Will the Solve-the-Riddle Hiring Trend Affect IT?

An anonymous reader wonders: "It's probably harder to find a good developer, than for a developer to find a job. Seems to be a Google-riddle trend; rather than caring about references/diplomas/resumes, employers are using solve-this-and-you-have-a-job approach, not even caring about any usual information. Does that give decent graduates/talented unexperienced devs/homegrown coders a chance at the corporate job, or does it alienate potential matches?"

Comments

It Seemed to Work for Bletchley Park

[eldavojohn]

From the Wikipedia entry:

Some 9,000 people were working at Bletchley Park at the height of the codebreaking efforts in January 1945, and over 10,000 worked there at some point during the war. A number were recruited for various intellectual achievements, whether they were chess champions, crossword experts, polyglots or great mathematicians. In one, now well known instance, the ability to solve The Daily Telegraph crossword in under 12 minutes was used as a recruitment test. The newspaper was asked to organise a crossword competition, after which each of the successful participants was contacted and asked if they would be prepared to undertake "a particular type of work as a contribution to the war effort". The competition itself was won by F H W Hawes of Dagenham who finished the crossword in less than eight minutes.

Solving a crossword in under 12 minutes was the entrance exam. That's interesting. I remember reading about this in Simon Singh's The Code Book in the Chapter on Alan Turing.

I think the ability to solve puzzles is tightly correlated with the skill set desired by IT. Because it takes an inquisitive and unrelenting mind to hit the hardest puzzles. If they like to do this for fun, surely they can do it well for a living.

Perhaps it's even more important than the education because of the way IT problems arise? I constantly tell my boss that I complete the crossword everyday at work without fear of repurcussions. I feel this keeps my mind nimble and prepares me for the day.

Isn't a college degree just a symbol that says, "Look, a whole bunch of people with good reputations threw a bunch of puzzles at me. Some were hard, some were easy, but overall I did well enough to pass through these puzzles. I retained some of the information and processes but that's not really important. What's important is the fact that I'm able to solve problems and paid to do it for four years."

So, in the end, I predict this will have little or no effect on the IT world at all. In fact, I think it's a better shift towards hiring the most qualified person. For financial reasons, I went to the University of Minnesota but people on the East coast imagine a backwoods podunk frozen tundra instead of an institution of learning when I mention it. If I'm a good puzzle solver, it shouldn't matter.

Re:It Seemed to Work for Bletchley Park

[Aladrin]

You hit the nail on the head there, twice.

Real IT problems aren't as easy as the Sunday crossword. The problem being that different people are good at different puzzles. But if you're bad at problem solving, it'll show up in your lack of ability to solve puzzles.

Even our small network here at the office was ugly to diagram out. I was amazed at what a pain it was. And the guy who installed the IP-based phones could not do his job until he drew it out on the whiteboard. We ended up fixing his drawing, then revising it several times as we re-ordered the office a bit for convenience. But we had to draw EVERYTHING for him. With extensive labelling. And we also had to call things by the names he learned in school. It had to be 'FQDN' and not 'domain name' or he'd be lost. (He did eventually figure that one out and start correcting us when we just said 'domain name', though.) He's exactly the sort of 'tech' the puzzles would have made sure they never hired.

Relying solely upon the puzzles is as crazy as relying on any other single part of the interview process, though. Our office is extremely smooth, and most people get along with most everyone else. A year prior to my hire, the office was not like this. Most people hated coming to work, including the owners. They instituted a personality test during the interview process and things got better quickly.

Just 1 more thing to help weed out bad apples, that's all.

Re:It Seemed to Work for Bletchley Park

[EastCoastSurfer]

These kinds of tests usually tell me I don't want to work at that sort of place. If they are going to expect an interview candidate to write a web server in a few hours, then what will they want when you work there? Expect to have demands like 'we need an entire application written this week, don't worry about design or figuring out what the application really needs to do, just write something.'

Re:It Seemed to Work for Bletchley Park

[Monkelectric]

Yea, but C++ is all about the details, and no external libraries meant you had to implement *EVERYTHING* ... no parsing tools (flex/yacc), no threading libraries, no file utilities. Not even simple things like path normalization.

In the "real" world you'd be a fool to implement any of those things.

It filters for one type of person

[pauljlucas]

Does [solve-the-riddle] give decent graduates/talented unexperienced devs/homegrown coders a chance at the corporate job, or does it alienate potential matches?

If I do say so myself, I'm a pretty good software architect and developer. However, I don't do well under the kind of pressure typically experienced at a job interview when asked to solve oddball problems in real-time. Often, my biggest insights come when I'm not consiously thinking about the problem, e.g., while in the shower.

Those who do well at solve-the-riddle interviews are certainly intelligent and can solve problems, but it's not necessarily true that they can solve ill-specified problems -- real-world problems that need solving aren't usually as completely specified as a riddle or puzzle.

There are other ways to conduct interviews that yield good candidates. Get the person to talk about his past work -- technical people who have done good stuff love to do this with great enthusiasm. You can then ask about trade-offs in thei designs and implementations. You can usually figure out whether the candidate was a key player in the work being discussed.

Another way is to describe a real-world problem facing your company, but without actually asking the candidate anything. A good candidate will be interested in yoru problem, ask questions, offer suggestions. If the candidate just sits there, s/he's not a good candidate.

Re:I've used them

[RingDev]

While I appreciate the unforgiving rock solid logic that is math, asking someone who has been out of math based education for a number of years to prove the irrationality of sqrt(2) is a bit of a stretch.

I'd much rather go with a series of standardized logic questions (pattern recognition, basic math story problems, etc...) and one question buried in the test that is intentionally vague or poorly worded. Because well defined problems are easy, it's the problems that are not well defined that really test us in IT. Seeing how a potential employee handles themselves in a confusing situation is just as critical as how they handle themselves in a well defined situation. I would stay away from anything that depends on a complex understanding of any given topic, because at this point, we're not looking for someone who has the quadratic formula memorized, we're looking at someone who can look at a situation and pull values from that situation to plug into a formula.

-Rick

Just when I thought interviewing techniques...

[Quiet_Desperation]

...couldn't get dumber. This is flavor of the week type of stuff, folks. I'm lousy at riddles, but I win design award after design award plus bonuses in my engineering job. I have several patents. I'm sorry, but I just have very little patience for these Grand Unified Theories Of Everything when it comes to dealing with human beings. It just strikes me as HR people looking for ever lazier ways to hire the talent.

Also, Our IT people have that site blocked. I wonder what that riddle means?