plenom was a message in my voice mailbox (even though I hadn't provided my phone number), from Mitch Friend, president of Plenom Americas, who said he wanted to talk about my development project. Duh

I called back anyway and started by basically repeating my last message to M. Friend, explaining that I didn't work in a call center and would be paying from my own pocket. The funny part came when M. Friend reassured me that his company was in great health, so there was no reason to fear bankruptcy. He claimed I was the only one who had asked about this so far, visibly trying to convince me I was at the faulty end of the conversation. Then came the worst part. M. Friend asked if I would do the same with Microsoft and ask them to change their policies. I was caught off-guards and failed to point out that I wasn't asking Plenom to change its policies, or that most of Microsoft's hardware implements the HID protocol, or even that Microsoft had interest in keeping the cross-platform support of devices as low as possible.

The conversation certainly didn't help convincing me to get a kuando Busylight, but it had a bit of constructiveness when M. Friend mentioned there was some Busylight-related code available on GitHub.

Ultimately I didn't get any permission to redistribute the documentation, nor any further explanation of Plenom's apparent unwillingness to help itself. I even realized after that Plenom requests personal information just to let you download their end-user software. At that point, it seems safe to conclude that Plenom won't offer either redistributable source code or interface specification for kuando Busylight anytime soon.

After giving up on kuando Busylight, I searched a little for alternatives. I found a couple:

1. Jabra Busylight
2. Embrava's Blynclight

The first is not USB and apparently in a very different category, specific to phones, so not an option for me. As for Blynclight, it seems worse than kuando Busylight. There is visibly no support for GNU/Linux, no source for any controlling application offered, nor even any specification

So, if you're aware of a well-working and reliable availability display device, please let me know. (Meanwhile, if you see me stepping through a stack tens of levels deep, feel free to find someone else to discuss your backyard.)

Early in my eleventh year, I fell in a painful love for the first time. My parents bought me a Pentium 120 MHz machine, my family's first personal computer. Instantaneously, my reading time dropped by 98%, and all of my spare time (or more) went to this buggy, unstable, very limited, yet how-irresistible creature.
A tragedy since, like any love, it takes the place from other possible loves. And a tragedy since, like any real love, it can never be completely fulfilled.
Before this tragedy started, I thankfully had the time to study and read a lot and become excellent in maths, science, French... and any school topic which didn't require creativity So I could pick any high school I wanted, and my parents sent me to the International Education Program, which - at the time - was still only accessible to the elite. I learned I was accepted in Rochebelle high school's International Baccalaureate program from the acceptance paper my mother had wrapped in the Christmas tree.

In spite of my mother's efforts to present this to me under a positive light, during all the months before I started at Rochebelle, as I imagined how it would be, I anticipated one downside. Not the higher academic expectations we would have to meet, but the less pretty girls I thought I would study with. In my 11-year-old mind, the intelligent girls which were selected had to wear glasses and be physically unattractive. Either I developed this preconception from the girls I saw at the admission exam, or - and I have no doubt it's the latter - I had already been exposed to enough Western culture to associate geekness with lack of physical attraction. My preconception was right in that both of my elementary school ex-"girlfriends" were in the regular program, but thankfully, there were more than enough interesting girls in my program too.

During my first grade, I would spend my evenings at my friend Guillaume's house, since he not only had a computer, but was lucky enough to have dial-up Internet access. At the time, we had to connect to the Internet, and to do so, we needed to enter our ISP password. One day, I saw Guillaume enter his password in the wrong field, unobfuscated. It was hard not to remember that his password was "12345". Even though Guillaume was my best friend at the time, some time after, while I was home, I stole his password so I could connect to the Internet from my home (ah, the good old days of dial-up). I probably never even thought that my own parents could catch me when they would notice that the phone line was busy. But it's Guillaume himself who caught me while I was using his account to use the irresistibly cool Palace chat application.

Thankfully, my theft had unexpected consequences. Instead of rejecting me, Guillaume remained my friend, and his mother accepted my family's proposal to share his Internet connection. I would bring them 15 CAD per month so we could use their connection, and we shared for a few months, until we got sick of having to synchronize who could connect or not all the time and each got our own.

Long story short, when I entered high school, my mind was constantly thinking about the Internet and several interesting girls. Girls to whom I didn't dare to disclose my attraction, and who - I thought - may also have been keeping secret reciprocal interests.

During my first years in high school, I had this idea of a system allowing all students to secretly indicate who they were interested in, which would tell 2 students if they were interested in each other. There were design difficulties though; how would the system ensure that interest was genuine, and not just a dummy indication someone would enter to know who was interested in them?

Fast forward many teenager crushes, even more computer crashes, and many less tricky ideas which would replace that one in my dreams. With the advent of social media, this kind of system has become much cheaper to develop. So much so that in fact, a couple months ago, Facebook started deploying one, called "Secret Crush". Which got me aware of the fact that Orkut had done it last decade.
Welcome to the future, the geek paradise! (Just too bad it's 20 years later and I've forgotten with whom I was hoping to use it)

Microsoft's electronic book business wasn't profitable enough. So Microsoft is closing its DRM servers:

Philippe, happy not to own (if that word still makes sense) any DRM-encumbered book

7,99$ pour un sac de pommes sans pommes, ça fait cher :

Marketing qui ne vaut pas cher au supermarché Provigo, sur l'avenue Myrand

One of the things I remember best from my high school and college days is the communist pantry. Back then I would share a locker at school with friends to store shared food. We would buy several types of non-perishable food appreciated by teenagers and store it in that "communist locker". In these good old days when we took our health as granted, one of my favorite contributions was packs of Pringles chips. One pack would sell for 1.99 CAD and weight 200 g.

Shortly after, I was disappointed to see the contents diminished from 200 g to somewhere around 170 g. I suspected that Procter & Gamble would eventually increase quantity back to where it was increasing the price. But I didn't suspect just how premeditated the move was.

Before increasing back, Procter & Gamble (or the current owner, Kellogg's) would reduce again, to 148 g. Without changing the can's size at all. And it's only last weekend that I could finally see the result of this well-planned crime, when I saw a tall Pringles can called a "Mega pack", with 30% more Pringles. This can now sold for 2.75 CAD (as a promotion) contains a Mega stack of 194 g of Pringles - which is less than the original can.

I now wonder what Kellogg's next moves will be. One possibility is that the can keeps growing by 30% every 15 years. In that case, I would estimate that Pringles cans can no longer fit in communist pantries in 2 to 4 centuries. But there's another avenue which I suspect Kellogg's considers. Instead of restarting such an inflationary cycle again, they could market a communism-friendly and "eco-vegan Pringles can", with the ecological achievement of reducing the packing material by 25%.

I will not make an exact prediction about what Kellogg's is premeditating, but with climate change, one thing is sure - the stackstakes are higher than ever.

P.S. This is written with all respect for Procter & Gamble, which used to sell the most dense pack of chips on the market.

2020-12-23 Update: Kellogg's is now making Pringles cans eco-friendly by reviewing their composition.
2023-11-12 Update: The Mega pack is now called the Party stack, and now sells for 3,57 CAD. And rampant inflation has gotten the practice Procter & Gamble pioneered a name: shrinkflation.