Brainbench perl test?
chris_jack at msn.com
Tue Sep 4 17:59:30 BST 2012
Piers Cawley pdcawley-london.0dd185 at bofh.org.uk wrote
> On 4 September 2012 14:41, Dominic Humphries <djh at thermeon.com> wrote:
> > On Tue, 2012-09-04 at 14:31 +0100, Matt Freake wrote:
> >> For that reason, I would have thought there were other, better, recursion
> >> problems out there I could use.
> > Tower of Hanoi? :)
> Tower of Hanoi (with a proper description of what the problem _is_) is
> always a better example for solving with recursion than the
> fibobloodynacci sequence.
Tower of Hanoi is one of those "aha" solutions that I would argue has little to do with day to day programming. If you've worked it out sometime in past or had it explained in a lecture then you're unlikely to forget - otherwise I don't think I've ever come across a problem with a comparable solution. Although I do have fond memories from my uni robotics course of having to program the robot arm to do it.
In regards to Fibonacci: knowing about memo'izing (or even the performance issues around calculating Fibonacci) could arguably be effectively asking if you've read "Higher Order Perl". It's an interesting book but I wouldn't suggest high up the list of books I would recommend people read about Perl unless they're doing something very specialised. I haven't yet had a problem which I felt was worthwhile of a memo-ized solution - but that might just be indicative of the sort of perl work I do.
Similarly, discriminating against people on the basis of web programming versus perl experience - is a massive presupposition about what people use perl for. Probably 90% of the perl work I do has nothing to do with the web. If you haven't read up on web security issues, SQL injection is not immediately obvious and there are various legitimate reasons for avoiding bind variables.
I think we can often treat interviews through the filter of our own experience - I went to one interview where my interviewer seemed to think it was incredibly important to know about closures.
I think it is more important to broadly assess the competency of the candidate . Which is what a lot of posts in this thread seem to have been alluding to.
More information about the london.pm