Ah, cool. Thanks for the clarification.

haha! I've had a run in with rook before!

speaking of trolls did you guys read the story about gawker exposing the real life identify of this guy violentacrez, a reddit poster who is supposedly the biggest troll in history. pretty interesting story, he ends up getting fired from his job and losing his home in the aftermath of exposure

Ya, reminded me of some of the poorer threads in 4L and how much some of those posters would enjoy being outed. The age of anonymity on the web is pretty much over.

you could do all sorts of things! a game? a fancy visualisation for one of the onboard sensors (GPS, compass etc.)?

Wow just read it. Quite interesting

seems like the 4l crew was more playful and ridiculous. this guy seemed to be a stone cold pedophile and all around sick dude.

I don't think it ever existed lol

http://www.cigital.com/papers/downlo...r_gambling.php

lol 1999. in the modern world, no one ever fails at basic crypto tasks like random number generation or salt hashing.

Interesting read... very surprised that the company published the code for their RNG

very good read. amazingly simple too.

You're not gonna discover the issues you want to test with automation until lots of people actually use the system. How big is the company? Live fire in house testing is a pretty good and very cheap method. Just get bunch of people to logon at once. We do live fire tests before every release. Our software is different though. The stress doesn't come from many users. It comes when a few users simultaneously make extremely computationally expensive queries.

Not true in our case. I suspect its rarely true but I guess its one of those it depends/YMMV issues. Every time I've done load testing I've learned tons of valuable information.

We're small so this isn't really an option. Although we have a small number of users its just not quite the same.

I think in a lot of cases load testing isn't necessary - but when you have a specific point in time that you know is going to really increase load and increase attention to your site I think its important to make sure you're not going to fall over*.

The beauty of a good load testing suite is that even in cases like you're describing (few expensive operations) you can simulate that and see how your overall performance degrades. It's probably nice if you can serve your "simple" customers well even when your "complex" users are experiencing performance problems.

* See the stories about the Romney campaign's 'app' disaster. That's a perfect example of an application that needed thorough and automated load testing.

Thinking about this a bit more - I suspect this is a false cheap. I've spent about 3 and a half days on writing tests for The Grinder that give me a pretty decent coverage of what users can do on our system. With another few days of work I'll be able to launch multiple EC2 machines all generating user-like requests for our app and generating useful load/performance information.

Once you're there all you need to do for a new release is update/add any tests specific to the release and re-run the tests.

So in about 6 dev-days of work I have a reusable load testing system that can scale in the cloud. If I find performance problems we can fix them and then re-run again with minimal extra effort (say an hour each time).

I think 6 dev-days up front with an extra dev-hour per load test after that quickly becomes better than the little up front but heavy cost of manual fire-testing.

agree with jj. that sounds infinitely better than manual fire-testing.

but you guys HIS APPLICATION is DIFFERENT.

random thought: How difficult would it be to create a script which would automatically open up notes and place into certain area of the screen(on 2nd monitor) whenever you was in a flop against villains (on pokerstars)? (sometimes annoying having to keep on hovering over notes especially when they get long)

Im a bit of a beginner at programming and dont even know where to start with this.

Posted that in the BFI simple questions thread. Probably good here as well.

Zynga and Groupon..there's other reasons not to work for them. Groupon is run by horrible people and they treat their employees badly and Zynga is just evil.

From the companies you selected I guess you should include Amazon (the tech side) and possibly Apple as well.

Google a bit on the recruiting dunno what you want to apply for but for programmers Google interviews are sufficiently hard. They'll ask you a bunch of algorithm and data structures questions usually and you go through multiple rounds with multiple people and are asked to solve non trivial stuff.

Might want to investigate ERP businesses if you feel you could work in that field i.e. SAP,Oracle,salesforce (I'm assuming you're mostly/only interested in the big companies)

I read an article a while back about how Amazon is a terrible place to work at - it might of changed now though or been a disgruntled employee ranting

I've heard mixed reviews. I know a couple of people that work there and quite like it and I've talked to a few ex-employees that found it frustrating.

What's the deal with Zynga?

My only experience was a small Q/A with a senior person at Zynga (so obviously not an objective source) and it actually sounded quite interesting. If you're interested in things like A/B testing, usability and how you build features that people want - they do a lot of really cool stuff. In fact if you were interested in starting a consumer oriented website or mobile app working for Zynga would probably be an invaluable experience.

As for the stigma about the fact that they do meaningless **** - this guys answer was that what they do makes people happy. Just because it might not be something you're interested in doesn't mean its not worth doing. Seemed fair to me.

Edit: Also, I suspect Zynga does a LOT of working with personal data.

all large companies will have good projects and bad projects, good team and bad teams, good managers and bad managers. it's completely YMMV situation. amazon does pay well, even for seattle, so that goes a long way to treating people the right way. also, there are so many people coming over from microsoft to amazon, the culture is definitely changing to be more like microsoft. that's a bad thing, but it's not a terrible place to work at... yet.

sylar, what do you do there?