Coding interviews are not about coding skills, they are about communication skills.
I thought I had discovered a clever hack when I was interviewing. You ask enough closed questions about the problem that the interviewer tells you the answer. Then you repeat the answer back to them, as code or aloud, and you pass.
Now that I am interviewing, I am desperate to find anybody who will ask me enough questions to have me reveal the answer, and then repeat it back to me.
This article leans rather heavily on fallacious arguments.
Taste is a weak sense -> Wine tasting is bullshit.
On a subjective measure people varied within a range ( of only 8 points ) -> Wine tasting is bullshit.
People are subject to anchoring effects -> Wine tasting is bullshit
Other senses play a powerful role in perception -> Wine tasting is bullshit
Worrying argumentation. The studies themselves could be interesting though.
The thing I got from the article is the idea that a wine is a set score like a 95 is arbitrary silliness and so much of the wine community is set up around things like that. I do think that savvy consumers and aficionados have realized this for ages and have grown out of simply looking at scores and more strongly consider pairings and preferences. I, for instance, only really pay attention to the first digit of the wine score. Most people that are well schooled can pretty easily tell the difference between a 70 and a 90, I think the bullshit factor comes in when you're trying to grade these out and say this bordeaux is a 95 but this one is a 91.
I actually forced myself to learn to use this. It seemed cool, and potentially very powerful, so I used it exclusively for 6 months on my phone.
I got to the point where it was completely coded into muscle memory, I can still type on it without the letters displayed. While it feels really cool when you are using it, there are some pretty fatal flaws.
First off, even when you are going full tilt, it is slower than using the built in keyboard. Tapping the screen is much faster than making a loop on it.
The bigger problem, which I think would be interesting to see addressed, is that there is no tactile feedback to let you know where your finger is. Sometimes you will accidentally hit the middle partway through a figure, giving the wrong character. Other times you are on the wrong side of the dividing lines when you get to the centre. These things happen often enough that it gets annoying having to go back and delete characters.
A quick fix for the "being in the wrong spot" problems would be to make the centre into a square. A more complicated fix would be to try to look at the patterns that advanced users are making and analyze them to figure out what they are actually typing ( If my finger is moving down, I am probably trying to type the top quadrant ).
When you get going, it is pretty cool, but you are still slower than other keyboards, and when you try to relax a bit you will hit the lines in the wrong places from time to time.
A bit of work on making it smarter, and giving some sort of tactile feedback, and it would be a really great way to touch type on a smartphone, but for now it is not worth it.
I also used it extensively. There are options to make the center spot much smaller that make it far easier to type with.
Also, my major complaint (and the reason I use Swiftkey now) is simply that the built in dictionary is simply not good enough at predicting N-gram sentences that I use often. There are times on swiftkey where I just tap the middle suggestion to write a sentence (although perhaps that's more revealing about my low-entropy text messaging habits than anything)
Because people find this attitude so western, and entitled, I would like to propose a slight change. Rather, I will propose some sound bytes, and hopefully somebody else will express the sentiment eloquently.
Learn to love the process, not the results.
Find your foe not in achieving success, but in destroying the obstacles.
You can't get anywhere if you cannot motivate yourself to go there.
Or something poetic like that. People should probably not assume that they have the right to do what they love. What they should learn is to love what they do.
I was under the impression that jQuery cached selectors, so the advice "Cache jQuery Objects", while smart for many reasons, would not actually affect performance much at all.
It depends on the selector and the browser. Something like $(".myclass") is worth caching on IE7 because it requires a pass through the entire document; there's no `querySelectorAll` or `getElementsByClassName`.
I wouldn't start by caching selectors all over the place unless you profile and see that it's slow. Doing a lot of selecting in a `scroll` or `mousemove` handler would be a bad idea though: http://ejohn.org/blog/learning-from-twitter/
This is a problem that I had down to the word ( except being able to give presentations, you got me on that ), and fixed.
Here is my short guide.
1: Read everything written here - http://30sleeps.com/blog/ , particularly the old stuff is relevant.
2: Find 2-3 groups of people who you know none of. Many people do this at university. This is a clean slate. Worst case, you never talk to these people again.
3: ( This plan does not involve becoming a drunk ) Booze. Varying quantities of it. In different situations. Watch how you act, if you act differently or get over your social anxiety, watch how you do it, and the results that you get.
4: Back to sober. Make an absolute fool of yourself being overconfident. Don't be a huge asshole, people do have feelings. Keep in mind things that you learned in steps 1 and 3. Watch what happens. Make mistakes. Do it wrong.
5: Apply lessons learned in everyday life.
That is abstract. Tangibly, I did 1, 3, then 2. Step 2 was local games of Manhunt, and university Debate.
Step 4 was me acting much like my drunk self. Mixing in the behavior I saw in more confident people. I did this at manhunt and I did this on trips to debate tournaments. When it got too stressful, I listened to "Come on Sea Legs" by Immaculate Machine. Over and Over again.
Gradually ramped it up, made myself say yes to every opportunity. Went out on a limb more and more.
It probably helped that debate forced me to stand in front of people and talk with no more than 15 minutes preparation. I was really really godawful at first, and got better. Remember, you will fail a lot. Don't do Toastmasters. Do open mic night, do karaoke, ideally do debate at your local university ( they will have absolutely 0 problem with this at most universities )
My end result was a personality which shares a lot with my drunk ( more honest, to the point ) self. It also shares a lot with the more confident friends I have had through my life. In no way do I feel that I am "faking" it (anymore). Getting rid of the social anxiety wholesale solved everything else. Including the not knowing how to have an interesting conversation.
One that I think is most relevant is CUSID( Canadian University Society for Intercollegiate Debate )
It was very much an organization which had a problem with women feeling pushed out, harassed, and uncomfortable. The result of that was that very few women were getting involved or staying involved.
The TL;DR; of it, is that every tournament is obligated to have an approachable "Equity Officer" as well as an anonymous way of submitting "Equity" complaints. These are comments on things that made someone uncomfortable, feel offended, or feel harassed. The equity officer then takes action, at their discretion, usually informing the offender that what they did was not cool, with no tangible punishment. Sometimes ignoring the complaint. Or in rare cases, taking drastic action against an individual.
Having been at the receiving end of two complaints, they take this policy very seriously. There is a very vocal group that believes that the policy is harmful, because it is too restrictive. There is another group that feels it is necessary and should get stronger. It is a political debate, and the two groups keep each other in check, at the current level.
The IMPORTANT result. In 2001, very few women were involved, particularly in upper levels of the community. This year, female involvement has risen to the point that fully half of the 32 finalists in the BP Nationals held two weeks ago, were female.
There are still issues, and a new discussion is being had on refinements to the policy. The point is that confronting the problem head on, did a lot of good. Particularly the point of having a formal way to complain and have the offensive parties made aware of the inappropriateness of their actions, and hold them accountable.
The philosophy that I have, that most people disagree with is "Given that Ignorance of a thing is or would be bliss, that thing is not immoral"
So if I steal and give it back, but they would never have used it in the interim, I am fine, even if they found out and got angry, because Ignorance would have been bliss.
If I violate somebody's privacy, but never change my interactions with them based on it, and never reveal what I know to other people, I am fine. Because they are doing precisely as well as they would if I had not.
Unfortunately, the fundamental problem with diabetics, and the reason that the (now) 4 people who have commented here never created this app, is that diabetics are lazy. This needs to be soooooo easy to use. Which it is, to the extent that it is a webapp.
This really somehow needs to get turned into a mobile app, which will remind you to do the checks, and prompt you for a reading. If you were to provide a simple web API, I would probably be plugging into it tonight on Palm.
If this app texted me, and I could respond with readings, I would be in love.
Particularly for basal tests and bolus tests on an insulin pump, prompted readings with a dead simple interface are key. When you are supposed to be doing this 5 times a day, you are going to be as expedient as possible. This is the primary problem with logging systems. Personally I either forget, or don't want to run upstairs every time.
The only real missing feature of note is the lack of an interface for basal and bolus testing. Though that only applies to a subset of diabetics.
Most of the tech people we talk to think a mobile app is an essential feature. But at the moment we noticed we have an older and less tech savvy user-base. So its not at the top of the list at the moment. But definitely in the future.
Btw, there are about 4-5 iPhone apps for tracking diabetes = in the app store if you are looking for one. Not sure about Palm.
I thought I had discovered a clever hack when I was interviewing. You ask enough closed questions about the problem that the interviewer tells you the answer. Then you repeat the answer back to them, as code or aloud, and you pass.
Now that I am interviewing, I am desperate to find anybody who will ask me enough questions to have me reveal the answer, and then repeat it back to me.