lyrics to they know remix …

Eric Dane Shirtless …

On my present project we try to keep things very simple and organized, no obscure make features I think the worst we do is $(patsubst).

Of course, our codebase is much much smaller than the one you have to wrangle so we don’t need all that “power”.

But, c’mon, give us some details! Is it the standard subset of make or GNU make? Do you use the brain-crippled functional facilities such as $(call) and $(foreach)? What about indirection via $($x)? How do you deal with slowdowns caused by command evaluation - by trying hard to not invoke the shell, or by checking for $(MAKELEVEL) to cache results of process invocations? Is the whole dependency tree checked by one hairy Makefile or is make invoked recursively? Do home-made tools count on make’s cute habit of exporting variables as env vars or is it all orderly command line argument passing?

The public wants to know! At least the parts of it buried in the stinky trenches of Makefile maintenance.

I think the most important thing is not to screen resumes just based on someone not having the exact experience you need for the job, but reading between the lines and taking into consideration that specialized experience isn’t always that valuable.

As to the question you pose: I think I’d interview them both and see how they are in person.
Is the Windows programmer aware that there’s life outside of WinAPI? Is he interested in software? Does he actively learn new technologies?

It’s nice that the new graduate read the book, but does he understand the material? Does he have any interesting ideas on solving real world problems in RT systems? Does he have sufficient experience to write a shipeable RT software? Is he aware of pitfalls he might run into when developing an RT system?

I guess there is a case for or against both, but I’d definitely interview both of them.

That being said - Specific knowledge in a technology (e.g. XML) is a silly requirement. I always feel awkward to write my technology list (ODBC, XML, COM, DCOM….) - how irrelevant is this - heh?

Lets take something extreme. A new position for someone writing RT software is open. You have on one hand an extremely clever, experienced programmer who only worked on WIN32 applications - doing some complicated stuff - but no RT.
On the other hand you have an extremely clever 24 year old software engineer fresh out of the uni - who did basic and advanced courses on RT systems, and already read the books you were going to give him to read when he started working.

All other things being equal - who would you prefer?

As to negative aspects of people with lots of experience. I think you have to call them in for an interview to tell. There are people with an amazing amount of experience who are very flexible and always into learning new things. And there are those with little experience who have one way of doing thing which you can’t budge them from.

These are definitely things to look out for in an interview - but for that you have to call them in, and not trash their resume because they lack 5 years of experience with XML.

I don’t know whether this effect has “commercial significance” comparable to the problems caused by lack of experience. Maybe I’m overestimating the importance of this because generic rule-based judgment (as opposed to problem-specific logical arguments) really, really bugs me

Another problem with having lots of experience is the loss of drive. “C’mon, I’ve been through this a zillion times already…”

When I think about it, the strength of C++ lock-in is amazing. I mean, you have Wine on Linux, but you don’t have that easy a way to interface with C++ libraries from Java. Captures one’s imagination.