help on C++/programming interviews...

[Noah Roberts]

Re: help on C++/programming interviews...


Chris Schumacher wrote:

On 7 Nov 2006 10:41:11 -0800, "Noah Roberts" <roberts.noah@g mail.com>
wrote:
>

Hopefully you don't really interview people this way. If you do maybe
your employer needs to see what else is in the management market
because that is NOT how to do an interview. I mean, it really is a
rather abrasive and confrontational way to talk to someone and you're
competing for my skills as much (or more possibly) as I am trying to
sell them.

>
Hell screw that, I'd work for the guy in a second. It's exactly what
I'd do in his position. That was a nice case of someone passing the
WWGATD* test.

Yes, because everyone would love working for or with Thrawn.

[Earl Purple]

Re: help on C++/programming interviews...


Richard Heathfield wrote:

Earl Purple said:
>
<snip>

Companies always want the best C++ programmers and are then happy to
shove them into maintenance of legacy code. Why?

>
Because maintenance is more difficult than development. Any fool can create
bugs, but it takes a special kind of fool to track them down.

Bring in the good programmers from the start and you don't have the
mess there to clean up in the first place.

[Richard Heathfield]

Re: help on C++/programming interviews...

Earl Purple said:

>
Richard Heathfield wrote:

>Earl Purple said:
>>
><snip>

>
Companies always want the best C++ programmers and are then happy to
shove them into maintenance of legacy code. Why?

>>
>Because maintenance is more difficult than development. Any fool can
>create bugs, but it takes a special kind of fool to track them down.

>
Bring in the good programmers from the start and you don't have the
mess there to clean up in the first place.

Yeah, I know, but that's what they tried to do at the start, only they
didn't manage it - and now it's too late, because it isn't the start any
more. Companies have to run with what they have, not what they'd like to
have had at the beginning.

--
Richard Heathfield
"Usenet is a strange place" - dmr 29/7/1999

http://www.cpax.org.uk

email: normal service will be restored as soon as possible. Please do not
adjust your email clients.

[Alf P. Steinbach]

Re: help on C++/programming interviews...

* Earl Purple:

Richard Heathfield wrote:

>Earl Purple said:
>>
><snip>

>>Companies always want the best C++ programmers and are then happy to
>>shove them into maintenance of legacy code. Why?

>Because maintenance is more difficult than development. Any fool can create
>bugs, but it takes a special kind of fool to track them down.

>
Bring in the good programmers from the start and you don't have the
mess there to clean up in the first place.

Although off-topic in both groups posted to, I think that comment
deserves a follow-up.

Yes, it would be ideal if that was how projects are managed.

However, a typical project leader in a typical firm would not be a
project leader if he or she couldn't point to past "success stories".
With success being narrowly defined as accomplishing stated goals within
some time & budget frame, ignoring the likely consequences for what
happens later, because it hasn't happened yet. Thus, Darwinian natural
selection operates, ensuring that the project leader's focus will be on
short term advantage, not on negative consequences for those who must
maintain the mess later on (some project leaders may of course focus
also on that, but then, unless there is a long-term follow up policy in
place, it's /better/ that the next lead should fail, because that's then
one less person to compete with for a limited number of positions).

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

[Noah Roberts]

Re: help on C++/programming interviews...


Alf P. Steinbach wrote:

* Earl Purple:

Richard Heathfield wrote:

Earl Purple said:
>
<snip>
>Companies always want the best C++ programmers and are then happy to
>shove them into maintenance of legacy code. Why?
Because maintenance is more difficult than development. Any fool can create
bugs, but it takes a special kind of fool to track them down.

Bring in the good programmers from the start and you don't have the
mess there to clean up in the first place.

>
Although off-topic in both groups posted to, I think that comment
deserves a follow-up.
>
Yes, it would be ideal if that was how projects are managed.
>
However, a typical project leader in a typical firm would not be a
project leader if he or she couldn't point to past "success stories".
With success being narrowly defined as accomplishing stated goals within
some time & budget frame, ignoring the likely consequences for what
happens later, because it hasn't happened yet. Thus, Darwinian natural
selection operates, ensuring that the project leader's focus will be on
short term advantage, not on negative consequences for those who must
maintain the mess later on (some project leaders may of course focus
also on that, but then, unless there is a long-term follow up policy in
place, it's /better/ that the next lead should fail, because that's then
one less person to compete with for a limited number of positions).

Sometimes it's a matter of knowledge as well, especially with startups.
Take my position...comp any was started out of a garage by
non-programmers who learned how. Original was written in basic in the
80's. Then it was translated to C, again by the same people, for
win16. Then it was translated again to win32 and yet again they
switched to C++ and started using OO.

Code that has been around for that long and been translated that many
times has a tendency to get a little unclean. It all works, but yes,
there is plenty of rot and places nobody wants to go because it is
almost impossible to maintain. Super long functions, monolithic
objects, the works...

I think the agile crowd has shown that TDD and such work, and work
well, so there isn't much need for such stuff to continue...but you
have to be able to deal with legacy code or you're just not worth your
salt. Even TDD with refactoring can get messy when a deadline is
comming up, or even when not and the design is just flawed. The cool
thing is though that refactoring works great to understand a piece of
software and you can slowly get the product under test as you add
features and change things.

Point is, it doesn't matter how it happens and it can happen in many
ways...but legacy code and code rot occur and you can't just start over
when it does. Not in a commercial product worth a fortune. You have
to continue working on it, adding features, while trying to keep it a
viable product. This necessitates a clean as you go kind of philosophy.

[Michael Angelo Ravera]

Re: help on C++/programming interviews...

Noah Roberts wrote:

Michael Angelo Ravera wrote:
>

I've sent more than one interviewee away crying after exposing that
they made up their experience and knowledge. I've been known to ask the
interviewee about how he stands on the acetylsolosin QID controversy in
C++ or whether they have had any experience with Feci Tauri methods.

>
I would be one to answer these questions in detail. You better be an
expert too or I'll convince you they are real.
>

When they start to enumerate their vast experience with (or passing
exposure to) each of them, I know that I'm going to have fun in the
(usualy very short) interview. If at all possible, I will arrange to
get called away and leave the interviewee alone to stew for a while
while I go joke with the buddy of theirs who recommended them.

>
But, the joke would be on you because I would know what you where up to
and have decided I probably don't want to work for, or with, such
people.
>

When I come back, I usually try to discover their experience with NAFC
objects and thank them for coming in.

>
NAFC as in having to ask trick questions to discover someone's true
experience?
>
I actually have extensive experience with NAFC objects. NAFC managers
too.
>
Hopefully you don't really interview people this way. If you do maybe
your employer needs to see what else is in the management market
because that is NOT how to do an interview. I mean, it really is a
rather abrasive and confrontational way to talk to someone and you're
competing for my skills as much (or more possibly) as I am trying to
sell them.

I only interview people this way when they start off as liars and self
indulgent ones at that. People who come in saying "I heard that you are
looking for a snot nosed kid who can program in c++. Well, here I am!"
have the job unless I get overruled by someone else.