How would you improve this network?

[Pashers]

If in an interview someone asks 'how would you improve this network?' what is the 'correct' reply? Surely you can't improve a network just by looking at it can you?! Or have I missed something along the lines? Is this meant to be a trick question? As in you just say it's about the data flow etc. Or are they actually wanting to know if I have any suggestions on improving a network (maybe on the physical side)?

[Gibson335]

You need to ask them those very questions. Such as:

 

'In what way do you mean 'improve'? Do you mean physically, the infrastructure, server load-balancing, the end-user experience, support?'

 

Get them to be more specific. That will open the door for you.

[pagelad]

'Upgrade to win 3.11'

[IanT]

Windows Me!

[pete]

"What problems are you currently experiencing?"

"What do you want to do that you currently can't?"

"What (probably government) requirements are driving a need to improve?"

"Depends, how much cash have you got?"

 

i.e: "I can suggest quite a lot of stuff, some of which probably will improve "x", but without looking at your setup and getting some more info, that's a bloody stupid question to ask"

[Dom_]

They will generally take you around the school and give you some background info, but it is sitll a stupid question!

[matt40k]

Could just be a PR campaign

 

Capita rules!

[Hightower]

I found as a golden rule, never talk about money in an interview. The money might be there to spend but from past experience I've found they don't like to hear that you are wanting to come in and blow a wad.

 

Maybe talk about Open-Source products that cost nothing/next to nothing that can improve the network.

[kmount]

My advice would be to outline that you'd ask the users themselves for what they felt was the most limiting factor of the system and base your improvements as direct solutions to known problems rather than giving people stuff they don't necessarily want.

 

Depends who's interviewing you; if it's a techy person you can gibber on about anything you want; if it isn't they aren't interested in your failover network topology etc.

[pete]

I found as a golden rule, never talk about money in an interview. The money might be there to spend but from past experience I've found they don't like to hear that you are wanting to come in and blow a wad.

 

Heh, nor would I - the "depends, how much cash have you got" was a throwaway and leads into the "this is a stupid question without a frame of reference" part.

 

Thinking about it for a moment though, if I'm interviewing at a bunch of schools and one is in severe financial constraints, whereas (all other things being roughly equal) the other two aren't, I'm more likely to lean towards the job where there's sufficient funding to do stuff properly rather than the make-do and mend approach.

 

Cash flow problems are a valid concern to the interviewee if it'll have an unexpected negative effect on their ability to do a good job.

[srochford]

I found as a golden rule, never talk about money in an interview. The money might be there to spend but from past experience I've found they don't like to hear that you are wanting to come in and blow a wad.

 

Maybe talk about Open-Source products that cost nothing/next to nothing that can improve the network.

 

I think that last bit is good - it shows you can think outside the box a bit. It's all too easy to do the "oooh, that looks expensive" but asking "How do you monitor now; have you thought about Nagios etc" or "what's your current maximum throughput?; well, there's no need in going to gigabit yet but we could make sure that the switches are configured to give best throughput"

 

Without knowing exactly how the question was asked it's difficult to say if it's good or bad as a question. In general, though, open ended questions are good at interviews - it gives people a chance to show thinking skills and how well they can present their thoughts.