Yeah that is tough. Probably a little much for a whiteboard interview.

Pitch is more like "Hi I'm Free Willy and if you become our customer our first goal is to understand your business processes and supply chain inside out. To do that we need you to commit to X,Y,Z if you can't do that that's fine. Tell us now and we'll be on our way. On the technical side, we have really smart people working for us, see their work here, here and there. Feel free to call any of our customers, we can set up a meeting for you if you prefer. Your tech guys can look through our source code if they want we'll answer all questions for a week free of charge or send a guy to implement some quick prototype with your engineers to get them a feeling for the system (also free of charge of course)."

But your point still stands. The answer is simply that we weren't interested in doing business with people that prefer programmer Peggy. Our CEO's argument was basically "if they don't belive we can learn enough from them and they can learn enough from us it's not worth it for either side anyways. If they think we should know more than them about their business before we start that's guranteed failure."

Yup. But maybe good for a discussion - talking through how you might try to do it. Depending on the position, that might be getting a bit too mathy/academic, I suppose.

Pretty much this. SSDs are the stone cold nuts.
[+try Linux ldo :P Well guess not because of PS]

Unfortunately, I think you come from a pretty common place. Business software as a sector is full of crappy software, snake oil sales people and the like.
Even stuff that creates really solid business value and isn't sucky is often embarassingly simple in the sense of wow they paid THAT much for this? Pretty much a shiny database frontend with non-hard data requirements can we sold as ZOMG-WOW stuff.
There's also practices that are borderline distortion that the mafia would be proud of. Automotive industry (only know it a couple of years back might have changed) comes to mind where you pretty much have to buy the software that the GM's of the world order you to buy if you want to sell them parts etc.

I dunno I guess that's why the labour theory of value was so hard to get rid of. I still feel pretty sad when a SMB buys some supersimplistic software for a lot of money even if it's +++EV for them.

I missed that. It's probably promising way too much but it seems like an interesting enough idea. The announcement text is cringeworthy though. Not great that the example given is a standard hackme...CTFs work pretty well for security stuff but it's the transfer to other domains that's interesting.

When brainstorming ideas for the startup lecture I hold on request (I claim to know nothing just read and summarized the books so it's essentially a 90 minute...is this interesting, get the tl;dr type of lecture) I actually used "company that builds custom programming games for talent assesment" as the example (for the lean canvas etc.). From reading that post I wouldn't be shocked if I had thought about the problems more than them (a whopping 30 minutes or so) :P At least I think the key is getting the employer side right (and typical how to fill a marketplace issues) not the employee side. My assumtion was that if you build interesting programming challanges enough programmers will try it to get you started. Main competitor I identified was "do sideprojects and post on github" not standard recruiting/job interviews.

sim·i·lar
ˈsim(ə)lər/
adjective
1.
resembling without being identical.

Wasn't a snarky remark on my behalf.
I actually think it is close enough to reasonably call it the sudoku problem instead of the n*n-grid-with-unique-numbers-from-1-to-n^2 problem.

Question for those of you who are holding actually programming jobs :

If you encounter an error that happens irregularly, which option do you normally choose :

1) Figure out what is causing the error and fix the cause
2) Patch the effect.

An example might be an element is supposed to be positioned at a specific offset/position. For some reason, it gets placed incorrectly maybe 10% of the time.

suzzer was clear about all this when he first posted about the problem. I will say that the solution he posted didn't create an object though, just functions.

Mea culpa. I totally missed the original post where suzzer described the problem.

I always try to properly debug the problem, but practical reasons will sometimes force you to use a band-aid. It's really hard to answer in a vacuum, since it is so dependent on the project you are working on and the bug.

If I had to try to generalize, I'd say that the earlier you are in development, the more it's worth it to spend the time getting the proper fix.

That's not really the decision of the programmer in most companies. It will come down to how serious the defect is, how often it happens and if it happens to the customer account manager in particular...

3) ignore it until QA finds it

Same. I browse this forum in bursts and missed out on context.

4) Ignore it until a customer finds it and complains.

Seriously at my work, if you find a production bug that's not immediately under your domain – the entire system is set up to discourage you from doing anything about it. You can file the bug with QA, but then it becomes "your bug". You'll be asked approximately 50 times to explain the bug to various groups from QA to the product team, to the dev who eventually has to fix it. If you actually want to get the bug fixed you will have to champion it, and get it prioritized along with all the bugs and change requests that the business actually knows/cares about.

I went through that once. Never again unless the bug is some kind of major security hole or other damaging issue. It kinda sucks for dev morale though. We spend so much time fixing even the tiniest, 1-pixel, most bizarre edge-case bugs during development. Then once we go live we orphan the thing, and it takes a much more major bug to ever get fixed.

It really comes down to ignorance, and not always in a bad way. A small business doesn't, and I'd argue shouldn't, be involved in technology.

The argument boils down to this: if you had to chose between hiring a volatile dev team of 2, then you are looking at paying, at the minimum, $20/hour, which is > $6000 / month.

One can quit, one can be a dud, they may not be able to work together, etc. Finally, unless they get extraordinarily lucky and get genuine talent at the pay scale, they are going to have middling programmers assessing new talent, and it stands to reason that they will find programmers like themselves, or if they do find an expert, said expert won't take on the job at that pay.

How do you manage people when you have no clue how to do their job?

Even though you think, "oh, wow, so expensive," the deal is very simple. A company may be paying $5,000 / month for a product, but part of the deal is that they get whole team of (theoretically) vetted talent and a complete, ready-to-use product. They don't have to worry about HR issues, etc, which is monetarily be cheaper than hiring two people who won't even get the product out within several months.

Kind of depends on where the software is in the product life cycle. If you an an early point in development I would recommend fixing the problem not hiding it. At least understand why it is happening. There is a point in the life cycle where the decision on whether or not to fix it is pretty much out of your hands like the others said.

Anyone in the uk looking for a job? Great opportunity here! https://drive.google.com/file/d/0B9p...dJN1hCck0/view

from hn obv.

Ideal preference:
Under 30 years old

Pretty sure that's illegal?

Edit: Well, pretty sure it's a joke ad unless I'm confusing myself lol

I was waiting to read "salaries up to $9.50/hour" at the bottom.

In the US, yeah

Anyone here have any experience with SMS api's/services? I recognize the name twilio and they look legit but is there any other company worth looking into? I don't trust google searches for product reviews anymore so thought I'd ask here.

It's been years since I used them, but Clickatell are legit and at least were very good back when I used them.