Thursday, 19 March 2015

10 Stupid Questions

Here's a list someone composed of actual job interview questions from some big employers. Some of these questions are just silly, others are the kind of dumb crap I really hate. Here's a quick review:


A note on ratings: I always worried about being a TA for a philosophy class because I would be reading the dumbest things I'd ever read and giving them B+'s. Keep that in mind.

1. If you could high-five one person, living or not, who would it be?
This is a goofy question that I can't quite understand the point of. If someone thinks this reveals something about the person answering, then I don't know if I want to work for that person. My answer, "You, right now" (if they won't do it, just leave the interview). My rating: B+

2. If Labatt gave you a million dollars with the condition that you'd have to pay it back in three years, what would you do with it?
Assuming this is for a position where a person has to actually make investment decisions, it could be a good question. But for regular people I think it badly conflates risk aversion, financial savvy, current capacity to take a loss, and ability to properly estimate current capacity to take a loss. Too many variables.

My answer, "Figure out how much interest I could get on it with a recognition that I can't afford to lose the principle." My rating: A

3. If you only have one choice, do you prefer to work on time or correctly?
What a stupid, stupid question. What the hell are we talking about here? Are we talking about selecting the correct cables to hold up the suspension bridge or are we talking about making sure there are no typos in our report? And how badly does someone need the report, and by how much are we going to err on those cables?

My answer, "It would depend completely on the circumstances, sometimes errors are inconveniences and sometimes they cost people their lives. Sometimes missing deadlines is an inconvenience and sometimes it costs millions of dollars. There is no meaningful way to figure out which is more important in general." My rating: D

4. How can we move Mount Fuji?
Come on now.

My answer: "Well, the first thing to do would be to start negotiations with the government of Japan and locally with the people who would be affected. Moving an active volcano seems like a pretty big risk of completely devastating the island it is located on, so you might be talking about razing Tokyo. Once you've convinced millions of people that this project is worth it and figured out where they are all going to go if the largest city on earth is buried in lava, you obvious have millions of crazy people willing to help you. So at that point, you move it the same way they built the pyramids, chunk by chunk with a never-before-seen amount of manpower." My rating: F

5. If you were an inanimate object, what object would you be?
I mean, the clear answer is 'a corpse' because if I were an inanimate object right now, that's what I'd be, but presumably that isn't what they are asking, and it wouldn't go over well.

My answer: "A corpse." My rating: B+

6. What would you name a book about your life at this point?
I think if you have an immediate answer to this you are either in the midst of writing your memoir or you are an insane narcissist. Maybe that's the point of the question. If you don't hem and haw and say, "Wow, I don't know" then they should stay away from you.

My answer: "I don't think it's a good idea to pick a title for a book before it is written. I'd probably want to make a clever pun about Borderline Personality Disorder." My rating: C

7. How do you get a plastic ball out of a cup without touching the cup?
This is even worse than moving Mount Fuji. This kind of question tells you that they are looking for a particular answer and will judge any other answer to be a violation of the unstated rules of the question. This question tells me not only that I don't want to work for the person asking it, but also that I don't want to do business with the company.

My answer: "I'd just reach in and take the ball out with my hand without touching the side." Their reply: "Uh, you can't, the ball is too big." My answer: "Oh, okay, I'd just let the ball fall out because the cup is upside-down." Their reply: "Stop being a dick." My reply: "Lowering something sticky in, cutting the cup in half, shattering the cup, tilting the surface the cup is on, asking someone else to get the ball for me, blowing the cup over, pushing the cup over with a stick, putting several more plastic balls in until the cup is full enough of them that I can take the top one out and noting that you only asked me to take 'a' plastic ball out of the cup without specifying that it had to be the ball already in the cup, using similar logic to take a plastic ball out of a bigger cup that I can just reach into that I brought myself, tongs, bringing the cup into a zero gravity environment and waiting for them to drift apart, filling the cup with water so the ball floats out - it was that last one that you wanted, wasn't it?" (Seriously, my guess is that they are looking for 'fill the cup with water') My rating: Fuck off, you are not clever.

8. If you own a grocery store and you have to sell apples, how would you determine how many different types of apples would you sell?
This seems like a pretty good question that will actually tell the asker how you approach unfamiliar situations.

My answer: "Since I don't know anything about the grocery business, I'd probably start by looking at other grocery stores and trying to carry the same kinds of apples they do. Once I had some appled in the store I could look at how each kind sold, and occasionally experiment with other kinds of apples to see if they were valuable additions. Of course, the amount of time and effort I put into these experiments would depend on how much of my business actually depended on apples - I wouldn't want to spend a disproportionate time make decisions about apples that end up having a negligible effect on my business." My rating: A

9. How will you keep HootSuite weird?
This isn't even an oddball question. Apparently HootSuite is a business that depends on being weird. This is the equivalent to a "How would you keep our business efficient?" for a company that values efficiency.

My answer: Who knows, but obviously I had better have one if I am applying to work at HootSuite. My rating: Shouldn't be on this list.

10. How many traffic lights are there in Canada?
A simple question to see if you approach problems.

My answer: Step one is to define how much time they really want me to spend on this question in the interview. Step two is to define the terms (is a 'traffic light' an entire intersection that is equipped with lights, or is it a single lamp?). Step three is to estimate how many traffic lights are in Toronto based on my knowledge of the city (how many major blocks there are east to west and north to south), and scale that from Toronto's 2.5 million people to the population of Canada. Obviously Toronto is an exception, but I don't even know which way it is an exception. There are way more traffic lights but also way more population density than in rural areas. If Toronto has a thousand intersections with traffic lights, that would mean one set of lights for every 2.5 thousand people. Does a town of 10,000 people have 4 traffic lights? Does it have more or fewer? I really don't know. At any rate, I'd guess Toronto has more than a thousand but way less than ten thousand, so I'd put Canada in the 20,000 range and be pretty confident I'm within an order of magnitude. My Rating: A

Tuesday, 17 March 2015

Health and Physical Education

Ontario released its updated Health and Physical Education curriculum a few weeks ago now. Last time they tried to update the curriculum they had to put it off because people were outraged by the idea that kids would be taught about things like masturbation and homosexuality. I'm not going to tell you that the document isn't hilarious, because it is:
Teacher prompt: "Sexting – or the practice of sending explicit sexual messages or photos electronically, predominantly by cell phone – is a practice that has significant risks. What are some of those risks? What can you do to minimize those risks and treat others with respect?"

Students: "Photos and messages can become public even if shared for only a second. They can be manipulated or misinterpreted. If they become public, they can have an impact on the well-being of the persons involved, their future relationships, and even their jobs. There are also legal penalties for anyone sharing images without consent." "You shouldn’t pressure people to send photos of themselves. If someone does send you a photo, you should not send it to anyone else or share it online, because respecting privacy and treating others with respect are just as important with online technology as with face-to-face interactions."
Ontario Health and Physical Education curriculum, page 196
That's how Grade 7 students talk according to the Ministry of Education. The examples of Grade 1 students include at least one 44 word grammatically correct sentence with no contractions. These are bizarre reversals of Plato's vision of Socrates talking to moronic straight men in the forum where we have the student being some kind of perfect robot for reciting the truth.

But stupidity of the text aside, let's take a moment to consider the sexual radicalization that will soon be foisted on our youth.

This, of course, is the physical education curriculum as well as the health curriculum, so a lot of the document is focused on developing motor skills and getting exercise. What we really need to pay attention to is the Healthy Living section of each grade, specifically the statements in the learning summary under the heading "Human Development and Sexual Health."

Let's see, in Grade One our children are expected to be able to name their body parts. In Grade Two they are expected to be able to outline the stages of human development. In Grade Three they are expected to be able to describe the characteristics of a healthy relationship, like listening, showing respect and being honest.

So by Grade Three our children are hopeless corrupted already. And the juicy stuff is yet to come.

In Grade Four they are expected to be able to describe the changes that happen during puberty. In Grade Five they are supposed to be able to identify the parts of the reproductive system and describe their function. In Grade Six they are supposed to consider their self-concept and how they will make decisions in relationships.

If that seems pretty obscene, then you might want to cover your eyes as I get to the final two years of elementary school.

In Grade Seven they are going to talk about reasons why people may choose to have sex or not have sex, how people choose what is appropriate for themselves in physical relationships, and identify common sexually transmitted infections. But the real kicker is in Grade Eight when the things they have talked about in previous years are continued and built upon, furthering their thoughts self-concept, decision making in relationships, and supports available to them if they need help.

It seems like the furor over the curriculum has died down pretty quickly. Not before a federal MP told the federal leader to intervene in provincial politics to stop it, and not before people tried to imply that it was created to groom children for sexual predators, but it is mostly out of the new now. I can only imagine the anger and intrigue that will be generated when they reveal the next math curriculum.

Monday, 16 March 2015

Wild Wacky Action Game

I wrote a while ago about making some kind of hybrid Sandcastle Builder / Kittens Game inspired idle game, and I have something that is probably about ready for testing. My biggest problem is that I really can't seem to get the pacing right. In my first iteration it was actually just insane to play, requiring you to constantly swap between tabs and click things. I've drastically reduced the need for clicking and I've increased the length of a time period in the game to make it less frantic and more methodical.

But as it always does, what it comes down to with an idle game is the balance between clicking things and leaving the game alone. Part of my attempt to solve this problem was a hidden time dilation mechanic, where leaving the game alone caused time to slow down until you once against started playing. This meant that leaving the game for 10 minutes had a fairly expected result, but leaving it overnight would produce substantially smaller outcomes than it would initially appear if you just multiplied everything out. I tried removing this when I increased the basic unit of time from one minute to three, but I still feel like too much happens if you leave the game alone for a night. On the flip side, if I increased the time period further, to five minutes, say, then the beginning would be far too tedious.

On the other hand, I also have a special problem that leaving the game alone is too devastating. Sure, things progress while you are gone, but without some maintenance you are likely to come back to find that things have gone terribly and it will take some time to rebuild. I guess that's sort of how I drew things up, but having to rebuild every morning and every day after work is pretty harsh if rebuilding takes several three-minute time periods. Before I release a play-test version I think I may need to add a few "autocast" features to use certain abilities that are key to keeping things going.

I'd like to put some more features in before I show it to anyone, but just like my previous attempt, at some point I feel like I have to release what I have or I will just shelve it forever. Then again, there's always the chance I release what I have and *then* shelve it forever. So far that's what I've done with all but one of the six projects I've put up on this blog. Then again, one hundred percent of the projects I've never shown anyone have gone nowhere, so 16% is not so bad.

Monday, 9 March 2015

SolForge Campaign

It had been a long time since I loaded up SolForge, but when I heard they finally released campaign mode, I thought I might as well go and see how it was.

I'd almost say it wasn't. Campaign mode is six opponents. Four of them are battled with preset decks and are basically unloseable. The last two allow you to bring your own deck and have absurd special abilities and decks to make them difficult to beat.

I understand it will be expanded further in the future, but honestly a menu of computer opponents isn't that much of a campaign. After all this time I would think they'd be a little embarrassed to call it one.

But anyway, the two hard opponents have seriously unreasonable abilities to overcome. The first one gains a very large amount of life. Normally this wouldn't be a big issue but he plays with the dragon that gets big when you have over 100 life. And big is really big in this case, going from an 8/9 rank 1, to a 16/17 rank 2 to a 100/101 rank 3.

The second one kills any creature that damages her. Fortunately the AI doesn't take full advantage of this by leaving all of your creatures unblocked, but this is still pretty absurd.

Making matters worse, both of them put a level 4 creature on the board when they hit rank four or five. These creatures are the first level 4s in the game, I think, and they are suitably insane. One has attack and health equal to your life total and gains you life equal to the damage it inflicts. The other is 7/40 and not only kills any creature she damages, but also kills any player she damages. And she has mobility 1.

This was kind of neat because I had to use decks that were suited to each opponent. For the first I needed to be very aggressive to make sure I kept his life total low. The second I have to avoid relying on keeping creatures in play. Fortunately among the decks I already had built was a very aggressive Spring Dryad deck and a spells deck that burns opponents out. Using these decks I beat the two challenge encounters pretty easily, earning a copy of the legendary cards that go to level 4 from each.

I was having fun with SolForge so I updated my decks a bit and played some games. I think they've had a lot of cool ideas for cards since I last played. There are cards that you only play once and then get removed from your deck, cards that add other cards to your deck, cards that only have one level but scale in their effect based on what rank you are, and other cool variants.

One problem is that I feel like they somehow broke the AI. The Hard Computer AI used to be able to beat me some of the time using my decks when it got a favourable matchup. Playing against it now it seems bascially impossible for me to lose thanks to it doing things like overwriting an 11/7 Spring Dryad with the copy of an Ether Hound and other absurd moves. In the games I played against the computer I rarely even got to rank 3 because the computer played so badly it lost too fast.

I'm not going to play very much SolForge but I don't regret giving the markers some of my money once upon a time, and I wish them all the best. But for the next few days I'll probably bash a few heads and profit. I'll also likely check in on it once to twice a year just to see what else they have come up with.

Wednesday, 4 March 2015

Waiting For People to Die

Came across some Pew Research today via BoingBoing. No need to read, I have it in summary here:

Support for Marijuana Largely a Generational Story

Democrat-leaning members of the boomer generation aside, the pattern is pretty clear. Legalizing marijuana is not so much a Republican vs. Democrat issue in the US, it is an old vs. young issue. I'm willing to go way out on a limb and guess that research would produce a similar pattern in Canada.

If you spend time on internet message boards and comment sections you'll often encounter "tone policing" - criticisms of people based on their manner of expression rather than the content of what they are saying. These are usually coming from people who are supposedly friendly on the actual issue. "Sure that thing is insulting to women," they say, "but yelling about it just makes the people who advocate it defensive, it won't change anyone's mind."

Instead, they say, being conciliatory will change minds, connecting with people emotionally will change minds, even being deferential may be necessary.

There's a lot wrong with that point. For one thing, it's almost always another white man telling non-whites or non-men that if they just all did what he thought was right then everything would get better. The idea that he is going to be given a lot of charity is far-fetched, he probably isn't that one exceptional white man who actually has everything figured out.

Today, though, I want to address the actual Machiavellian argument being made. The argument that we should be kind for our own good for the sake of making people believe us - that this is how we will change people's minds. Look at that chart above. Nothing changes people's minds.

But that isn't the same thing as saying that trying to change things can't work. It can and it does. It's just that if we are going to be Machiavellian about it we need to understand our proper goal. The goal is not to convince people who disagree with you, because you can't convince anyone. The goal is to increase the amount of signal in the public space that is devoted to the issue.

It is obvious that marijuana should be legal, especially in America where the war on drugs has had an incredible toll. If someone things marijuana should be illegal, it is because they haven't actually thought the issue through. But we all have tons of things we haven't thought through. It's inevitable that we mostly accept what our culture gives us and only question a subset of that.

If there is enough interest in an issue that it causes people to question it, then younger people will tend to be more right about it than older people. If there is not, then there is no reason for younger people to be more right.

Last century's heroes of social change are Nelson Mandela, Ghandi and Martin Luther King Jr. All advocated for peaceful resistance and all eventually attained their goals. But Apartheid lasted 46 years; the struggle for an independent India went on for 32 years; and while the civil rights movement in America spanned only 14 years, looking at the criteria for success from MLK's famous "I Have a Dream Speech" only about half have ever been satisfied - the civil rights movement is a name for a particular period of the struggle against racism in America that has been going on for hundreds of years and certainly hasn't ended.

When you look at time periods like 32 and 46 years, it's hard not to think of the change as a generational shift. A new generation came of age that said to themselves, not just that Ghandi was right, or that Nelson Mandela was an admirable man, but that British occupation of India didn't make sense and that apartheid was a bad policy.

Peace hasn't been the only road to success, either. The Irish Troubles lasted 30 years of riots, bombings, and failed ceasefires before people eventually concluded that if Northern Ireland wanted self-governance, it should have it. It wasn't because they thought bombing was a good thing, it was because rejecting the will of a nation for self-governance is just policy and they knew it. Sinn Féin, believed by US and British intelligence to be directly linked to the IRA and bombings, holds seats in the Northern Ireland House of Commons and the European Parliament.

I think peaceful protests are better than bombings because bombings are terrible, but for achieving your political end, I'm not sure either is more effective. Nor is being sympathetic more useful than being outraged in an argument. Whatever ideas you advocate, the Machiavellian thing to do is to figure out if they are on the right or the wrong side of history. If they are on the right side then scream them from the rooftops, get as much bandwidth as possible, and let the next generation make up their own minds. If they are on the wrong side then keep quiet about them so they can go unquestioned.

Actually convincing people of things in arguments is essentially hopeless, and if you want to be effective, it shouldn't even enter the conversation.

Monday, 2 March 2015

I Swore I Heard Something

I've played Skyrim before with some different builds. It seems like the most efficient way to club your enemies to death is to use a club. You can get your damage so high through smithing and enchanting that you can one- or two-shot dragons. I've also played magic-heavy builds that end up relying most on summoning dremora who used clubs.

Directly attacking with magic seems pretty weak, and I didn't play very far into a stealth game, so I decided I'd go with bow and stealth this time.

Leveling up pickpocketing and stealth means you go up levels with skills that don't directly affect combat. That ends up putting you in a tricky spot when the fights are harder. So basically if you want to do a stealth build you have to actually utilize the stealth or you'll get killed by things that just have more health and do more damage than you. Playing with stealth has some pretty comical results.

When you do something noisy, like shoot someone in the back, all the enemies within earshot notice that something is going on, but unless they have a direct line of vision to you, they don't know where you are, only where the noise came from. So they start running around looking to see what is going on. If you remain hidden for enough time they eventually give up the chase and go back to their starting positions and mutter to themselves about not being able to find you.

So I'm up on a little ridge overlooking a camp where five bandits are hanging around. I put and arrow into one of them, killing him outright with my triple damage sneak attack, and back away from the edge so they won't be able to see me. After a little while they give up and head back to where they were standing, and one of them says, "I could swear I heard something."

Yeah, your friend is dead on the ground.

So another arrow goes into another head and another look-around-to-see-what's-up begins. Soon it's back to what they were doing.

I think my favourite was getting a mission to assassinate someone in the middle of a city. The person was leaving a tavern with someone else standing a few feet behind. I waited until they got to a side street so no one else was watching and carefully aimed from behind them to hit the person in front rather than the nearer person. As fatal shots often do, the impact launched the victim forward with rag-doll physics, and they skidded and rolled to a stop a fair distance away.

The person walking because them didn't appear to notice at all, but, upon reaching the dead body, he stopped to say, "What happened here?" He lingered for a moment while I looted the body right in front of his eyes before continuing on his way home.

The stealth system can definitely be frustrating - especially in certain fixed encounters where the enemies appear to cheat and automatically know where you are - but it certainly has its rewards