About the employment situation in the city

73 posts in this topic

3 hours ago, engelchen said:

What do people lie about? Do they really think that they could get away with it? (Now I'm just curious)

Usually they overstate their skills or simply have no experience with it.

We use several tricks during the interview to find it out. First we give them a chance to self identify their level on a certain subject, for example C++. Then we go into detailed questions to see if they really know what they are saying.

 

Example of a C++ developer question: "Have you used C++11/14/17? Could you tell me what new things are you using from these standards?". One guy answered: "Templates"!  <--- completely wrong!

 

Another type of lie is overstating their previous roles in a project. The trick to get them lying is asking them to design a system concept of their previous project. If they were actually so important for a project, they will know how it works.

 

3

Share this post


Link to post
Share on other sites
2 hours ago, starkebogen said:

 

This doesn't match my experience.

I hope you are being sarcastic, because SW developers are probably the most autistic professional group.

I could spend hours telling crazy stories.

1

Share this post


Link to post
Share on other sites
3 hours ago, engelchen said:

 

What do people lie about? Do they really think that they could get away with it? (Now I'm just curious)

 

I think a lot of people lie on their CV and try to make it fit the job better than the reality.

On the other hand, companies are always on the look-out for recent graduates with at least 15 years of experience in "the field" and willing to accept the lowest wage, since they are fresh out of uni. Nothing really fits together, does it?

 

Where I work, we currently have a management trainee who, according to his CV and what he tells you, must simply be Superman,

As some of you might have already guessed - he isn`t!

Enough said.

3

Share this post


Link to post
Share on other sites

I can relate to some of that.  

 

Many devs overstate their qualifications, but to be fair, I think there are so many projects where the blind are leading the blind they don't even know that they don't know what they're talking about.  My current project is just...I can't believe these people have degrees.  They know NOTHING about basic architecture, oo design, etc - I say "encapsulation" and their eyes glaze over.  

 

I mean, one time I got a "senior" oo dev in for an interview who couldn't explain polymorphism AT ALL yet he'd been deving at Nokia for 10 years.  I can't even grasp how that's possible but it happens.  A lot.  Not in the US so much but in Germany, I've found this lack of fundamental skill to be a huge problem.

 

I can't fully agree with the autistic devs thing.  Yes devs are geeky, for sure, but most I've worked with are a lot of fun and have no problem engaging with others.  But a problem I've encountered since moving here, is an abject lack of teamwork and it's along the lines that they just don't seem to even know what that is or how it can be useful.  Communication is AWFUL but I think it's largely a product of the German uni system where studies are completed soooo independently (contrasted to my uni experience where we had group or pair projects all the way through) as well as management just not valuing any kind of comprehensive, coordinated effort.

 

it's so frustrating.  I hate my job at this point as no one has any clue what I'm talking about even on the most basic dev topics, and they see zero value in coordinating what we're doing.  Such a waste.

 

  

1

Share this post


Link to post
Share on other sites
24 minutes ago, robinson100 said:

I think a lot of people lie on their CV and try to make it fit the job better than the reality.

 

oh yes!  But this is another problem I've encountered in Germany:  you can have all of the really important, hard to develop skills, but if you're missing one or two minor points (in the grand scheme of things) you will not be considered at all. 

 

German employers don't seem to understand that a good dev can learn the small niggly bits in very short order.  It's just baffling to me as it's so contrary to the US style:  show us you have some solid core skills and you can think and we'll work out the small missing links later.  Here, it's most often just blindly formulaic so I can sort of understand why people lie - to a point. 

 

I've had German friends and coworkers encourage me to lie to fit a job description as they say the employer will reject my CV based on any "miss" - which is, in my experience, absolutely true (here), but completely stupid.

 

4

Share this post


Link to post
Share on other sites
11 hours ago, MikeMelga said:

Example of a C++ developer question: "Have you used C++11/14/17? Could you tell me what new things are you using from these standards?". One guy answered: "Templates"!  <--- completely wrong!

 

You've got me until this.   This is a very silly question to ask and deciding if the person is skilled in C++ based on this question is extreme and actually borderline dumb.

 

Expecting that someone know by heart all the differences in the standards is really silly and doesn't say if the person is a good C++ programmer or not.  And actually Variable Templates is one of the differences in the standards.

 

If you are a Windows developer, Visual Studio 2017 still didn't finish implementing everything in standard 14.

 

11 hours ago, lisa13 said:

I mean, one time I got a "senior" oo dev in for an interview who couldn't explain polymorphism AT ALL yet he'd been deving at Nokia for 10 years.  I can't even grasp how that's possible but it happens.  A lot.  Not in the US so much but in Germany, I've found this lack of fundamental skill to be a huge problem.

 

 

There are plenty of things any random good IT guy can't explain, because IT is huge.   You might think that polymorphism is something every person should know, specially if you've been working in oo for some time.   But actually, working for long time in the same company normally ends up in you working and becoming specialist in a small set of things and losing/forgetting plenty of other things.

 

Then in your example, if you want to explain polymorphism in general you have to do it in a very short way.   Normally a programmer will think polymorphism is how polymorphism is implemented in the language they know the most.  But Java polymorphism is not really the same as C++ polymorphism.    Then you can start the discussion if method overriding and method overloading are polymorphism or not.

 

And even if the person is totally wrong about all of that, that still wouldn't say if the person is a good programmer or not.

1

Share this post


Link to post
Share on other sites
59 minutes ago, Krieg said:

 

Expecting that someone know by heart all the differences in the standards is really silly and doesn't say if the person is a good C++ programmer or not. 

 

I agree with this.  Moreover it's rarely up to individual developers which new features get leveraged - eg you've got an existing code base, are you suddenly going to convert all of your boost-based implementations to use the stuff that has since been incorporated into the standard?  Hardly ever happens.

 

and there is the problem of arriving at the lowest common denominator as well when an individual dev does go off on a toot.  Eg someone starts throwing lambdas around like candy (just because the can - "ohhh shiny!") and the rest of the team has no earthly clue what is going on.  

 

1 hour ago, Krieg said:

Normally a programmer will think polymorphism is how polymorphism is implemented in the language they know the most. 

 I don't agree with this at all.  Well, maybe, insofar as this is what separates "programmers" from "software engineers" - if your only grasp of polymorphism is in terms of implementations details, I think that is a problem.  The whole concept is about abstraction and if you don't understand abstractions (literally and figuratively) you are missing the boat wrt any object oriented language.  And yes I already stated these gaps come from the limitations of previous work experience and I do believe that they really don't know they're not senior level oo engineers :)

 

the point is, if *you yourself* claim to be an oo engineer (let alone a senior), you should be able to explain polymorphism as it's absolutely fundamental to oop/ood.  I don't think that is a stretch at all, 

0

Share this post


Link to post
Share on other sites
9 minutes ago, lisa13 said:

 I don't agree with this at all.  Well, maybe, insofar as this is what separates "programmers" from "software engineers" - if your only grasp of polymorphism is in terms of implementations details, I think that is a problem.  The whole concept is about abstraction and if you don't understand abstractions (literally and figuratively) you are missing the boat wrt any object oriented language.  And yes I already stated these gaps come from the limitations of previous work experience and I do believe that they really don't know they're not senior level oo engineers :)

 

the point is, if *you yourself* claim to be an oo engineer (let alone a senior), you should be able to explain polymorphism as it's absolutely fundamental to oop/ood.  I don't think that is a stretch at all, 

 

You cleverly edited out the most important part of my post related to polymorphism and moved to the next sentence leaving my now edited comment without a context.

 

Again, if you try to define polymorphism as a general thing you have to be extremely short, one simple sentence would be better.  Something like "The ability for a function, variable or object to take multiple forms".   When you start to add more things to that someone will tell you "Oh, but that's polymorphism in <insert language here>, it is not like that in <insert other language>".

 

If I am conducting the interview and I asked a silly question like "Tell me what polymorphism is" and the applicant told me whatever that points to polymorphism, that's good enough for me, even something silly like "Java already has polymorphism by just extending every class from the Object class".   Not that I would ask those kind of questions anyway.

 

I actually wouldn't care if a person applying for a programmer position does not know every single detail of all this.   These are things that can be easily learned.  

 

 

P.S., We do not let managers conduct technical interviews because they might end up like @MikeMelga asking about relative new standards and thinking that not knowing them means you are not worthy.   The manager of the position can be present in the technical interview, but it is normally conducted by a technical person, like a lead programmer, an architect, and so on.

1

Share this post


Link to post
Share on other sites
1 hour ago, Krieg said:

This is a very silly question to ask and deciding if the person is skilled in C++ based on this question is extreme and actually borderline dumb.

After all, candidates may decide that this particular company is not the right one for them after questions like this one.

 

A candidate can :(){:|:&};: a job interview with wrong answers, yes, but the interviewer can as well :(){:|:&};: the interview by asking wrong questions...

 

2

Share this post


Link to post
Share on other sites
42 minutes ago, Krieg said:

 

You cleverly edited out the most important part of my post related to polymorphism and moved to the next sentence leaving my now edited comment without a context.

 

Again, if you try to define polymorphism as a general thing you have to be extremely short, one simple sentence would be better.  Something like "The ability for a function, variable or object to take multiple forms".   When you start to add more things to that someone will tell you "Oh, but that's polymorphism in <insert language here>, it is not like that in <insert other language>".

 

If I am conducting the interview and I asked a silly question like "Tell me what polymorphism is" and the applicant told me whatever that points to polymorphism, that's good enough for me, even something silly like "Java already has polymorphism by just extending every class from the Object class".   Not that I would ask those kind of questions anyway.

 

I actually wouldn't care if a person applying for a programmer position does not know every single detail of all this.   These are things that can be easily learned.  

 

 

P.S., We do not let managers conduct technical interviews because they might end up like @MikeMelga asking about relative new standards and thinking that not knowing them means you are not worthy.   The manager of the position can be present in the technical interview, but it is normally conducted by a technical person, like a lead programmer, an architect, and so on.

 

there, are you happy?  Please ratchet down the paranoia - I did not "cleverly" omit anything, merely quoted the part I was responding to. 

 

You've got it wrong imo.  The answer to a polymorphism question does not have to be short, but it should be GENERAL.  the concept,value and proper use of polymorphism does not change based on the language and that is precisely my point. 

 

You don't get it.

1

Share this post


Link to post
Share on other sites
1 hour ago, dj_jay_smith said:

After reading this, I'm almost scared to say that OOP is dead!

 

(discuss!)

 

ha!  sadly I think you are right. 

 

But pray tell what superior methodology has replaced it in the realm of large scale, complex applications?  I'm more interested in a discussion on that :)

 

1

Share this post


Link to post
Share on other sites
1 hour ago, franklan said:

After all, candidates may decide that this particular company is not the right one for them after questions like this one.

 

A candidate can :(){:|:&};: a job interview with wrong answers, yes, but the interviewer can as well :(){:|:&};: the interview by asking wrong questions...

 

 

oh yes. The whole point of an interview is for both sides to feel each other out and any interviewer who doesn't understand that is shooting him/herself in the foot.

 

IMO the biggest mistake interviewers make is asking questions that have nothing to do with the project at hand or the current dev practices of the group.  I had one interviewer "quiz" me with a litany of ridiculous, overly complicated, HORRIBLE code snippets - I could answer the questions (this seemed to inspire him to keep going - gah!) but the whole line of questioning was utterly pointless.  I finally asked him:  "do these examples represent things I'd find in your current code base?"  "ohhhh no - we would never do something like this!  it's awful!"  "yes.  it is...sooooo?"  I excused myself shortly after.      

 

I find it equivalently stupid when the first question is to write a sorting algorithm,  WHY?!  do you actually roll your own in this day and age?  It's always fair game to discuss the fine points of algorithm efficiency but IME that's not the way to do it.  Any ninny can memorize a bunch of dumb CS 101 algorithms like this.  It tells you very little.  I prefer to provide a few different algorithms then try to get a candidate to discuss the pro's and cons of each approach.  My favorite questions are "what's wrong with this picture?"...those are really fun.

 

2

Share this post


Link to post
Share on other sites
17 hours ago, engelchen said:

What do people lie about? 

 

I have worked on a "job application verfication" system.  I was in the uk, so maybe its different elsewhere.

 

A huge number of people move start and end dates around a little to cover gaps, or modify job titles or details of what they did, but a suprisingly large number of people flat out lie about big things like qualifications (degrees) or jobs that they have never held. 

 

I have myself interviewed one guy who had on his CV "ran a training course on EJB" but in the interview could not tell me what an EJB is and when pressed admitted that it was a course run by a school he worked for and not something he himself was involved in.

 

I have also had friends ask me to write them a reference for a job they didnt have. I didnt do it, but I heard other people were happy to write references from their companies.

 

All in all, Im pretty sure that a large proportion of applications are seriously dishonest.  The company I worked with claimed about half of all applications have major lies (a degree, or job that was completly fabricated).  Dont know if thats true, Id assume their biases inflated the number somewhat.

 

17 hours ago, engelchen said:

Do they really think that they could get away with it?

 

I used to know someone, though I wouldnt call them a friend, who claimed to have got their job in the civil service (I think it was in the prison service, though my memory is a bit hazy) partially due to A level results that she had simply invented.  

 

We have rejected some applicants because we didnt believe their qualifications, but I think that in general people do indeed get away with it most of the time.  

 

I guess its easier in the UK as we tend not to send copies of qualifications and actual zeugnisse etc.  I know that very few companies actually bother to take up references at all.  Germany is different of course, even then Id guess most low res scans are taken at face value and there must be a fair amount of forged documents that noone checks up on.

2

Share this post


Link to post
Share on other sites
33 minutes ago, lisa13 said:

 

ha!  sadly I think you are right. 

 

But pray tell what superior methodology has replaced it in the realm of large scale, complex applications?  I'm more interested in a discussion on that :)

 

 

Disclaimer:  I am not a programmer, so I am not in all the details.

 

I read an article recently (unfortunately I no longer have the link) where the author was saying OOP is dead and that he is doing functional programming and that this is the future.

I didn't agree with this fully.  I think that each method has its applications.  Certainly, the application of OOP in all situations is not right, but neither is dismissing OOP as dead.  Functional programming might be making a comeback, but we should not dismiss OOP just because it is no longer trendy.

 

 

1

Share this post


Link to post
Share on other sites
23 minutes ago, dj_jay_smith said:

Functional programming might be making a comeback, but we should not dismiss OOP just because it is no longer trendy.

 

Why can't we just combine things?  What would you say about a mostly OO language but with some functional programming features?  We could share syntax with Java so people can get convinced easily to move to it and even let them use Java itself too withit in.

 

Well there you have Groovy.   My limited experience with Groovy (using Gradle) has been a love-hate relationship.   Sometimes it feels like an Apple product, it just works (and I have no idea why).

0

Share this post


Link to post
Share on other sites
16 minutes ago, lisa13 said:

wow...so what's very old and clunky is new and sexy again?  Ha!

 

fucking hipsters 

Porting 35year old PLC code to a new (ahem, 10 years old) generation of PLC makes sexy money... It's old farts like me and not lumberjack imitating tarts who do that...

2

Share this post


Link to post
Share on other sites
5 minutes ago, franklan said:

Porting 35year old PLC code to a new (ahem, 10 years old) generation of PLC makes sexy money... It's old farts like me and not lumberjack imitating tarts who do that...

 

yeah but that's just moving one grave to another - not digging a new one :)

 

(I really am teasing.  truly)

0

Share this post


Link to post
Share on other sites
4 hours ago, Krieg said:

 

You've got me until this.   This is a very silly question to ask and deciding if the person is skilled in C++ based on this question is extreme and actually borderline dumb.

I asked a very broad question: C++ standards. From there I develop the rest of the conversation. Example: if the guy says std threads, I develop the conversation into multithreading topic. If he goes into smart pointers, I go into that direction. If he says templates, he clearly has no idea and completely overstated his C++ knowledge.

 

4 hours ago, Krieg said:

 

Expecting that someone know by heart all the differences in the standards is really silly and doesn't say if the person is a good C++ programmer or not.  And actually Variable Templates is one of the differences in the standards.

C++11 was a landmark. If a C++ developer does not know that, then he is not a "above average" developer as they call themselves.

 

4 hours ago, Krieg said:

 

If you are a Windows developer, Visual Studio 2017 still didn't finish implementing everything in standard 14.

Again C++11 was the big thing. And there is much more to it than VS on Windows. We are using C++17 in windows, for example, because our compiler supports it and we found some usage for the new features.

1

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now