Starting on my new team today. I am the only person at this site working on this scrum team, and the team is made up of people in 5 different locations across 3 time zones. Should be interesting.

If I run into this issue again, I'll definitely look into this.

I do the same thing as you: look at Arch Wiki for documentation. It really is an exemplar of what documentation should be like for all Linux systems. Most of it is portable, but some things are isolated to how your flavor works.

***

More impression from trying to get work in SF.

Phone interview last week. I directed the guy to my music theory articles and he said "wow! that is some cool stuff!" It was for a music-based job, but still, kind of nice to get some positive acknowledgment once in a while.

Took a test yesterday. This is on a live OS disk since my OS had to be rebuilt and I wasn't sure when the questions were due. It was 6 questions, mostly fairly simple I/O and generating collections.

I probably messed up some of it due the time constraint, even though the quiz specified "quick and dirty" for the problems. Out of desire to do them all (so much better than Pokemon Go?), I took about an hour twenty for a one hour test.

If there are two classes of problems and algorithms that quickly expose my self-taught status, it is heaps and graphs. I know just enough about each to know that I don't know anything about them. I can sort of do a graph as long as it doesn't have any requirements beyond it being a graph... ironically, I can do graphs just fine in SQL, but not in a real programming language. Heaps are just total jibberish to me. One question involved a graph with some unusual requirements, and I just wrote a function that did what two functions are supposed to do (I think?) and then wrote out some of my thought processes about where it fell short of the full requirements and where I thought it was almost there. I really got stuck on that one.

What was the programming language Dave? I am going to start a thread on interview test questions. Might be a good source for solutions.

I always do my coding tests in Python. It's the language I can write the fastest. I seldom have to Google to remember anything.

You know, you should take a day and spend it learning how a heap works. It won't take longer than that. They're very simple and they lead to a few useful things.

Like, it doesn't happen often, but sometimes someone will want you to demonstrate or explain a sorting algorithm (well, a good one). Heapsort is by far the easiest to demonstrate or explain.

Graphs though, have basically infinite depth in terms of how much there can be to learn. I took 2 full semester courses on graph theory (as a math class, aside from CS stuff) and didn't even scratch the surface. My graduate work was in graph theory and computational geometry, before I, uh, quit to join a start up in 1999. ****in internet.

Every point you raise is exactly what makes it a WTF-worthy story. But it's 100% true.

The Sedgewick algorithm classes on coursera do a pretty good job of covering this stuff for practical purposes. I find it pretty hard to retain this type of algorithmic knowledge though.

Haven't fully read this yet, but apparently DogIsHead (poker player) is now a software engineer or something and this week's Hacker Newsletter had an article by him about negotiation.

Yeah, it appears that's his new thing - the AirBnB thing linked at the start of the article was mentioned here and made it to NVG http://forumserver.twoplustwo.com/29...kills-1605947/

never heard of the guy until his airbnb post but my god is he insufferable.

The sticky headers/footers and floating popups that appear on that site when you scroll up and down are insufferable.

I can describe a heap by rote, but I have a hard time building one up, if that makes sense. I get the impression that those courses give you a lot of ground material to work with, which is basically, you get a file of functions that already have the heap for you.

I never did that, so suppose I generate a list of 20 numbers, grab distinct random numbers, and push them into a function that generates a heap. This is where I end up with trouble.

Also, didn't coursera close all of their classes (the Sedgewick book isn't very good, IMO)?

Does anyone know for SEO purposes, does a site map include pages only accessible by login? Google cant scrape those anyways correct?

#1 - usually not

#2 - correct

What's the trouble? If it helps, a heap is valid if it's root is the smallest member in the set - that's the only condition for a valid heap. Last time I implemented a heap, I did it as a fixed array. array[0] was the head, array[1] and arary[2] were it's children. array[3] and array[4] were array[1]'s children etc. The nice thing about this is that some heap operations involve swapping the "end" of the heap with the root, and that's super easy to do with an array. It's also very efficient.

I just liked the lecture videos. He shows the important parts of the implementations and walks through examples with visualizations that are much easier to follow than any book explanation. They're good for getting a feel for the algorithms and data structures.

If you struggle with writing implementations of whatever algorithm you just need to practice and compare with others fully working code. There isn't really any other way to learn.

Also I'm not sure if coursera is still free/public but there is a torrent for the algorithms videos. Some of it is on youtube too.

https://www.youtube.com/watch?v=Hz_P-_R2l60

You can also use robots.txt to keep them from indexing things you don't want them to.

Most of my site is user based stuff, so it feels like the sitemap would just be like FAQ and terms and stuff. Almost feels pointless. I guess I want them searchable though so I have to

Not to go all "I can do fizzbuzz," but here's my implementation of a max heap (without looking up online, lol):

In an odd way, I get it, but I don't at the same time. It takes a lot of pen and paper to see exactly how it is stepping through the list (print statements can serve as a slight guide as well), but it's still a bit confusing to me.

I get that, in essence, the heap is saying take an Array[n];
for each n, the element in position floor(n/2) is less than the n. as you iterate from n to 0, the numbers get larger, therefore:

A[5] > A[5/2] > A[(5/2)/2] > A[((5/2)/2)/2] .. A[0]

I feel like I'm missing something super obvious and important here, but I'm just shy of figuring it out.

darn it, it is supposed to be A[n] < A[n/2] < A[(n/2)/2] < A[((n/2)/2)/2] .. A[0]

I'm pretty sure I can separate my life into two different columns: before and after purchasing a gtx 1080 & ultrawide UHD monitor.

I have a 32" dell that does 2560x1600 and I have a hard time working on anything smaller.

http://cs.boisestate.edu/~jhyeh/cs42...mer16/lab3.pdf

This is the last program I have to do for my summer course in CS 421 -- Design and Analysis of Algorithms.

Looks brutal, no?

I have an A in the class but maybe not after this assignment xD I'll start working on it tonight.

I just installed SUSE and I'm finding that a certain utility takes an absurdly long time to execute if I output to a new file as opposed to an existing file or the console. I'm finding the times to be something like:

Output to existing file - 5 secs (normal)
Output to console - 20 secs (normal)
Output to new file - 2 mins (abnormal)

If I ls in another window I can see the file growing very slowly. This doesn't happen in Ubuntu - it's 5 secs to a file regardless of whether it needed to create it or not. I'm invoking this from within a script, so a simple workaround is to just add a touch command before it, but I feel this shouldn't be necessary. Any ideas why this happens?

Do you know how to use strace? That's what I'd look at.

http://linux.die.net/man/1/strace