The cynic in me wants to suggest putting up ads a la "Sick and tired of grinding tests and accumulating student loan debt? Drop out and join us" at your local university :P

shabby,

craigslist yo.

Though the trick with Craig's List is to make sure to write your ad in a way so that you don't look like yet another shady character, as that site is so loaded with them.

Don't forget to use the words "Rock Star" in your ad.

Rock Star guru ready to work 26 hour days because he's so committed to excellence!

Being on the IT recruiting (well account management, and only contractors) side, I 100% understand why you want to do it yourself.

Employee/ Friend referrals are your #1 by far.

I would say also check if there are active Local User Groups, as people there may know who is looking for a job, but this is more of a longshot and probably not going to move quickly.

Also you might be able to leverage a recruiting company to pay much less of a fee if you will work with them exclusively and basically guarantee you will hire someone.. some bottom feeders will take low fees.

He should also be an expert at Ruby on Rails, Python, Django, Erlang, Clojure, jQuery (javascript not necessary but it'd be nice), HTML5, XHTML, HTML, CSS, CSS2, CSS3, and yeah, he better have read the WHATG and W3C docs because he's getting quizzed on second-child pseudo-classes. He should be up-to-date on all of the latest and greatest development culture and knowledge, have some experience in iPhone and Android development (coding in blackberry is so 2009, man!) and should be a bona fide expert in Visual Studio, Vim and Emacs. Please send links to at least 4 dynamic (non-static) websites he's built from scratch (no CMS sites please!). 40K/year w/ profit share (didn't we say were a start-up?) for the right person.

My company is also looking for a web etc. guy/gal not-so-rockstar because that crap cost $$$$, so I feel the pain also. Nice thing is that said guru only has to deal with me for the most part. Thankfully, I don't have pointy hair, know some coding, and don't have particularly strong opinions on it. Well, said guru should know Lisp (booyah!). j/k.

The problem is that my company really doesn't know what they want. Maybe they want a) but then again only if a) is really needed or something like that. This is that area where I have to stick my neck out and (hopefully) get my head cut off. Love making mistakes, getting the blame for things I didn't do, and otherwise dodging bullets for no good reason. I'll even let said not-so-rockstar laugh out loud at my frumpy and totally denormalized database. Actually, that could be the "coding test." See my database: if they say "That's really cool, man," I'm probably dealing with someone that knows less than me as well as someone who refuses to open their mouth when they see something wrong, which is both really bad (kind of what happened to the last guy). I haven't taken out the time to normalize it, so make the interviewee do it.

Thanks for the suggestions guys.

As for the Joel Test - We don't score a 10 on it - but everything we're a no on I think its better to be a no than yes.

For example:

Do you have a spec? - Nope. Every story we write has a spec and there's a long term vision for our product but specs are lame and useless.

Do you have an up-to-date schedule? - Well... We have a great schedule for the next release (one week). A rough schedule for the next couple of months. And... a vague cloud-like idea of what we'll be doing a year from now. Seems like the right thing to me.

Do you have testers? - Nope. There are times when this is good and times when its bad, but for us its not that important (Developers should test their own **** yo).

i agree with this but it is orthogonal to the stuff a good tester is doing: integration, e2e, performance, security, ui, ux, exploratory, destructive, plus they're a second (or third!) set of eyes that have a different thought process than most developers.

I don't agree that your list of stuff (integration to ux) is orthogonal to what a good tester is doing. In many cases developers should be testing all that **** too.

Anyway, I do agree there are lots of uses for Testers. But I'm hearing more and more about companies that are doing really well without them (or at least not nearly in the ratio of 1 tester to 2 or 3 developers that Joel Spolksy is recommending).

Etsy is the latest example I've heard of that doesn't have many (maybe none?) dedicated testers. They don't even have a superb set of tests. Mostly they just have a really amazing deploy process where they can constantly push code to Production and monitor the impacts.

no i'm saying those *are* the things a good tester is doing. those activities are orthogonal to the kinds of functional tests i think of when you say "developers test yo ****". i guess i made a leap there.

you're right, it's certainly possible for developers to do that work and they should be involved with all of those systems. however ime, devs get weaker the further you go in my list. having a specialist becomes worth it once your system reaches a certain size or complexity. (actually most devs i've had the pleasure of working with haven't even been good at unit tests but i've read stories about better practices on the internet...)

and even if devs are doing that testing, i stand by the value of the unique viewpoint that a good tester brings to the table.

I did not know Joel Spolsky co founded Stack Overflow.

Re: the orthogonal thing - I just didn't write it very well but yeah - I think a lot of those things should be covered by developer testing.

Unique viewpoint is definitely good - but it doesn't need to be a Tester per se and you definitely don't need 1 per every 2-3 developers.

I tried to find the Etsy video I watched and couldn't find it. I found this though: http://www.livestream.com/etsy/video...d-21c898693282 and it seems like it covers a lot of the same stuff. Very cool look at how to do continuous deployment well.

Somebody is testing your application. It can be the devs, the testers, support or your users. All I know is it sucks when it's the users...

Well, it depends on the development model. In the classic XP model of an incremental, in house process... sure, let it be the users.

I'd definitely say that the majority of the time users probably shouldn't be testers.

But I agree that there are definitely models where it makes sense. If you have a bunch of non-critical features and can release changes to these features gradually to your users (so say only 1% of your users see a change at first) it's probably ok that the features aren't thoroughly tested. Any particular user won't see a bug very often and even when they do it won't be a big deal.

So I've been learning Ruby on Rails for about a month and a half now (from basically no programming/no web design background) and think I finally have a decent grasp on most of the fundamentals. My sample site now has a pretty thorough test suite. It's a basic twitter clone (which was mostly written via a tutorial book). I've since added support for comments, private messages, password resets/initial user email authentication, email notifications for various things, a few levels of different site administration, search using ajax, etc. Mostly everything has been written from scratch, I think the only gems I'm using are will-paginate, and gravatar.

I'm kind of running out of ideas to extend this site, and would like to start a new project. Anyone have any decent ideas for either websites to copy that would be a bit more complex (just for practice purposes obv) or a simple site idea that would be useful for learning rails on a slightly deeper level? Or more ideas to extend my sample app?

Also, where would you guys go about learning ajax/javascript. While I'm feeling more confident in my ruby, my web design remains garbage... so that's probably next on the list of stuff to work on.

well i got a VPS just dont know what to do w it

Hopefully your development model matches your customer's expectations. It doesn't do much good being agile and what not if your customer's test cycles take longer than your development cycle.

Depending on what software you develop users are the best testers for many things. Release early/often etc.
UI testing is usually part of the designer's job (like software testing is part of the programmer's job).
But yeah for many things it makes a lot of sense to have extra testers. Personally I think it's most important for security.

Oh neat...Mozilla Browserquest...anyone up for playing it a bit?

Have any of you done IT support for big companies? What's a ballpark expected hourly charge? When I say they a big company it's household name big. It's specialised work (support for our product, highly likely to include lots of C++ coding as well) and we just need a rough idea of what they wont bat an eyelid over so we can think about it from that starting point.

How does the time get broken down?

Do you want to charge them only when they use you, or a monthly fee? Do you want/need to be available 24/7? How many people are you going to dedicate to this client? How much support do you forsee them needing?

Nevermind, just "finished" the game

This is all stuff we're thinking about atm. They want a 3 year support contract, details are in the planning stages at the moment.

We're highly confident in our product in terms of stability, reliability and design quality so don't foresee much troubleshooting support at all.

My brother is the only one who can offer the support, but if the amount is enough we would happily hire someone full time who can work for us and manage that support contract (the total contract would at absolute minimum need to cover their wages though which for a good C++ person would probably be in the region of £50k+)

24/7 access probably not required although it is possibly. We're just trying to work out how much to quote them so that the slowdown on our own product development is justified and don't really know where to start from.