Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
The price of "paper" certifications

By blixco in News
Wed Jul 05, 2000 at 11:08:54 PM EST
Tags: Round Table (all tags)
Round Table

Recently I have been involved in interviewing new candidates for a QA-type testing position at my place of employment. It's common practice for my employer to have team members interview prospective team members. Most of us are techs or engineers and have little to no interview experience, so we stick to the roughest technical questions we can think up. I also like to throw in a curve...questions like "why are manhole covers round?" or "complete this sequence: OTTFFSSE...."

Update [2000-7-5 21:11:27 by Inoshiro]: I fixed the formatting :)


In any event, the candidates lately that our HR department has thrown to us have been, to put it bluntly, complete morons. To be more specific, the candidates can't answer most basic questions about computing, networking, or troubleshooting. This in itself would not be so bad if the candidate had not gone through a significant screening process...one that looks for the right combination of alphabet soup (MCSE, CNE, CCIE) and degrees over experience. Ultimateley, the candidate leaves in a cold sweat and our day is shot (interviews can take five to six people and can last four hours). The candidate is convinced that we are Evil, and we are becoming convinced that Everyone Is Stupid.

My chief concern is that a lot of small companies do not have technical people to interview their prospective IT people. They ask a few basic questions and get a few basic answers, and are impressed by a certification or three. They don't know enough to ask about specific issues, they cannot be expected to know enough to challenge a paper MCSE or CNA or what have you. The end results are poor network setups, terrible security on public boxes and networks, and a general bad taste in the mouth of corporations concerning IT, among other things.

Does certification mean nothing? I know the worth of an MCSE...they typically don't know much outside of the MSFT test study guides, and aren't typically very experienced. Those that are tend to use the certification as a shield, propping it up to fill in gaps in experience or education. Is there a standard? What would it take to enlighten the corporate world about the lack of essential, basic knowledge with most (greater than 60 percent in my experience) certified technicians? How to get the word out?

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o Also by blixco


Display: Sort:
The price of "paper" certifications | 137 comments (135 topical, 2 editorial, 0 hidden)
A little one-sided... (4.00 / 3) (#3)
by Matthew Weigel on Wed Jul 05, 2000 at 09:42:13 PM EST

The problem -- and the blessing -- of certification is that it can only specify a minimum ability. When that minimum is relatively high, in particular high enough to include some theoretical knowledge on the subject of the certificate as well as some hands-on experience, the result is pretty good (like Novell and Cisco certificates, from what I hear).


--Matthew Weigel
It's not even a minimum ability. (4.80 / 4) (#4)
by adamsc on Wed Jul 05, 2000 at 09:50:43 PM EST

The real problem with certification is that most of the tests are standard and never change, allowing the cram strategy to work. Some people buy a cram book, pass the test based on what's in their short term memory and forget 95% of what was covered. There are countless examples of MCSEs who lack even extremely basic knowledge ("regedit? What's that?") to suggest that the current tests aren't parrotable.

The tests which really work are the ones that require someone to actually understand the subject and not just parrot back answers. As an example, one of Cisco's tests is a hands-on tests where they show you a network, keep you out of the room while they break the network (sometimes in very subtle ways) and let you back in to fix it against the clock. Something like this is more expensive because it requires more work on the part of the people administering the test but is also a much more accurate indicator of skill. A network administrator who can pass that test will probably be able to step right in at almost any company.

[ Parent ]
Re: A little one-sided... (5.00 / 1) (#5)
by blixco on Wed Jul 05, 2000 at 10:45:09 PM EST

I guess I didn't really mention that non "paper" certs are fine. People who have both experience based skill and certifications are tough to hire, though: someone already has hired them. As for the CCIE, most Cisco folks are extremely skilled. The test weeds out people who can't actually do the work. The lab test for your CCIE is a complete nightmare.
-------------------------------------------
The root of the problem has been isolated.
[ Parent ]
Re: A little one-sided... (4.00 / 1) (#7)
by Matthew Weigel on Thu Jul 06, 2000 at 01:08:22 AM EST

Well, all of the certs are "paper" certs. The problem is differentiating the good paper from the scrap paper. And that should be the question, not 'are paper certifications crap?' but, 'which certifications are crap?'


--Matthew Weigel
[ Parent ]
Re: A little one-sided... (none / 0) (#74)
by Anonymous Hero on Thu Jul 06, 2000 at 03:36:02 PM EST

I've met some major idiots with Novell certs, and when I go to Brainshare each year I see a lot of impressive talents (Joe Doupnik, for example) who haven't any. Novell certified engineers have never been of any use to me. The top Cisco jocks I know aren't certified either...
--Charlie

[ Parent ]
One way to look at it... (none / 0) (#89)
by Anonymous Hero on Thu Jul 06, 2000 at 07:41:17 PM EST

The resume is like a book's cover, and the ceritifications are like the quotes on the back cover. Don't judge a book by its cover, and judge the quotes by their author and always take them with a grain of salt.

[ Parent ]
The best "certification" is a degree fro (2.60 / 5) (#6)
by skim123 on Thu Jul 06, 2000 at 12:13:20 AM EST

At least you know someone who's completed a respectable university course with a respectalbe GPA has the smarts to at least learn what is needed, if he/she doesn't already possess that knowledge.

I have only directly hired one person in my entire life, but from my work experience (working with college interns and being an intern in the past myself) I find college interns to be the most productive, most excited, and most affordable solutions!

Money is in some respects like fire; it is a very excellent servant but a terrible master.
PT Barnum


Re: The best "certification" is a degree (2.00 / 1) (#8)
by Anonymous Hero on Thu Jul 06, 2000 at 01:40:19 AM EST

At least you know someone who's completed a respectable university course with a respectalbe GPA has the smarts to at least learn what is needed...

Fitting this description myself I must disagree. Often all the above tells anyone is that the individual knows how to memorise solutions to a few problems, or cheat on exams.



[ Parent ]
Re: The best "certification" is a degree (3.00 / 1) (#51)
by hurstdog on Thu Jul 06, 2000 at 12:18:39 PM EST

I agree. I am a CSE student at UC Davis and there are some people here who graduate and know next to nothing about CS or Engineering. All they know how to do is work the system. Some of the highest number of cheaters are in CS classes (most aren't caught, but you can tell who they are pretty easily). People emailing programs back and forth, reusing freinds programs from a quarter ago, and blantent cheating on tests are some things I see regularly. there are very few students who actually care about what they are learning and are not just trying to get a BS in CSE so they can start making money. Its sickening.

[ Parent ]
Re: The best "certification" is a degree (none / 0) (#106)
by Anonymous Hero on Fri Jul 07, 2000 at 10:23:34 AM EST

You guys come from lame schools... At UMO 100 freshmen will start the CS program and 12 will get the degree... It is impossible to cheat in all but the LOWEST level courses... we're not talking about C++ syntax classes here... How do you cheat at solving induction problems in a class of 12? How do you cheat at writing a spreadsheet complete with design specs for a professor who's been teaching this class for 20 years? Our CS program may not be the best in the country but at least it seperates the Geeks from the Wannabe's.

[ Parent ]
Re: The best "certification" is a degree (none / 0) (#115)
by Anonymous Hero on Fri Jul 07, 2000 at 03:55:57 PM EST

(from the first Anonymous Hero)

I am not convinced. So you have a 12% sucess rate. That tells me the UMO admitted 88 students who should not have been there. CS and ElecEng (my two fields) are not hard. If you find them hard, then you have lame ass proffessors who can't teach. (There may be a significant ammount of work, but this is not an indication of "hard to learn." It is an idication of psycho profs who may or may not be able to teach.)

hurstdog described it as "work the system." Cheating was a poor word choice on my part. I am thinking of the students who cram for a final or midterm without ever learning anything but memorizing how to use the cookbook solution. Or the students whose first step in doing an assignment is to search through all the past assignments to find the same problem with different numbers and copy it changeing the numbers were necessary. These students will come out with a degree and a great grade and no idea how to learn. They will bump a long just fine at work doing just enough to seem like they know what they are doing until they are presented with a hard problem. Then all hell breaks lose.

To answer the original question: I feel that university grades are hired in places were you really want a technician. University should have taugh more theory then practice, and theory dosen't work in practice. When hireing technical people we usually hire our techs starting on a 3-6 month term. If they work out, we keep them, otherwise we look for a new one. We are fortunate that we have a large enough dept that we can handle getting a few duds. I look for a bit of brains and some personality. It has been my experience that we could hire an arts major and make them a productive member in short order. To test for brains I will ask a few not so tough questions. I find that many people get so anxious in a interview it is a wonder they can remember their names let alone the answer to a tricky technical question. I think that the orginal poster, blixco, is looking for too much in a new hire. If you want someone who has the same know how as you, then you better hire someone with the same experience as you.



[ Parent ]
Re: The best "certification" is a degree (5.00 / 1) (#20)
by reggyt on Thu Jul 06, 2000 at 05:33:15 AM EST

I don't agree that the best certification is a degree. In terms of attaining a first position in any industry a good result from a good education establishment will give you an edge.

In my experience (at a software house for 10 years), I find that people having completed courses at university are bright and able candidates, but there are a significant number of well qualified, and here I mean that they achieved there results through hard work and ability, individuals who for some reason have no social skills whatsoever. For a team situation (any service company) and in my experience a software and services company, this is, well disastrous is a strong word, but its certainly detrimental to the smooth ebb and flow of life in the services sector and can lead to an increase in staff turnover.

I have been involved in hiring people in the past. In interview situations its often difficult to judge how much of a candidates personality you are seeing (or not) due the the circumstances of your meeting. I.e. An interview.

For me, chased up references are king.


"The importance of being earnest" -- A man called Ernest.
[ Parent ]

References!!! (none / 0) (#99)
by yuri on Fri Jul 07, 2000 at 02:50:56 AM EST

I've been reading this whole thread (at least the top 60 posts or so, OT but great to see a good discussion here) and I was amazed not to see any one mention references until yourself.

I am not in the computer IT biz, rather I am an acamedic(sic) scientist. When we interview candidates, be it for admission to graduate school, postdoctoral positions or (gasp) faculty positions, the "certifications" (read pedigree, degrees? from where?) get our attention, the next thing we look at are References! Bottom line is that the candidate will never get an interview without good references. Even with this process (we generally have the luxury of too many great candidates for too few positions) we still end up interviewing idiots occasionally. Usually, a careful re-read of a reference letter reveals less then complete faith. The candidates that have those "walk on water" reference letters usually DO walk on water.

So that brings me to my question? Why have so few people mentioned references. Is it because many here are hiring people fresh out of school or stealing talent from other companies, ie situations where the candidate isn't likely to have many refs available? Or has the general assumption been that people will give good refs in order to get rid of someone from their company and therefore you can't trust the letters.

[ Parent ]
Re: References!!! (none / 0) (#128)
by Anonymous Hero on Mon Jul 10, 2000 at 10:09:00 AM EST

Most references that I have seen on applications, and most references that I have put on applications myself, have been worth about as much as the paper they were written on. I can see how in an acedemic situation it might be different as reputation seems to be quite important.



[ Parent ]
Re: The best "certification" is a degree (none / 0) (#22)
by e7 on Thu Jul 06, 2000 at 05:53:59 AM EST

My company once hired a BSCS (basically on the spot) to assist with our Windows 95 rollout. He was assigned the task of analyzing the configurations we would need to build.

After two weeks of looking busy, he emerged with a stack of screen shots depicting the various icons that our users had on their machines ... along with useful notes like 'What goes hear?'

What's sad is that he personally interviewed with the MIS director, who passed him with flying colors. (Are Dilbert references considered cliché yet?)



[ Parent ]
who needs certifications? (3.50 / 2) (#9)
by sergent on Thu Jul 06, 2000 at 01:54:45 AM EST

Bottom line, I don't think it's really possible to do a good job of interviewing someone for a technical position without giving them a technical interview. There are too many things that you have to understand to be able to ask good questions about. Most HR types aren't good at technical interviews. We usually do 4 technical interviews (with engineers), and an interview with the manager at the beginning or the end.

I can't say I've ever interviewed or met anyone with "alphabet soup" certifications who has been worthwhile. Maybe some of the people have had them, but if the certification is the most impressive thing you can find to put on your resume, then that's a problem.

A university degree can be a good sign of "knows how to learn things". People can get that skill elsewhere, too, but either way, it's a very important skill. Certifications do not do a good job of measuring that skill in my experience, because I have met people lacking that skill who have had many certifications.

I guess what I'm saying is that telling me you have managed a bunch of machines that N other people have used for the last year is a lot more relevant than telling me that you took a test that says you know how to. If you think that basic system administration is all that challenging, that's probably a bad sign (I'm not saying it's not time-consuming, just that it's not usually very challenging once you have learned how to do it).

The classes on this stuff seem bad too. You hafta be able to learn this stuff on your own or it's not going to work out. System administration is that way--you have problems and you have to figure out how to fix them. Good problem-solving skills are just as important, if not moreso, and the certification tests can't do an adequate job of measuring that. If you've ever taken a lab course at a university and had practical exams where you had to solve a problem in the lab within the period--realize that some people really like those exams (i.e. me) and some people have a hard time with them. Being a {system,network,database} administrator is like taking a lab practical every week.

I get really bothered when I know how to do the job of the corporate IT people better than they do. Unfortunately I don't get to interview them...

--jss, probably a bit too cocky about this.

Certs are worthless. (none / 0) (#11)
by Sylvestre on Thu Jul 06, 2000 at 03:01:42 AM EST

I look on certs (the paper kind, not the breath mint!) as totally worthless. The only time I understand people getting them is when they have to in the normal course of work. I used to work for a company that paid a $500 bonus and an extra $500 salary for the first two certs you got a year. They also paid for the classes and training, so it was a Good Thing.

When I worked at Microsoft, you could take the beta classes and tests for Microsoft certs. If you did so you got a discount on the real cert test when it was released. Lots of people took advantage of that.

When I worked at Sun, Java certs were essentially free. Everyone took the classes. This may or may not still be true, but people who picked up those certs were doing so to test the certification process and learn about Java. The cert itself was secondary.

I'm sure there are similar programs at Oracle and Netware and HP and the like. That's the only time I understand certs. Beyond that, if you're getting a cert, I look at it like dingle balls on a resume. It's just stupid.

IMHO

-- Firearms are the difference between free people and subjects.
Certifications are for "Armor Piercing". (none / 0) (#12)
by joelreed on Thu Jul 06, 2000 at 03:04:44 AM EST

I recently completed a job search and am also working on several "paper" certifications. I have a few years of good experience behind me, but I was finding it very hard to get in at places. I had to "boil" myself down almost to the point of uselessness to get HR depts. and recruiters to even comprehend my skill set and that skill set is certainly nothing out of the ordinary for IT. I have a farily good position now that is contributing to my skills and I don't feel that the preperation and learning involved in the cert. process is really contributing to the skills that I have. I think that however it will get my resume passed the HR/Recruiter BS, and to someone who knows what the hell their looking at. Am I going about this job search the wrong way? I think networking is important, but I'm early in my career and haven't built that up yet. Does anyone else have problems getting considered for positions based soley on experience? I don't think that at this point in my career I have "killer experience" but I don't think that I can compete toe to toe with the guy with no experience and a MCSE cliffnote set.

The Problem Is... (4.00 / 2) (#13)
by iCEBaLM on Thu Jul 06, 2000 at 03:09:20 AM EST

The problem is that even us clued seem to need these peices of paper to "prove what we know" just to get by HR and into an interview, thats the whole point of an interview really. I mean if you're finding that you're weeding out a bunch of idiots during the interview process, then the process is working, just not efficiently. Your HR staff need to redefine the criteria on which they grant interviews it would seem.

You have to realize that a lot of great potential system admins or technical support staff may not have gone to college because the whole school system turned them off (they're geeks remember! they got bored and picked on and don't want to repeat the process, let alone pay to repeat the process) and what they know is learned through running their own boxes at home and helping friends and family members.

-- iCEBaLM

Re: The Problem Is... (5.00 / 1) (#56)
by Rand Race on Thu Jul 06, 2000 at 01:09:34 PM EST

I cut my teeth on an IBM 360 at my father's workplace. I've owned home computers (Commodores, Apples) since the late 70s (and I was born in 71). I've run Unix boxes since the mid eighties, IBM PCs just as long, and have tons of vocational electronics training (ISCET certified with microcomputer specialty). Unfortunately I was a complete screwup in school and have no degrees or diplomas. Even though I had two years of experience I had finagled my way into (I can fix that!), I had a hell of a time getting a job in the field. So I dropped $250 to get an MCP and got a job a week later... the funny thing is that M$ cert got me a Sysadmin job working with predominately Macs and BSD machines (thank the gods).

So yea, these things are great for impressing the non-savy and often that's what is required. But as a indicator of technical knowledge, they aren't that great.

I've seen it work the other way too. My brother is one of the finest PC techs I have ever known but has the worst luck with certification tests. He's working for a headhunter that cannot get him a decent job due to lack of paper, but employs him personally because the headhunter has seen him in action. Wierd.


"Question with boldness even the existence of God; because if there be one, He must approve the homage of Reason rather than that of blindfolded Fear." - Thomas Jefferson
[ Parent ]

Why are manhole covers round anyway? (none / 0) (#14)
by torpor on Thu Jul 06, 2000 at 03:11:05 AM EST

Something to do with the strength (resistance of buckling) of a circle compared to a straight side? I don't get it. Sorry if I'm stupid, but I've never even thought about this question before.

I do know the answer to the sequence question though ... "NT, etc".


j. -- boink! i have no sig!
Re: Why are manhole covers round anyway? (none / 0) (#15)
by Inoshiro on Thu Jul 06, 2000 at 03:14:36 AM EST

Actually, it has to do with the properties of circles. There's no way to jam a circular man hole cover down the hole if it's slightly larger, but a square one can easily slid in and hurt a work[wo]man if it's angled just so. So it's a safety feature :-)

--
[ イノシロ ]
[ Parent ]
Re: Why are manhole covers round anyway? (none / 0) (#26)
by Anonymous Hero on Thu Jul 06, 2000 at 09:12:51 AM EST

They are also easier to move around.

[ Parent ]
wrong... (none / 0) (#124)
by Anonymous Hero on Sun Jul 09, 2000 at 01:01:35 AM EST

Being easy to move is NOT a intended feature. 1) Manhole theft is a problem. 2) You never need to move a manhole cover more than a couple feet from the hole.

You never need to roll them.

The safety features of being round outweigh these considerations.

[ Parent ]

Not just that (none / 0) (#28)
by jabber on Thu Jul 06, 2000 at 09:40:34 AM EST

If one (cover) falls in the manhole, it's a b!%@# to carry back out, so we keep them from falling in in the first place. Those things are DAMN heavy.

[TINK5C] |"Is K5 my kapusta intellectual teddy bear?"| "Yes"
[ Parent ]

Oh, doh. Wait a minute. (none / 0) (#16)
by torpor on Thu Jul 06, 2000 at 03:15:02 AM EST

I found the answer:

http://www.bluegum.com/Humour/Quiz/LatSol1.html

Which, of course, brings up an interesting point ... how many cert tests give you access to the Internet to find out answers to questions you don't know? Shouldn't extra cert points be awarded for those that:

a) Admit they don't know something
b) Go find an answer quickly on the Internet?

For me, this seems like a more important skill than knowing glib answers to arcane questions ...

What tests exist to measure how prone someone is to admitting they don't know something, but are then able to rapidly and competently find an answer? *THAT* seems like a certification worthy of respect, in my opinion..

(Weird how I started off not being able to participate in this article due to lack of interest (and I really honestly didn't know the answer to the manhole question), yet here I am ... contributing to the thread!)

j. -- boink! i have no sig!
[ Parent ]
Re: Oh, doh. Wait a minute. (3.00 / 1) (#18)
by Inoshiro on Thu Jul 06, 2000 at 03:37:33 AM EST

"What tests exist to measure how prone someone is to admitting they don't know something, but are then able to rapidly and competently find an answer? *THAT* seems like a certification worthy of respect, in my opinion.. "

I'd call any test which measures the ability of a person to figure something out or, if they can't figure it out, seek help quickly from someone who does know the answer an intelligence test. Just knowing something isn't helpful if you can't read and understand what you are reading (directions for installation of an operating system, or a man page for some particular daemon you have to setup but have no experience with), because every day has the potential to go beyond what you already know.

Obviously a simple paper with some alphagetti from a big company which also happens to sell the same product is not going to be good. This is why indepedant third parties (universities) provide a nice means to determine intelligence, because you need a lot of it (plus luck) to make it through the hairer courses ;-) Plus, if you're a Ph.D or similar, you have to have actually researched and proved something that has been peer reviewed. Hardly the same as a 40 question multiple choice test.



--
[ イノシロ ]
[ Parent ]
Re: using Internet during a quiz (none / 0) (#21)
by e7 on Thu Jul 06, 2000 at 05:43:25 AM EST

Shouldn't extra cert points be awarded for those that

  • Admit they don't know something
  • Go find an answer quickly on the Internet?

An interesting idea. I'm teaching myself basic Oracle skills (not trying to pass myself off as a DBA, mind you!) and every time we get a weird error message, the fastest way to find out what's going on is to look up the ORA-xxxxx message in context, on deja.com.

Almost all of my knowledge has been picked up informally. It always floors me when our DBA/developer/webmaster/whatever gives us the wrong answer and stalls, and we're able to locate the information without his help.

I guess the moral of the story is that since we live in a connected world, we might as well take advantage of it ...



[ Parent ]
Re: using Internet during a quiz (none / 0) (#48)
by Anonymous Hero on Thu Jul 06, 2000 at 12:02:37 PM EST

regarding your ORA-XXXX error messages: if you use the oraerr program, it may save you some time and it's quite handy (most of the time... sometimes it's just useless).

[ Parent ]
Re: Oh, doh. Wait a minute. (none / 0) (#81)
by lhand on Thu Jul 06, 2000 at 04:33:43 PM EST

One of my favorite sayings is "An expert doesn't necessarly know the answer, but he knows how to find it." I've found this to be true in every job I've had. Getting someone who can find out the solution to any problem is far more useful than someone who has memorized the solutions to a few.

[ Parent ]
Certs have a place - in HR... (2.33 / 3) (#17)
by Deus Ex Machina on Thu Jul 06, 2000 at 03:23:42 AM EST

Having been a sysadmin for not too long, it has nonetheless been my experience that the greatest pressure in an administration job comes not from the technical difficulty so much as the pressure of responsibility. Certifications are nice only in that they are what they appear to be - they say that you know "this stuff" but give no guarantees on anything that isn't covered by the official criteria. However, in terms of job competance, certifications are totally useless. Responsibility is the only means that I have seen of attaining respect in the tech world - the more responsibility you have, the more people depend upon you, and when you help those people, they _know_ you. This is how it seems to work from my perspective, and although I am very new as an administrator, I know that when I can make a computer do something that the person sitting at that computer couldn't, I have just jumped a notch in respect (and responsibility). Of course, certification looks pretty for the people in Human Resources, so sure - certification does have a place on your resume, and might make a difference in getting a job. But it has no place on the job itself, and no one cares how educated you are if you can't deal with the pressure and responsibility of Real Work.

--
I might have went off on a tangent there, but if you look really hard, you might find a point. Or not.

And HR has a place in the dole queue (none / 0) (#101)
by Anonymous Hero on Fri Jul 07, 2000 at 03:54:43 AM EST

Sorry.... just wishful thinking!

[ Parent ]
It would be nice, but hard (none / 0) (#19)
by duxup on Thu Jul 06, 2000 at 03:48:48 AM EST

In several work places I've often found that having techs involved in the interview or evaluation parts of an application is helpful too. Although I should note I usually try to assess the prospect without a inquisition to the point where they feel we're being rude. I sympathize with the difficulty of finding qualified people. For better or worse in more recent experience I often just looked for people whom I think can learn relatively quickly.

Personally I think certs are nice to see. I don't weight nearly them as much as experience. I believe it does show they've taken some initiative to learn at some point. It is nice that with a few certs of my own that we may have in common, I do then know some of what they do and don't know on a low level.

I often see lots of flaming and knocking of certifications that I think that is just due to the large number of certified people out there people's own insecurities. I've been in tons of situations where people bitch and moan about how much person A or person B knows and how person A has his/her CNE (or any other cert) but couldn't fix problem Y for 20 minutes . . . yada yada yada. I think much of the moaning about how valuable a certification is often has more to do with egos than any real discussion. Person A and person B will go out and learn as best they can and they'll do good or bad regardless of what certs they have or do not have.

As for general certs I doubt that will ever occur (but that would be darned nice). It would be hard to see say something like "TCP/IP" covered well by some general certification at any reasonable length and give someone close to the knowledge experience gives you. You would have to cover an awful lot to really cover the vast majority of popular TCP/IP implementations or applications, let alone the difficulty of teaching troubleshooting or diagnostic skills.

In the end certification will never come close to experience. For that matter education also helps but again falls short of experience.

Re: It would be nice, but hard (3.00 / 1) (#82)
by lhand on Thu Jul 06, 2000 at 04:52:53 PM EST

As for general certs

The ICCP has (had? I got certified in the 80's and my cert expired ten years ago.) a general certification program with a series of different specializations. They are not vendor specific and require a broad range of knowledge of computers and data processing to pass the core exam. They also have an experience requirement to earn a cert. When I took the test all those years ago (an all day exam) I was not so much impressed that I passed, but that so many of my co-workers did not. They took lots of prep courses for the exam, read study guides, etc. and still failed. I just had some years of experience and interest in lots of different subjects.

Alas the ICCP is not well known so I never bothered to keep up my certification. But I would consider a CCP certificate as an indicator of someone who had good general knowledge of computers.

I don't think I would say the same of any of the vendor specific certs out there. They only tell how deeply someone has been indoctrinated into the vendor's product.

[ Parent ]

Your comment re: Small companies (4.00 / 1) (#23)
by Photon Ghoul on Thu Jul 06, 2000 at 07:45:12 AM EST

I work in a very small company (4 developers, 1 dba, 1 qa, 1 desktop support, and 1 webmaster) and when we interview, they get grilled on their technical knowledge of NT, Solaris/Unix/Linux, networking, etc. I know that my interview made me quite nervous about the results....

Now, when I worked at a _very_ large bank which will remain nameless, I just had to impress some very non-technical management types.

Remember: generalities are stupid
no sig
Re: Your comment re: Small companies (none / 0) (#34)
by blixco on Thu Jul 06, 2000 at 10:50:18 AM EST

Remember: generalities are necessary.
-------------------------------------------
The root of the problem has been isolated.
[ Parent ]
Re: Your comment re: Small companies (none / 0) (#72)
by Anonymous Hero on Thu Jul 06, 2000 at 03:31:23 PM EST

"No generality is worth a damn, and neither is this one" -Geo. B. Shaw, in a radio interview, replying to criticism of his statement "Those who can, do; those who can't, teach."

[ Parent ]
Quick definition (none / 0) (#24)
by blixco on Thu Jul 06, 2000 at 08:43:51 AM EST

A "paper" certificate is a cert that is worth the paper it's printed on. The type of person who gets this certificate often has little to no experience, and became certified as a way to make a lot of money. I guess the real difference between a paper certificate and someone who actually *has* the knowledge is that the latter is actually interested in the technology, in learning it, and in doing the best possible job with skills that they try their hardest to learn and maintain.

My concern is that Joe Marketing Guy is getting his MCSE after having successfully booted an IMac twice. These are the types of people I am interviewing: people who just simply have no idea. They crammed for the test and passed, and now they are industry recognized computer professionals, all it took was money for the classes and books.

Comparisons: any martial arts folks out there? Have you seen the big corporate "dojos?" The ones where, with just 19.95 a week, you too can be a black belt? OK, how about artists...any artists out there? Would you consider someone an artist if they sent off some line drawings to a PO Box in a matchbook cover? The difference would seem to be the *depth* of knowledge and understanding of the subject.

And the typical small shop (or large non-computer related corporation) will hire a paper MCSE or CNE based on these certifications alone these days, since people are so hard to find.

My hope is that maybe a new, more visible cert can be added to the mix, one that is based entirely on actual knowledge, one whose test psychometrics involve changing formats and unknown, human controlled scenarios. This would at least help weed out those that have some base knowledge from those that cannot operate a can opener.

And not every MCSE is a moron. Just 60 percent of the ones I have interviewed.
-------------------------------------------
The root of the problem has been isolated.
Re: Quick definition (none / 0) (#36)
by Anonymous Hero on Thu Jul 06, 2000 at 11:03:13 AM EST

I teach computer applications part time at the local JC, as well as holding down a full-time job (not really IT, but computer related, OK?) One of the students indicated on the feedback form that she'd like to see all instructors MS certified. Yeah, right. I use the entire MS Office suite in my day job, and they're more than happy with what I'm turning out. Get a cert for a $15/hour part time job? OK, let me figure it out.... It would take me about 2.5 years at the part time gig to pay for the cert. Talk about your basic disincentive. I agree that certs have their place, but a firm grasp on realism in the HR department would be a big help too. "If you can't explain it to an 8-year old, you don't really understand it." -- Albert Einstein

[ Parent ]
Also..... (3.00 / 1) (#25)
by blixco on Thu Jul 06, 2000 at 08:50:27 AM EST

The next letter in the series is "N"

think:
12345678....9
-------------------------------------------
The root of the problem has been isolated.
Interviewing prospectives (3.70 / 3) (#27)
by jabber on Thu Jul 06, 2000 at 09:31:00 AM EST

Manhole covers are round for two reasons:
1. They're easier to move that way, they're heavy and rolling them is easiest.
2. Round is the only shape that WILL NOT fall through it's own hole when placed at an angle.

OTTFFSSE...
... NTETTFFSSENT...

Clever questions are good to break the ice, and to see if a person is actually thinking in the interview, or if they are on autopilot. Some people get very stressed out in a new situation, and this may not be the candidate you're looking for. Conversely, a person may be really anxious to get in the door, but can take technical challenges in stride once on the job. I hope you will not eliminate people for nervousness. Interviews are tough, especially for geeky types who would rather hack code than schmooze with HR contacts.

That said, bludgeoning a prospective employee with the toughest technical obscurity you can find is not a good tactic either. It immediately sets up an adversarial relationship and ruffles feathers. This new person's background is different then yours, and s/he may not be in the frame of mind to tackle your latest bug at the interview.

Two things are very important, and you should (IMO of course) try to identify these in the initial interview:
- How would this person approach a problem? Give them an example of a relatively small problem that you've recently faced. Ask how they would decompose it and solve it. Don't expect an actual solution, but pay attention to the steps, the logic and the attitude. Would they call the vendor tech support as step 1? They're a deadbeat. Would they ask team members about similar problems solved int he past? Probably a good trait - why reinvent the wheel? Would they rush in, solve it their way, but hide their solution? A prima-donna waiting to happen.
- How does this person 'jell' with your team? Do they have similar interests? Similar sense of humour? Similar personality? Are they just different enough to spark the interest of your group, to include them on the personal level? Are they 'friendly' or 'vicious' in their competitiveness? Good competition builds a good team, bad competition tears one apart. How do they give and handle criticism?

Skills can be learned over time, but if you can identify a person who thinks in a logical, project/team oriented way, and can work with the other people on the team, then you have got a strong candidate. Hiring an antisocial encyclopedia is not a good bergain, hiring a new team member who is willing to put effort into the work, has something to contribute and isn't afraid of learning, is a good bet.

If you want to test technical knowledge, give some warning. Consider the initial interview a test of thinking and of personality. If it works out well, ask the candidate to come again for a follow-up 'technical' intertview. Give them a week to think about it, and let them know the sort of skills you're seeking. No one can learn Java or Perl given only a week, but they CAN brush up on a rusty skill. They can learn an obscure feature or protocol in that time, and will be prepared to demonstrate competency - or they'll tell you straight out that they lack this skill but are willing to learn it.

In the first interview, both you and your candidate are flying blind, and setting landmines (as you're already pointed out) intimidates the prospect and frustrates the interviewers.

How would YOU like to be treated in the initial interview?

[TINK5C] |"Is K5 my kapusta intellectual teddy bear?"| "Yes"

Re: Interviewing prospectives (none / 0) (#31)
by FFFish on Thu Jul 06, 2000 at 10:35:22 AM EST

Circular manhole covers are *NOT* the only shape that won't fall through the hole.

But I'm not going to tell you the other shape. :-P



[ Parent ]
Re: Interviewing prospectives (none / 0) (#32)
by jabber on Thu Jul 06, 2000 at 10:44:18 AM EST

Equilateral triangles don't roll half as well. :P
Spheres are not fair, they make the road too bumpy. ;)

[TINK5C] |"Is K5 my kapusta intellectual teddy bear?"| "Yes"
[ Parent ]

Re: Interviewing prospectives (none / 0) (#33)
by squigly on Thu Jul 06, 2000 at 10:44:31 AM EST

Triangular. Not so convenient for people though.

--
People who sig other people have nothing intelligent to say for themselves - anonimouse
[ Parent ]
Re: Interviewing prospectives (none / 0) (#44)
by spankenstein on Thu Jul 06, 2000 at 11:44:46 AM EST

In Marilyn VosSavant's column a few weeks ago, this question was actually brought up.

Circular manhole covers are best when there is a lip. This is like manhole covers that we see everyday. There is no possible way for it to fit through. If we were building manholes without a lip equilateral triangles would be best.



[ Parent ]
Re: Interviewing prospectives (none / 0) (#53)
by jimhill on Thu Jul 06, 2000 at 12:32:50 PM EST

Vos Savant is wrong so often it's not even funny. On the plus side, she is a perfect example of the limited utility of "IQ testing".

_ALL_ polygons can be used as manhole covers if the hole is lipped. The purpose of the lip is to keep the cover from falling through. Most polygons can be oriented so that they fall through the hole. Triangular covers cannot, which one might think makes them ideal. One would be wrong. Triangular manhole covers (1) have to be oriented for replacement, (2) create stress concentrations at the corners, (3) don't match the cross-sectional shape of the human body, and (4) don't roll worth a damn. Circular covers suffer from no such shortcomings.

I'm too disinterested to look it up, but several somebodies out there have webpage after page after page detailing the frequent shortcomings in Marilyn Vos Savant's responses to questions.

[ Parent ]
"Marilyn is Wrong" site (none / 0) (#100)
by Patrick Bateman on Fri Jul 07, 2000 at 03:38:38 AM EST

can be found here.

---
I have to return some videotapes.
[ Parent ]

Re: Interviewing prospectives (none / 0) (#137)
by FFFish on Tue Jul 25, 2000 at 12:04:06 AM EST

Whoops. I lied. I stumbled across my "prior postings" list and thought I'd check out the replies. So, now I'll also give an answer:

The other shape that won't fall through is a "soft" triangle. No idea what the shape is called, but it's a triangle made with curved lines; I believe the curve's centrepoint is the opposing corner. Equilateral, of course.

Drawing it: Draw an eq.tri., place a compass on one corner, make its radius = distance to other corner; draw curve. Rinse, repeat.

IIRC, etc.

[ Parent ]
Re: Interviewing prospectives (none / 0) (#70)
by Anonymous Hero on Thu Jul 06, 2000 at 03:25:51 PM EST

Yeah, but you have to ask some technical questions. I interviewed one candidate for DBA who claimed to have extensive experience working with terabyte-size databases, and when I asked some questions that I thought were so simple I apologized when I asked them, he started making up nonsensical answers as he went along, just trying to bluff his way through. If somebody has good problem-solving skills that's great, but if that's all he's got I want to know about it up front.

[ Parent ]
Re: Interviewing prospectives (none / 0) (#105)
by jabber on Fri Jul 07, 2000 at 09:30:24 AM EST

Agreed. Hence the need for a technical interview. And there certainly are justifiable technical questions... The depth to which they run should be relatively low as far as your needs are concerned though. Simple questions are good, they don't just test knowledge but also wether a person can think technically in a stressful situation.

I'd go only as deep as the average type of problem that might be faced in the course of an average day. If you do a lot of javascript web design you'd ask about mouseover or some such... When working with Java, some typical API questions or parameter-passing oddities are fair game, since this sort of thing is 'in-your-face' every day. But, speaking of Java, I would not ask about the intricacies of garbage collection unless your perticular environment is very tight and the candidate MUST know such 'details' implicitly.

[TINK5C] |"Is K5 my kapusta intellectual teddy bear?"| "Yes"
[ Parent ]

On rectangular manhole covers (was Re: Interviewi (none / 0) (#112)
by converter on Fri Jul 07, 2000 at 12:55:02 PM EST

My last real job, back in the eighties, was at a rotational molding plant (LLDPE). One of our products was a rectangular pad that was installed on a lip near the bottom of a particular company's rectangular manholes in order to prevent damage to both the concrete structure at the bottom of the hole and the manhole covers, which for some reason tended to fall into the holes quite often <grin>.

I wonder if the lawsuits from injured workers, pedestrians and auto insurance companies put the manhole folks out of business.



[ Parent ]
Where I work (1.00 / 1) (#29)
by Ex Machina on Thu Jul 06, 2000 at 10:05:55 AM EST

I'm an intern (incoming college freshman) at a small .com where the development platforms are C++ and VB. The entire interview process for me was basically my boss (the CTO) trying to crush my will with evil, evil questions on C++ syntax and internals. Basically he tried to rip my still beating heart from my body through a complicated quiz on the heap and pointers. I got the job though.
/* ooka looka */
Certifications aren't all the same (none / 0) (#30)
by DemiGodez on Thu Jul 06, 2000 at 10:25:51 AM EST

I think it is important to note that all certifications are not the same. Often, they are misunderstood as to what they are. I don't have any experience with MS certs, but I know a lot about Lotus Notes certs and Java certs. The Lotus Notes ones totally suck - they are worthless. Anyone could sit down and pass them even if you hadn't seen the product.

On the other hand, I think the Java certs are pretty good. I've taken the programmer and architect ones and I have talked to a lot of people about the developer one. They are IMHO, very good at testing what they are designed for. The problem comes when they are misunderstood. For example, the programmer test tests language knowledge. So, datatypes and operators and really core language stuff. But a lot of companies will hire to people to do, for example, servlet programming, based on that cert. That cert has nothing to do with servlets specifically - just language.

If more research was done on what the certifications were for, that would help.

(3.00 / 4) (#35)
by fuzz on Thu Jul 06, 2000 at 11:00:48 AM EST

it boils down to this:

certifications get you in the door.
experience keeps you there.
degrees make you promotable.

period.



Re: (none / 0) (#40)
by Anonymous Hero on Thu Jul 06, 2000 at 11:21:23 AM EST

And where do *you* work? I'm not sure I want to work there. If certifications are all that get you in the door....must be a pretty evil place to try and solve a problem.

[ Parent ]
Re: (5.00 / 1) (#110)
by tzanger on Fri Jul 07, 2000 at 12:15:57 PM EST

certifications get you in the door.
experience keeps you there.
degrees make you promotable.

Interesting, I would have said it this way:

Degrees get you in the door
Common Sense keeps you there
Experience promotes you



[ Parent ]
Re: (none / 0) (#127)
by fuzz on Mon Jul 10, 2000 at 08:26:46 AM EST

degrees can get you in the door as well, true.

when i say "experience keeps you there" i mean that common sense and knowledge that you gain from experience keep you there. so we're of the same opinion on that one.

if experience and/or certifications get you a job at a company, great. but without a bachelor's degree, you won't be going anywhere on the promotional ladder, regardless of your experience level.

certifications are a great way to be considered for a job in a high tech field if you don't have a bachelor's degree. if the cert gets you through the first round or two, or even getting you the job, great. then you have to rely on your skills (your experience) to perform that job well. then, you need a degree in order to be promoted (99% of the time).

plain and simple, a lot of emphasis is placed on degrees. your boss can't promote you very high, because at some point in time he might have to hire someone to work under you that might possibly have a degree. and generally, that just doesn't work well. so if you don't have a degree, chances are you'll sit on the bottom rung.

[ Parent ]
Re: (none / 0) (#131)
by tzanger on Mon Jul 10, 2000 at 05:48:08 PM EST

if experience and/or certifications get you a job at a company, great. but without a bachelor's degree, you won't be going anywhere on the promotional ladder, regardless of your experience level.

I don't agree with this at all, but to continue your quote...

plain and simple, a lot of emphasis is placed on degrees. your boss can't promote you very high, because at some point in time he might have to hire someone to work under you that might possibly have a degree. and generally, that just doesn't work well. so if you don't have a degree, chances are you'll sit on the bottom rung.

I agree that a lot of emphasis is placed on degrees. However the rest of your statement I don't agree with. Experience should always preceed a degree. Always. How's that saying go? Education is what's left after you've forgotten everything you've learned in school? That doesn't sound right but the jist is in there somewhere. :-)

If I've been at a company 5+ years and some wet-behind-the-ears punk out of university thinks he's getting ahead of me on the ladder just because he's paid $50k for his degree and I didn't go that route he's got another thing coming. Oh wait... That's why this sounds familliar. It happened to me. Twice. I'm still top dog in the department. Why? Because of my experience, drive and plain old dumb luck. That's what got me here and the drive and experience will continue to allow me to bring the guys up from entry level and teach them what really happens when the authors of the textbooks go home. Degrees don't get you promoted. Being valuable to the company does. If you've got your masters in x and a doctorate in y and you can't move and shake out in the industry you won't be promoted. Period.

Don't take my arrogance the wrong way. I am going on no sleep (bad night with the kids and chicken pox) and I know I'm coming off stronger than I want to. I don't ever imply that they don't know anything or that I'm smarter than they are, but rather that they don't know it all and that despite my lack of formal education my different outlook and life experiences may have a few advantages that the academic life has blinded them to.

So long as I can teach them things they didn't learn in school they tend to see me as a knowledge source and realize why they're not (yet) managing me. There's been one who came in and demanded a top position because of his degree and the ring on his finger but he got cut down awful fast. If I become lazy then yes, someone hungrier will take over. If I stop pushing then yes I'll be overtaken. But so will any of these guys. That's natural and the right way to do it; that degree don't mean anything.

I'm 24; I'm not looking at throwing in the towel for another 15-20 years. Three kids haven't slowed me down (much <g>) and I don't see myself sliding too far back in the immediate future.

Wanna know the best part? In North America (I hear it is very different in Europe) people hire and pay based on experience, not education. After your first job nobody really cares about your education, they only want to hear what you did and what you can bring to the table. At least in the technical industries. I would imagine that for sciences and whatnot it'd be quite different.



[ Parent ]
Okay, I admit it... [slightly OT] (3.00 / 1) (#37)
by Denor on Thu Jul 06, 2000 at 11:05:55 AM EST

  I'm horrid at logic problems. Absolutely terrible. I've bought those brain-teaser books with 100 pages or so of logic problems, and I'm able to get through about the first three pages. Then I'm done.
  This wouldn't ordinarily be a hindrance, except nowadays interviewers are using logic problems to determine whether you're qualified. I may not know logic problems, but I do know programming - while the same skills could be useful, I don't think my brain works quite the same way on logic problems as it does on programming problems.
  For instance, I once interviewed with Microsoft (quit looking at me like that! I was young and needed the money!) and they had one ["You are given six numbers in the range 1-5. Write a program to find the repeated number using only one loop"] which I really did try, but failed miserably. I never heard from them again.
  Am I the only one who has this trouble? Is anyone else out there a good programmer, but bad at these kind of problems? Or is my brain just mis-wired :)

  (And I understand the answers to the manhole question - those were obvious once pointed out. But how does one get the answer to the 'finish this sentence' question?)


-Denor


Re: Okay, I admit it... [slightly OT] (none / 0) (#38)
by Inoshiro on Thu Jul 06, 2000 at 11:19:28 AM EST

1,2,3,4,5
One extra num from 1-5

My pseudo code after considering for a moment (not too many minutes, I tend to be good at algorithmic programming problems ;)):
ASSUME numbers are in array nums[1-6]
ASSUME you get i and a second array (you could use something else to flag numbers)

int i;
int flags[6];

memset(flags, 0, sizeof(flags)); /* Clear flags so we don't get any false positives */

for(i = 0; i < 6; i++)
{
   if(flags[nums[i]])
      printf("Duplicate is %d\n", nums[i]);
   else
      flags[nums[i]] = TRUE;
}
printf("Done.\n");

The logic: you flag each number as you find it. If the flag is already on when you try to flag it again, you've a dupe. Print a message about it :)



--
[ イノシロ ]
[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#46)
by Denor on Thu Jul 06, 2000 at 11:52:37 AM EST

Dang....

  That would have been rather impressive to the person interviewing me, considering that works rather well, and it wasn't the answer he had... :)

  The answer they had was:
int target = 1 + 2 + 3 + 4 + 5;
int sum = 0;
int i;

for(i = 0; i < 6; i++) {
  sum += nums[i];
}

printf("%d is the duplicate\n",sum - target);
  Deceptively simple, but I couldn't lock myself out of the 'store the numbers we have, then search for them' mentality I had at the time :/
  And I still don't get the OTTFFSSE problem :)

-Denor


[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#50)
by joshv on Thu Jul 06, 2000 at 12:15:46 PM EST

You didn't say there was only ONE duplicate! :)

That makes it a little bit easier. I like their solution, although elegant, it is unlikely that most would come up with that during an interview.

-josh

[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#59)
by Inoshiro on Thu Jul 06, 2000 at 01:15:52 PM EST

You didn't say there was only ONE duplicate! :)

My program still printed each duplicate. You could've had inifite duplicates, and it would've flagged them all. It's fairly generic in that it only tests if a number is new, rather than it being duplicated. "Known" numbers are duplicates. (this is why I don't like the MS answer, it's fairly "specific" instead of being generic logic fit to a special case)



--
[ イノシロ ]
[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#57)
by blixco on Thu Jul 06, 2000 at 01:12:02 PM EST

ottffsse = first letter from each word of the following sequence:
One Two Three Four Five Six Seven Eight.....
so the next in the sequence would be "N" for nine.
-------------------------------------------
The root of the problem has been isolated.
[ Parent ]
Re: Okay, I admit it... [slightly OT] (4.00 / 1) (#71)
by Anonymous Hero on Thu Jul 06, 2000 at 03:29:12 PM EST

pretty poor question for the type of person you are looking for. The correct solution is not the mathematical analysis (which is how I tried to solve it) but an observational one. It is also locale based which makes it harder for those who don't use English as their first language. Unless their English is PERFECT they tend to think in their native tongue and have a quick 'interpretation layer' in their brains that translates things on the fly.

The manhole question is also quite puzzling because I have seen manhole covers that AREN'T round. In the same way I could ask you: "why are all power outlets in the UK equipped with on/off switches?". Just like with the manhole thing there are several valid reasons for justifying this. The correct one in this case is that you can disconnect the supply of current in case the cord or the plug catches fire.

If you want to get people that will fit your bill look at how they approach problems related to the job on hand. Pick a few problems that aren't trivial but also obvious enough for a competent professional to pick up straight away. For example one of the questions I asked in an interview was to find the bug in the following code:

/* FIXME: always prints "Bummer" */
int compute(int num)
{
  return num++;
}

int main()
{
  int num = 0;
  int n = 0;
  for(n = 0; n < 20; n++)
  {
     num = compute(num);
     if( num > 10 )
       printf("OK\n");
     else
       printf("Bummer\n");
  }
}
Even here I wouldn't take this as a benchmark of one's ability. It's a specific issue here that one may or may not have come across. There is no way to quantify person's cleverness. And in this light I'd make no assumptions about people's paper certifications. Some candidates will be worthwhile while others will be in it purely for financial reasons. Firstly try and weed out the latter.

While you might argue that this question is technical and requires knowledge of the C language it is much more universal (within the C programmers' world) than your English based one, two, three.

Being asked your 'brain teasers' in an interview I'd assume you are either not clever enough to ask any reasonable technical questions or so arrogant to expect everyone to show the thought pattern similar or identical to your own. Either way given a choice I'd probably decline your job offer.

Remember that even technical skills come in different flavours. Some people are better at low level nitty gritty optimization issues where others have a good eye for overall code design and keeping subsystems coherent and interfaces clean. You will exclude possibly bright people by asking them your 'trivial' questions and showing them their 'stupidity'. I didn't know the correct answers to your questions but I do have more than a over 500,000 lines of c++ behind my belt and that's just the code I wrote for living. When the original poster said he could write code but not solve puzzles, I'm prepared to believe. I'm just like him and I consider the two domains completely opposite. Programming and debugging requires one to have a creative mind whilst maintaing attention to details. Your questions test neither quality.

[ Parent ]

Re: Okay, I admit it... [slightly OT] (none / 0) (#86)
by FlinkDelDinky on Thu Jul 06, 2000 at 06:52:57 PM EST

Okay, I'm not even a programmer, my C skills are highly immature. Although I've read some books on C (for personal enjoyment) I've only done some simple stuff, maybe 50 lines (for a fractal type program I did using Mix C). And that was years ago. Although I've been getting back into it only because I'm reading the book "Beginning GTK+/Gnome Programming" (A+ book, I'll submit a review after I'm done and programmed a some).

Anyway, what I did was compile your program 'as is' and fixed some minor compile warnings. Now when I saw the num++ I thought that was probably the problem because 'what's going to be incremented after num (pre increment) is returned?' So I simply pre-incremented num. Oh wait, I just thought of a nasty way to do it, hold on, I'll be right back...

...Nope, I did a bunch of gnarly stuff using a pointer and while I found out that doing (*num)++ is <i>real important</i> I was unable to get post increment to work on the <b>return</b> line.

Obviously, if you've returned you cant even get to the post increment. But I learned alot about pointers trying. Thanks.

/* FIXME: always prints "Bummer" */

#include<stdio.h>

int compute(int num)
{
   return ++num;
}

int main()
{
int num = 0;
int n = 0;

for(n = 0; n < 20; n++)
{
   num = compute(num);
   if( num > 10 )
   printf("OK\n");
else
   printf("Bummer\n");
}
return 0;
}

[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#108)
by Anonymous Hero on Fri Jul 07, 2000 at 11:44:20 AM EST

You've worked it out well. Indeed this is the prefix/postfix operator bug. Brackets and pointers won't help in this case. But the overall point I was trying to make was that even though the question is quite C specific in our case it's still good enough to evaluate one's thought process. The question is sligthly more convoluted in the real test but still fairly evident. It tests one's attention to detail (an important factor for a programmer in a language that's poorly typed). It often proves that people use statements like
n++;
without thinking what really goes on. In c++ there often is a speed penalty attached to calling a postfix operator versus the prefix operator if it's called on a non basic type. Sometimes this may be an issue. I think you should carry on learning to program, you seem have the potential :).

[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#103)
by blixco on Fri Jul 07, 2000 at 08:17:18 AM EST

Keep in mind I'm not looking for a specific answer. Typically if I ask a question that has nothing to do with the position, I'm looking at the way the person thinks. If they think, that's a 1. If they don't think, that's a 0. You're assuming that all of our candidates have some specific knowledge about the job they are interviewing for. My problem (outlined in the original post) is that they do not have a clue about the subject at hand....which is OK if they can at least demonstrate something above brain stem function. I need people on my team that can think very quickly and act on that thinking. If I use certification as a measure of that thinking ability, I'm screwed. If I use *any* canned interview question, I'm screwed. So I like to throw them, and ask either some type of brain teaser or a mathematic / scientific / geographic question.

I do not base my recommendation for hire on the answers to these questions any more than I base my recommendations on the candidates certification (or lack thereof). It would be completely ignorant to do so, and I pride myself on being not completely ignorant. Maybe just mostly ignorant.

I don't ask difficult questions to point out how stupid people are...in fact, during the interview, I never give them the answer to the question I ask. If they do not answer it, I don't. If they do answer it, they've satisfied my intent, and I can begin to ask more directed questions about their background and qualifications.

As far as your observation that not all manhole covers are round....that would be a perfectly acceptable answer over "I just don't know." This isn't a job that requires looking up answers, it's a job that requires *finding* answers to problems that haven't been thought up yet.


-------------------------------------------
The root of the problem has been isolated.
[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#58)
by Inoshiro on Thu Jul 06, 2000 at 01:12:43 PM EST

The letter problem relates to the first letter of a word... one, two, three, etc. It's weird, and I was puzzled until someone else pointed it out in a post.

As for their solution, I think it's a tag inelegant for the problem. If you expand that option to counting PIDs (as someone suggested), then their solution would be O(2^N), whereas mine is O(N) (they have to add every single number to a "target" variable, this breaks if the numbers are non-contigous [possibly], and adds the equivalent of an extra loop if you have to initalize target from an N instead of a constant). Typical half-baked code, IMO.. but if you read the source code to many applications, you'll find it (and worse) tends to be common :-(..

(Of course, despite this ability to learn rapdily and understand algorithmic logic, I'm apparently unemployable :-P)



--
[ イノシロ ]
[ Parent ]
Re: Okay, I admit it... [slightly OT] (4.00 / 1) (#109)
by tzanger on Fri Jul 07, 2000 at 12:10:08 PM EST

As for their solution, I think it's a tag inelegant for the problem. If you expand that option to counting PIDs (as someone suggested), then their solution would be O(2^N), whereas mine is O(N) (they have to add every single number to a "target" variable, this breaks if the numbers are non-contigous [possibly], and adds the equivalent of an extra loop if you have to initalize target from an N instead of a constant). Typical half-baked code, IMO.. but if you read the source code to many applications, you'll find it (and worse) tends to be common :-(..

Depends entirely on the situation.

Personally I never, ever spend a wad of time thinking out the most optimized solution to a chunk of code unless that code will be in a fast loop or must execute fast. It is a waste of my time and energy squeezing every last byte and/or cycle out of code that gets executed only a handful of times in the computer's timescale. Unrolling loops, trading memory footprint for speed, etc. is all done but only at the appropriate points in the code.

That doesn't mean I have shitty (half-baked) code. The entire process includes a lot of time away from the computer, just thinking about how I'd code the thing, what kind of structures I'd use, the memory layout and program flow. However speed optimization is only peripherially evaluated until I've actually done come coding and can profile the code to see where my bottlenecks are.

I'm just about at the end of the product development cycle for one of our hottest selling products. Most of that code (8k of PIC assembly) is optimized since the system it runs on is so tight. But the same type of thing applies even there: the main loop code got a quick once or twice-over for obvious stuff but only a few the interrupt routines which got hit every 16us or so got cycle-optimized. I spent the better part of a week getting rid of 2 cycles (1us worth of time) in the ADC interrupt code because it had to be as fast as possible.



[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#114)
by mezzo on Fri Jul 07, 2000 at 03:23:43 PM EST

yay! that was what i thought to do.. and wondered if it was too silly.

however, i did get another stupid question at an MS interview (*ducks and hides behind denor* hey, MS interviews are fun, if only for the logic games =P).
It goes something like this:
you have 3 barrels of fruits. one contains apples, one contains pears.. the other contains both. give an efficient algorithm to find out which barrel had what.


"The avalanche has started, it is too late for the pebbles to vote."-- Kosh
[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#135)
by kraant on Tue Jul 11, 2000 at 11:44:09 PM EST

you have 3 barrels of fruits. one contains apples, one contains pears.. the other contains both. give an efficient algorithm to find out which barrel had what.

This one is easy

Treat each barrel as a Stack

Pop a value off each stack... two will be of one value and one will be of the other...

You now know that the odd one out contains all apples, pink licorice or whatever...

Now just keep popping values off the other two until one of them returns the "wrong" value...

That solutions complexity is O at worst case scenario so I can't see a better solution coming along

*shrug*
--
"kraant, open source guru" -- tumeric
Never In Our Names...
[ Parent ]

Re: Okay, I admit it... [slightly OT] (none / 0) (#136)
by mezzo on Wed Jul 12, 2000 at 11:42:09 AM EST

That's more or less what I said, to take a fruit from each barrel, and then for the last two.. just tip one barrel or take one from each alternately.

If you tip one barrel, then its O(n) worstcase (where n is the # of fruits in the largest barrel) if you take one alternately from each barrel.. worstcase is O(2n).

I am not sure if there is a better answer, since the guy just sorta nodded vaguely at my answer.

"The avalanche has started, it is too late for the pebbles to vote."-- Kosh
[ Parent ]
Why this is hard..... (none / 0) (#120)
by Anonymous Hero on Sat Jul 08, 2000 at 02:20:23 AM EST

This problem is sneaky, because it exploits programmers habits of thinking - programmers specifically!

Simply summing the numbers and subtracting 1+2+3+4+5 was the first thing that crossed my mind looking at this. I was wondering if anybody would post that solution.

But then again, I'm a science/math person, not a professional programmer. A math person wouldn't think of doing it any other way.


Programmers are taught to use fancy data structures and algorithms that scale well with N, so we see all these solutions that first fill a counting array and then search it for repeated values. This is THE way to pick out N REPEATED NUMBERS -

it just happens to be a crummy way to pick out only ONE REPEATED NUMBER.



[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#43)
by joshv on Thu Jul 06, 2000 at 11:38:36 AM EST

For instance, I once interviewed with Microsoft (quit looking at me like that! I was young and needed the money!) and they had one ["You are given six numbers in the range 1-5. Write a program to find the repeated number using only one loop"] which I really did try, but failed miserably. I never heard from them again.

Unless there is something more to the problem, I don't think I would hire your either. Just loop through the list and increment a counter for each of the digits 1-5. At the end you have an array which contains the counts of for each digit 1-5 in your list, anything with a count of over 1 is a repeat.

If they want to get technical and say it would require a second loop to go through the counts array to find repeats, just keep track of any digit's count you increment to a value higher than one in the original loop.

Am I missing something here?

-josh

[ Parent ]

Re: Okay, I admit it... [slightly OT] (none / 0) (#45)
by Denor on Thu Jul 06, 2000 at 11:46:25 AM EST

Unless there is something more to the problem, I don't think I would hire your either.
  But that's exactly the point I'm trying to make - I like to consider myself a good programmer (I'm not the best, but I'm always willing to learn), I just happen to be completely horrid at logic puzzles. It might have something to do with context (a problem like "you have 10 unique orders to process, but there are actually 11 there - find the duplicate" might be easier to solve for me) or it might just be that I'm bad at logic, good at programming.
  I was hoping there might be others like me :)

-Denor


[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#61)
by Anonymous Hero on Thu Jul 06, 2000 at 01:47:32 PM EST

Ah... but that uses two loops - you need to iterate through the list at the end to find out which is a repeat.

what i would do is to have a 5 bit wide bitfield, do a simple bitshift and AND test for each number in the list, printing out the repeat if the AND test was true (and setting the bit if it was not)



[ Parent ]
Re: Okay, I admit it... [slightly OT] (none / 0) (#77)
by Qtmstr on Thu Jul 06, 2000 at 03:53:15 PM EST

That's easy. char numbers[6] = { num1, num2, num3, num4, num5, num6}; char counts[6]= {0,0,0,0,0,0}; for(int i = 0; i < 6; ++i) { counts[numbers[i]] += 1; if(counts[numbers[i]] > 1 return numbers[i]; } Voila.


Kuro5hin delenda est!
[ Parent ]
Or without the if statement (dangerous code though (none / 0) (#96)
by Anonymous Hero on Fri Jul 07, 2000 at 01:24:57 AM EST

int find_dups(int *numbers)
{
    char counts[6]= {0,0,0,0,0,0};

    for(int i = 0; (i < 6) && (++counts[numbers[i]] < 2); i++);

    return i;
}

ok, ok, very VERY cryptic, but as long as the test data is valid, it will work... ;)

-- Tim

[ Parent ]

Re: Okay, I admit it... [slightly OT] (none / 0) (#111)
by Rasputin on Fri Jul 07, 2000 at 12:22:03 PM EST

A slightly more elegant (I think) solution:

char numbers[] = {num1, num2, num3, num4, num5, num6, '\0'};

for (int i = 0; i < 6; i++); {

if(strchr(numbers+i+1, numbers[i]))

return (numbers[i]); }

This just off the top in about 30 seconds ;) Do they still teach pointer arithmetic?
Even if you win the rat race, you're still a rat.
[ Parent ]

Re: Okay, I admit it... [slightly OT] (none / 0) (#104)
by Anonymous Hero on Fri Jul 07, 2000 at 08:26:09 AM EST

I believe it was Penn and Teller who said,

Anyone who loves brain teasers should have a gun put to their head and the trigger pulled slow, "Suck death, puzzle loving pig." -- paraphrased from Penn&Tellers Cruel Tricks for Dear Friends

Incidentally, Cruel Tricks itself would seem to be just as valid a way of interviewing people. Just see if they fall for the glasses bit or one of the other tricks in the book.

Of course, I think you should have an "interview box." You could use this computer to conduct interviews by deliberately messing it up, and then seeing how the interviewee got it up and running again.

Of course, as an old Infocom junkie, who got the Babel Fish without help, I should be happy that interviewers are using logic puzzles to determine my programming skills. Of course, I still don't think it is a fair way to test someone.

[ Parent ]

Re: Okay, I admit it... [slightly OT] (none / 0) (#125)
by Anonymous Hero on Sun Jul 09, 2000 at 12:14:19 PM EST

I think a more interesting form of the problem is:

Supposing you have a 10000 (n) numbers in the range 1 to 9999 (n-1), and exactly one number is repeated. Find the repeated number in a single loop without using an array i.e. using O(1) memory space.

[ Parent ]
This is a good problem. SPOILER. (none / 0) (#138)
by Paul Crowley on Sat Oct 14, 2000 at 07:33:04 AM EST

Sum them.

I had to think really hard about that one! Thanks!
--
Paul Crowley aka ciphergoth. Crypto and sex politics. Diary.
[ Parent ]
Interviews are fun (4.33 / 12) (#39)
by Anonymous Hero on Thu Jul 06, 2000 at 11:20:34 AM EST

Okay, I'm going to wander off-topic for a bit, but I'm in a story-telling mood today. This is a true story about how I landed my first big BoFH job, in case anyone is wondering.

About two and a half years ago, I was interviewing for a sysadmin position at a small ISP. The kind of place with about a dozen employees where everybody did a little bit of everything.

The day of my interview I showed up at the office/NOC/warehouse and sat down with the sales manager/HR guy. "The president", he said, "will be with us shortly to handle the technical side of the interview. Until then, let me tell you about the company and go over your resume."

Twenty minutes of small talk later, the company president, an old school VMS admin, entered the room to ask a few technical questions.

"How much do you know about UNIX?"

"A fair bit. I've worked with Solaris, Linux and --"

"Great. Here's a hypothetical situation. Suppose you have a unix server and had somehow corrupted the /etc/passwd file. Nobody can log in, not even root. You need it back right away. How do you fix it?"

"Interesting question. It depends on what kind of unix it is, but basicly I would boot it from a CD, mount all the filesystems, then go in and repair the damaged files, perhaps by restoring them from a backup."

"Good answer." He nodded, gave a meaningful look to the HR guy, and stepped back out of the room.

I chatted with the HR guy for a few minutes about the details of the job, and then he took me on a tour of the building -- Up until then all I had seen was the lobby and one office.

There, in the back room, were the company president and two of their developers huddled around their main unix server, which they had booted from a CD, desperately trying to repair the damaged /etc/passwd file.

Not surprisingly, they hired me immediately.

-D
dcross@cryogen.com

Re: Interviews are fun (none / 0) (#62)
by CodeWright on Thu Jul 06, 2000 at 02:33:49 PM EST

LOL!

That story is so good, it seems almost apocryphal.



--
A: Because it destroys the flow of conversation.
Q: Why is top posting dumb? --clover_kicker

[ Parent ]
Re: Interviews are fun (none / 0) (#65)
by Anonymous Hero on Thu Jul 06, 2000 at 03:05:44 PM EST

I like you as an editor

[ Parent ]
Show me numbers (none / 0) (#41)
by ejbst25 on Thu Jul 06, 2000 at 11:31:52 AM EST

Geesh....I am an RHCE. And if 60% of all the RHCE's don't know jack...that would be a HUGE surprise to me. Yea..I work with a couple of MCSEs here that don't know squat...hell..if you ask them something NT they look it up in a book. I swear all they were taught was how to use microsoft's support page. But there are some good and valid certs. A cert never meant you are a guru....most places equate it to from 2-5 yrs of experience. How much that really is should be left up to a technical advisor to your HR. But don't try to invalidate them all.

Everyone who has their cert and deserved it that I know is employed and not looking. ;-) How do you know that 60% of the ones you see actually represents 10% total? Unless you spoke with a large sample of certain certs (if 90% are employed your sample would have to be 90% employed and 10% unemployed) I don't think you can make such a judgement. I will let you pass judgements whether a certain person knows their stuff or not. You probably are qualified to judge that. But, the sample of certified people who are looking for a job with your company is most likely not an accurate sample. I know my company treats certified techs who know their stuff very well..and most wouldn't care to look elsewhere for a job. Maybe I am wrong...but it is something to think about.

Re: Show me numbers (none / 0) (#49)
by warpeightbot on Thu Jul 06, 2000 at 12:05:39 PM EST

Dif'rent bits of paper mean dif'rent things.

MSCE, as we all know, means about jack. (dunno about the higher grades, but, c'mon, it's a multiple guess test with no practical.)

RHCE means you know how to install and configure Linux and have had some hands on practice with busted machines and such. I hear they're planning an advanced curricula; those guys are going to be serious ubergeeks, although I hope it's not too limited to The Red Hat Way.

Cisco's various exams are killer. The ubermonster is a two-day written, followed by three days of sweating it out for real in the labs. Those guys are serious, and I bow to their superiour geekhood. You get a dude with a Cisco logo on his card, and you expect him to do networks with Cisco hardware, if Cisco says that's really him, hire his butt, fast. And pay him well, he deserves it.

The lesson here is still caveat emptor, and know what those pieces of paper really mean.

w.e.b., rhce

[ Parent ]

Re: Show me numbers (none / 0) (#55)
by Inoshiro on Thu Jul 06, 2000 at 01:06:16 PM EST

Geesh....I am an RHCE. And if 60% of all the RHCE's don't know jack...that would be a HUGE surprise to me....How do you know that 60% of the ones you see actually represents 10% total?

Well, I can't speak for others, but my only RHCE experience came from some nimrod who send "unsubscribe" to all the subscribers of a mailing list several times (even after we all pointed him directly at the FAQ, website, and unsubscribe address). One person isn't statistically sound, but one bad apple spoils the bunch in my mind ;-)



--
[ イノシロ ]
[ Parent ]
Phone Screen candidates and recruiters (4.00 / 1) (#42)
by mahlen on Thu Jul 06, 2000 at 11:33:25 AM EST

I think a couple tweaks to your interview process might help you here, although I'm not certain what you seek, so this advice may be just worth what you paid for it.

Unless the candidate has such great-looking experience on the resume that you don't want to take the chance of losing them by moving too slowly, have a tech-type interview them on the phone (you should come up with a standard list of questions to make it less ad hoc) for 30 minutes or so before bringing them in and using lots of people's time. While you can't ask, "What's wrong with this code?" type questions (unless you post it to WWW and give them address on the phone), you can certainly get a sense of the depth of their experience, what kind of work they seek, and their overall Nerd Quotient. I think that will seperate a lot of time-wasters out early.

Also, the "make 'em cry" school of interviewing isn't really needed, and is mainly appropos for a Microsoft, where being a jerk is a necessary job survival skill. Gradate your questions up from easy to impossible, and note where they fall off. Don't alienate the people you seek; even people who answer all your questions will think you're too annoying to see every morning.

Also, consider recruiters. Yes, expensive, but a good one can save you great stinking piles of hassle. I recently landed a Senior Java Developer job via Monday Technology Solutions (www.mondaysolutions.com), and they did great work for all concerned. So shop around for good ones.

I know nothing about certs in particular, since all I have is a degree from UC Berkeley. But yes, if you're getting resumes from the wild blue yonder and then interviewing the people who claim them, the world does appear to packed to the seams with pedigreed morons. Filter, filter, filter, and good luck.

mahlen

"Here's a pound for your troubles."
"I ain't got no troubles, mate."
"You have now, mate. That pound's a forgery."
--The Goons: "Six Charlies in Search of an Author"


Re: Phone Screen candidates and recruiters (none / 0) (#52)
by blixco on Thu Jul 06, 2000 at 12:24:52 PM EST

We do quite a bit of phone screening. Actually, we use recruiters too.

Funny story: my manager was doing a phone screen and the person on the other end was busy looking up "best answers" for the questions on the internet. She could hear the typing, and the guy was lagging on his answers to the point where she said "If now isn't a good time, I can call you later."

Also, we rarely "make 'em cry." We do ask tough questions related directly to the job function they would perform.
-------------------------------------------
The root of the problem has been isolated.
[ Parent ]
Re: Phone Screen candidates and recruiters (none / 0) (#63)
by porovaara on Thu Jul 06, 2000 at 02:35:18 PM EST

Second the notion on recruiters... a good one can save you soooo much time. I've also had one good placement through Monday Solutions... then again that was out of about 40 resumes sent. You have to be picky.


[ Parent ]
Interview styles (4.00 / 1) (#47)
by BlueFox on Thu Jul 06, 2000 at 11:59:58 AM EST

One of the skills I picked up at my current employer that I value quite a bit is interviewing. I'm far from an expert (or even a skilled) interviewer, but I have learned how to find the information I'm looking for.

When I started out interviewing I tried to take the "ask technical questions" approach, and made some good and some bad hiring decisions. The HR person who supports the IT department here helped develop an interview process that looks at how someone thinks more than what they know. The process is based on the idea that past behaviour is likely to be repeated in future.

The process starts with clearly defining what you are looking for in a candidate, and picking specific traits you need. An interview is then structured using standad questions that are designed to find out if the candidate has these traits. An example might be "problem solving", in which case we have the candidate describe a difficult problem they had to solve in the past.

The key is to have them walk through the process in detail, and we stop them for more detail when needed. By asking for a description of past action, we avoid the canned answers on how they "would" solve a problem. At the start of the interview we outline what we're looking for is the candidates personal action. If they were part of a team, we still want their personal actions, and not a desccription of the teams work. If the candidate drifts off into "we did this" the interviewer asks for clarification of the candidate's specific role, and if the description becomes "I would do this" they are asked to describe what they did.

It is actually quite difficult to BS your way through this style of questioning, since people will naturally switch to using "would" and "we" if they are not describing an actual event or personal action.

The whole interview is kept to 10-15 scripted questions but can take some time to complete. "Points" are given for specific things being said that identify the traits being looked for.

The other advantage to this process is that by clearly defining what you are looking for in advance, personal traits become much less of a factor. How well we like a candidate based on initial impressions is not recorded in the scoring. Since we're talking about actual past events, people are usually more at ease than they might be if being grilled on technical issues.

This process also works quite well for hiring entry level candidates who don't have a high level of knowledge to experience in the field. A description of a difficult problem they had might be troubleshooting a problem with their car. The process is the same, since we can still look to see how they thought the problem through.

It is important to keep in mind when interviewing that the interview is not at all like the normal work environment. Most problems are not solved while sitting in a room with four people staring you down. Instead they are solved by looking at the problem from a variety of angles with the tools in your hands. Similarly brain teasers are fun, but I'd put little weight in them. If the candidate doesn't think through puzzles in the same way as you they can be more harmful then good by setting the interview off on the wrong foot. More often than not, brain teasers are really a test of whether the candidate has seen it before or not. (Personally I get stumped by sequences like yours every time - especially if only given a minute or so to think about it, but fare very well with visual puzzles).



Very useful, but that was kinda off topic (none / 0) (#66)
by Anonymous Hero on Thu Jul 06, 2000 at 03:10:00 PM EST

His problem is that HR is sending the people without the necessary knowledge and skills to him. It's a case where he's being sent people with certifications that don't know what they're talking about and training that kind of person is a difficult and timeconsuming (at best) task. Usually that type of candidate doesn't have the right aptitude anyway.

[ Parent ]
Re: Very useful, but that was kinda off topic (none / 0) (#73)
by BlueFox on Thu Jul 06, 2000 at 03:35:20 PM EST

Point taken.

If the people you are looking for are not clearly defined in a way that they can scan for, HR have no choice but to use certifications as a basis of selection. Working with them up front taps into their skills at selecting people. HR people are the same as technical people in that they work best with clearly defined needs outlined.

Technical people are famous for saying it is the users that are the problem. I'll bet HR staff say the same thing.

[ Parent ]

Does the sun always rise in the east? (none / 0) (#54)
by Anonymous Hero on Thu Jul 06, 2000 at 12:48:32 PM EST

Everyone (well, many people) have already seen those problems. God help you if you hire someone who thinks Marilyn von Savant really is the world's smartest woman. :*)

Anyway, here's one that I like because it the person's reasoning gives you some granularity -- and the people who immediately jump at the boundary conditions are the people you can trust to know that constants aren't, disks fill up (and hence writes can fail, etc.)

DOES THE SUN ALWAYS RISE IN THE EAST?

Re: Does the sun always rise in the east? (none / 0) (#60)
by squigly on Thu Jul 06, 2000 at 01:32:02 PM EST

DOES THE SUN ALWAYS RISE IN THE EAST?

No. Only on Earth.

So who is this Marilyn von Savant who has been mentioned in at least 2 posts here? Or is that like asking "Who's this Alan Turing chap I keep hearing mentioned?"

--
People who sig other people have nothing intelligent to say for themselves - anonimouse
[ Parent ]
Right answer, wrong reason (none / 0) (#64)
by Anonymous Hero on Thu Jul 06, 2000 at 02:43:26 PM EST

Nope, that's the shoot-from-the-hip answer that causes me to eliminate you from consideration from any senior position. It tells me that you don't question your assumptions well enough for the senior-most roles.

This question refers to earth, today. (Well, this year). The answer, once you figure it out, is extremely obvious in a "frame grabbing" sense.

MvS has a column in the US national weekly insert "Parade" magazine, and she frequently answers reader's puzzles. Her claim to fame is taking so many IQ tests in such a short time that her measured IQ is/was the highest on record -- which is pretty convincing proof that IQ tests are meaningless when you're in the top few percentile.

[ Parent ]
Re: Right answer, wrong reason (+ IQ tests) (none / 0) (#78)
by squigly on Thu Jul 06, 2000 at 04:07:03 PM EST

On the subject of IQ tests - They can only be guarenteed to measure accurately ones ability to answer IQ tests. This tends to have a correlation to logical reasoning, but are not a direct measure, especially considering that meanings can be subjective (Like in this case, I was assuming that you weren't necessarily talking about earth). I could of course have used the ambiguity in the question, and answered "No. I have been to the east, and saw the sun set."

Still not sure about the sun rising in the east question. I presume that in the Arctic circle, it will rise in the south at the appropriate time of year, but that doesn't sound right since you talk about the "frame grabbing" sense.

--
People who sig other people have nothing intelligent to say for themselves - anonimouse
[ Parent ]
Re: Right answer, wrong reason (none / 0) (#79)
by Anonymous Hero on Thu Jul 06, 2000 at 04:20:03 PM EST

Frame-grabbing my ass. When you're throwing out tricky questions, expect to have deeper answers. If someone asked you this on the street, the answer would be a definite yes. When asked in the context (or bullshit term: "frame-grabbed instance") of an interview, you'd better expect that whoever is answering is attempting to give you the most absolutely correct answer taking into account all possible ambiguities in the question posed. So shut your superior cake-hole.

[ Parent ]
Re: Right answer, wrong reason (none / 0) (#95)
by Anonymous Hero on Thu Jul 06, 2000 at 11:14:33 PM EST

I generally only consider the answer when evaluating a candidate for the senior-most positions, but I also use it as a stress test. Clients rarely articulate their requirements in a way that I fully understand the first -- or tenth -- time I hear it, and no company can afford a technical staff that gets easily frustrated and starts communicating a condescending attitude towards them just because they use different nomenclature. I doubt you would actually call an interviewer or client a "superior cakehole," but I'm sure you communicate it nonverbally and it's my job, as an interviewer, to see if I can elicit that response under the type of stress that you'll see on the job. If I can, it doesn't matter how good your technical skills are, you're tempermentally ill-suited for a job working with clients or even other employees.

For the record, the "frame grabbing" was short for the sense of a violent shift ("grab") in your frames of reference. One moment you think in terms of North-East-South-West, the next you're at the North Pole and realize that all directions are due south.

[ Parent ]
Re: Right answer, wrong reason (none / 0) (#119)
by Anonymous Hero on Sat Jul 08, 2000 at 12:54:14 AM EST

>Nope, that's the shoot-from-the-hip answer that causes me to eliminate you from consideration from any senior position.

I'm glad I don't work for your company. Any company where the HR person (or anyone with power) knows all the answers (and any other ones are automatically wrong) is going to be full of pissed off workers who can't get their jobs done due to red tape.

[ Parent ]
Re: Does the sun always rise in the east? (none / 0) (#68)
by Anonymous Hero on Thu Jul 06, 2000 at 03:17:21 PM EST

How nitpicky do you want me to be? My answer, if posed this question in an actual interview, would be "This morning it appeared to, but past behaviour is only an indicator of future performance, not a reliable predictor. Is there a specific problem that needs solving that this question pertains to? Stating a problem would establish the parameters more precisely."

But I'd only say that to cover my ass with the interviewer. If my 4-year old son asked me, I'd say "Here on Earth the Sun appears to rise in the east because of the direction of the Earth's orbit." (pausing to draw diagram on anything available) "On Pluto, which revolves around the Sun in the opposite direction, the sun would appear to rise in the west". When my nine-year old nephew asked me this question a couple of months ago I gave the same answer but used differing language - terms like "retrograde" and "ecliptic" occurred.

The point is, imparting a useful amount of knowledge by answering this question would generally require more understanding of the questioner's points of reference than it is possible to have in the context of an interview.

So what would you consider a "correct" answer?

--Charlie

[ Parent ]
Re: Does the sun always rise in the east? (none / 0) (#75)
by Qtmstr on Thu Jul 06, 2000 at 03:42:43 PM EST

I hate to be pedantic, but isn't Venus the planet that rotatates in the direction opposite the others? (Not revolves, of course).


Kuro5hin delenda est!
[ Parent ]
Re: Does the sun always rise in the east? (ANSWER) (none / 0) (#80)
by Anonymous Hero on Thu Jul 06, 2000 at 04:22:36 PM EST

Let me rephrase the question?

Does the sun rise in the east AT THE NORTH POLE?

This isn't a trick question on astronomy, it's a challenge to the assumption that "EAST" is always defined. Our coordinates have two boundary conditions - the northernmost point and the southernmost point - where all of our conventional directions are meaningless.

Believe it or not, I've known a number of people who hear the question and immediately burst into laughter. The common thread is that they all habitually challenge their assumptions, so when I ask this question they ask questions like "where is 'east' a meaningless term?"

[ Parent ]
Re: Does the sun always rise in the east? (none / 0) (#83)
by kovacsp on Thu Jul 06, 2000 at 06:25:00 PM EST

No. If you're at the south pole, it always rises in the north. And if you're at the north pole, it always rises in the south. There aren't any other directions at the two poles.

[ Parent ]
Re: Does the sun always rise in the east? (none / 0) (#84)
by kovacsp on Thu Jul 06, 2000 at 06:27:34 PM EST

Whoops. I swear you're answer wasn't there when I replied. Oh well. I was right anyway.

[ Parent ]
Re: Does the sun always rise in the east? (evaluat (none / 0) (#94)
by Anonymous Hero on Thu Jul 06, 2000 at 10:40:15 PM EST

No problem - I waited as long as I could before the nit-pickers threatened to take over.

My hierarchy, from ideal to get-this-person-out-of-here-now!, is:

0) The poles. Hey, they're also the only places on earth where the "day" is exactly one year long!

1) "East" is meaningless at the poles, but the sun rises there. These are the people you want designing and coding your most critical systems!

2) Whipping out my astronomical almanac.... They miss the bit that breaks the rules, but they can probably be trusted to implement things correctly.

3) "Isn't that the definition of 'east'?" I believe that's wrong (isn't it the "right hand rule" applied to the planet's rotation, not the (greatly abstracted?) direction of sunrise, but it shows the ability and willingness to follow formal definitions instead of gut feelings. This is someone who is willing to use man pages, read RFCs, etc.

4) "Due east?" I said nothing about due east, so they're reading more into the requirements document than is present. That's dangerous - mid-level programmers requiring regular status meetings.

5) Non-earthers. These people are reading too little into the requirements document. As others have pointed out, common sense is also an important criteria. (Criterium for the grammar nazis :-) If they mention Venus explicitly, call them wise-ass middle-to-senior level programmers. If they don't, Middle-to-entry level programmers requiring even closer supervision. If they mention Hyperion and expect you to catch the reference, go to the first entry.

6) Dunno/That's a stupid question/I hate riddles. I hate them too, but I really don't care how long it takes you to get the answer (wthin reason) as long as you're asking the right questions.

As I said, this question allows *far* better granularity than the OTTFFSS question.

The "manhole" question also allows some granularity, and for the record my answers to it are: 1) not all manholes are round - some of the earliest ones were square. In the unlikely event you need to track manhole covers for the oldest parts of London (IIRC)... 2) with the small lip it's impossible to drop them down the hole (N.B., not just because they're round, but without the lip the cover would always drop straight to the bottom!), 3) when pipes are used to encase the access tunnel those pipes are usually circular and round covers fit the hole exactly - no need to manufacture fairings, and 4) as someone else pointed out they're self-seating when the sewers "burp" or the cover is otherwise slightly offset.

[ Parent ]
Re: Does the sun always rise in the east? (none / 0) (#107)
by Rasputin on Fri Jul 07, 2000 at 11:41:55 AM EST

I noticed a lot of people assuming that the only possible direction at the [North|South] Pole was [South|North] but that particular assumption is wrong. If you happen to be at the magnetic North Pole (Just off Little Polaris Island in Canada's Arctic) then east/west are generally accepted to be defined in relation to true north. This is because you have to use a gyroscope (or similar non-magnetic technology) due to the compass becoming useless there. At the True North Pole (in the middle of the Arctic Ocean) you actually have the option of using Magnetic North because you are far enough away from the MNP. Interestingly, if you allow for the magnetic variation at Thule, Greenland, Magnetic North is actually where most people would assume west to be. From there, magnetic south is actually true east (well, it's actually off by a couple of degrees) .

This question is too loose for the type of precise answer that was probably expected.
Even if you win the rat race, you're still a rat.
[ Parent ]

Certifications -- Nothing more than a sport/hobby (none / 0) (#67)
by ejf on Thu Jul 06, 2000 at 03:13:00 PM EST

I have not yet taken any of the "well known" paper certs like the Cisco or M$ ones because I simply lack the money to do so. I did take some tests (and got some certs) at BrainBench.Com, and to be frank, I had the exact same feeling taking them I had when I was in those High School Computer Science competitions between schools. They are meaningless, but a fun thing to do if you have some time at hand.

Although some of the BB questions come pretty close to the RW and actually test some basic understanding of the topic, they are prone to be cramsessioned by some. All these tests do is test whether you have some understanding of the topic at hand, but they don´t qualify you for the Real World.

I will probably take the Cisco and/or M$, etc. certs some time (once I have a significant amount of money to spend), but from what I heard they are pretty meaningless as well. Yes, the lab tests might be more challenging than those pencil-and-paper tests, but no test is going to measure your experience and/or problem-solving skills accurately. The only people who SHOULD be impressed by them are those PHB´s.

I see those tests as a sport/hobby to pursue. I would not rely on them myself (well, not entirely -- it is good to know the person you´re hiring knows how a subnet mask looks like. It would be better to know whether that person knows what a subnet mask is.

Oh, yeah, my transcript is here. Just in case a PHB reads this and wants to finance my buying of Diablo II (I really am desparate, am I not ?)

my |-$.02|


--- men are reasoning, not reasonable animals.
Useless trivia. (none / 0) (#69)
by Anonymous Hero on Thu Jul 06, 2000 at 03:24:03 PM EST

OTTFFSSEN - First letters of the numbers one thru nine.

A round manhole cover has a uniform diameter, so it can't fall through the hole and injure someone below. It's also easier to move the cover about once it has been lifted off since you can roll it.

The sun doesn't always rise *due* east. For example, it rises in the south-east on the winter solstice in the northern hemisphere.

As someone else mentioned, being able to answer trivia questions like these is more dependent on having seen them before than being a good "problem solver" or "critical thinker". If you are basing hiring decisions on useless trivia, you will end up with a bunch of trivia buffs, not necessarily good employees.



Re: Useless trivia. (none / 0) (#76)
by Gomker on Thu Jul 06, 2000 at 03:46:01 PM EST

..."so we stick to the roughest technical questions we can think up. I also like to throw in a curve...questions like "why are manhole covers round?" or "complete this sequence: OTTFFSSE...." "

I think they were asking the non tech questions to make sure the person had some common sense. Some people can be "Certified" as a tech but are more like certified idiots.

[ Parent ]
stupid brainteasers (none / 0) (#85)
by cpt kangarooski on Thu Jul 06, 2000 at 06:27:55 PM EST

I enjoy puzzles, but I would really be cheesed off if getting a non-puzzle-solving job depended on them. It's about as stupid as making sure that FBI agents can routinely win at Super Mario Bros. just in case they need to rescue a kidnapped princess.

Employers should look at the quality and quantity of work, skills, references, etc. If that's still not enough, offer a trial period of a couple of paid weeks on the job. Or contract and then hire later.

Now then, here's a really fiendish one; if you're a wimp there are answers for *many* such puzzles on some web site somewhere that was mentioned on /. once. Here's the sequence: 1, 11, 21, 1112, 3112, 211213, 312213, 212223, 114213, 31121314. Now then, what is the next item in the sequence? Why? This pissed me off for ages before I figured it out.

--
All my posts including this one are in the public domain. I am a lawyer. I am not your lawyer, and this is not legal advice.
Re: stupid brainteasers (none / 0) (#90)
by torpor on Thu Jul 06, 2000 at 07:48:45 PM EST

Simple:

41122314

Right?

j. -- boink! i have no sig!
[ Parent ]
Re: stupid brainteasers (none / 0) (#123)
by FlinkDelDinky on Sat Jul 08, 2000 at 04:56:48 AM EST

Yes , that's it. See my reply to my reply of the original post. Did you figure it out for yourself? Had you encountered similar problems that 'trained' to think about sequences like this?

I don't know any math but every time I encounter 'number theory' type problems like this and the fibonachi sequence I'm always facinated.

I just don't think I would've seen the answer to this one. I'm pretty impressed that you saw it.

[ Parent ]

Re: stupid brainteasers (none / 0) (#132)
by torpor on Mon Jul 10, 2000 at 07:54:25 PM EST

Took me 2 minutes, and I figured it out without any help. I haven't encountered this particular one before, but I have been honing my skill of being able to 'observe the obvious' as a systems programmer for many years (needed to survive those late-night debug sessions, particularly my filesystem research).

I think that problems like this definitely cater to a certain 'debugger' ethos which is refined in systems programming over the years.

I'm really lousy at math, though. Just because the problem involved numbers doesn't mean it's a math exercise ... For me, it was instantly a pattern-matching problem, because I saw a general pattern in the first 3 steps of the sequence.

Could just as easily been:

A AA BA ABAA CAAB BAABAC CABBAC

j. -- boink! i have no sig!
[ Parent ]
Re: stupid brainteasers (none / 0) (#133)
by FlinkDelDinky on Mon Jul 10, 2000 at 09:44:54 PM EST

I spent hours on it before I googled it. There's just now way I would've seen that. I suppose as a systems guy you learn to look for 'bad' patterns (array overruns, unfreed mallocs, etc.) within the whole.

I was trying to figure out how the 2 went from the front at 21 to the back at 1112. I was investigating all kinds of complex crazy ordering schemes.

The answe was far simpler than my perception of the problem. It's cool that you got it by yourself though.

[ Parent ]

Re: stupid brainteasers (none / 0) (#117)
by FlinkDelDinky on Fri Jul 07, 2000 at 07:01:39 PM EST

I give up, please, mercy, I beg of you.

<p>Maybe not an answer but a hint. If it involves maths beyond add and subtract, I'm out.

<p>I thought it may have been some odd binary notation, kinda like hexadecimel can represent 4-bits. But now I don't think so. I'm wondering if it's another one those OTTFFSSEN things (which I never would've figured out).

<p>I my GOD, omigod awmigod ogmygod etc..

<p>I was just about to say that the 1, 11 had a fibonachi (sp) ring to it, then something lept out like a bolt of lightning:<br>
1 = 1, 11 = 2, 21 = 3, 1112 = 5 = DAMN IT TO HELL!!!! my idea was wrong!

<p>Help is needed...

[ Parent ]
Re: stupid brainteasers (none / 0) (#122)
by FlinkDelDinky on Sat Jul 08, 2000 at 04:50:11 AM EST

Okay, I went to google and searched:
"1, 11, 21"

Found a bunch of sights with the answer. Very simple. Don't think of difficult transformations or sequences like fibonachi (sp?). I never would have guessed it. You could figure it out with just the first two numbers but it gets obvious as it goes (just not obvious enough for me).

[ Parent ]

Re: stupid brainteasers (none / 0) (#129)
by Anonymous Hero on Mon Jul 10, 2000 at 12:04:36 PM EST

How about this for a sequence:
1, 11, 21, 1211, 111221, 312211, 13112221, 1113213211...


[ Parent ]
Re: stupid brainteasers (none / 0) (#134)
by cpt kangarooski on Mon Jul 10, 2000 at 10:27:07 PM EST

This is pretty similar, but there is a subtle bit of fiendishness in it.

I don't want to entirely give it away, but rather than deal with totals (which is what I was doing) it describes digits...

The next item in the sequence is 31131211131221 (and 3. ;)

--
All my posts including this one are in the public domain. I am a lawyer. I am not your lawyer, and this is not legal advice.
[ Parent ]
Brainteasers vs. "real" problem solving (none / 0) (#87)
by Anonymous Hero on Thu Jul 06, 2000 at 07:01:12 PM EST

I have to agree that making a hire/no hire decision based on someone's ability to solve a puzzle is stupid. But it can give you some good insight into how they think. (The "how many names are in the NY phone book?" question from PC Roadkill is nice -- do they guestimate or squirm?)

I always ask the manhole cover question in my interviews. No one yet has been able to answer it, but some people have better answers than others.

A good answer is any one which shows that you thought about it some and had a few ideas, even if they were flawed. A bad answer is saying "I dunno" and not coming up with anything even after a lot of prompting and being told not to worry about getting the one correct answer.

Decent answers: so far we've gotten the standard "won't fall in" and "rolls easily" answers, plus "so if the sewer burps it will roll like a quarter but still settle back on correctly", which was a new one.

Smartass answer: because the hole is round too.


Still, I have to agree that a nice "how would you solve this real-world problem" question, with reasoning and additional detail, is best.

And, finally: we once asked a guy to tell us about difficult technical problem he'd had to solve and how he did it... and he told us about how he was the first person in [his home country] to solve the Rubick's cube, back in '83, and write a book about it. He was very earnest, we were scared, HR went ahead anyways. Great.

-T

Re: Brainteasers vs. "real" problem solv (none / 0) (#93)
by blixco on Thu Jul 06, 2000 at 08:40:29 PM EST

We interviewed the *same guy* here.

Now *that* gives me the creeps.
-------------------------------------------
The root of the problem has been isolated.
[ Parent ]
Re: Brainteasers vs. "real" problem solv (none / 0) (#98)
by Anonymous Hero on Fri Jul 07, 2000 at 02:24:21 AM EST

Why does anyone ask the manhole cover question anymore? I've read about it a million times. Does anyone *not* know about it? I hope some idiot asks me it during an interview so I can pretend to think really hard and come up with a "good" answer.

Anyway, it was clearly demonstrated to me why manhole covers are round when I was 13. I was on a fishing trip in Canada. We had a cast iron wood stove in our cabin. One of my uncles took one of the (round) burners off and tried to push it into the oven. "See," he said, "It won't fall in. It can't. That's why manhole covers are round."

Luke Francl

[ Parent ]
"I know more than you" posts are annoyin (none / 0) (#102)
by Anonymous Hero on Fri Jul 07, 2000 at 07:59:11 AM EST

So, your uncle should be the benchmark by which a company judges all candidates? "Forget certifications, if they knew Luke's uncle, they're in!"

The point is not the answer, it's whether or not the person at the opposite end of the question has the ability to think of an answer. You wouldn't believe the number of times I hear "I don't know. I'll have to get back to you" or "that's a good one."

Actually you might believe the number of times I have heard that. Your uncle probably told you.

[ Parent ]
a small problem (5.00 / 1) (#88)
by Anonymous Hero on Thu Jul 06, 2000 at 07:08:40 PM EST

I am an adult with a peculiar language disorder that makes it damned near impossible for me to pass written tests. Its not that I don't know the answers, its just that when I read the test questions they often appear to make no sense. However if you ask me the same question out of the clear blue in the hallway after lunch, I can recite the correct answer without any difficulty. Very hard to describe in a way that someone without a speech and language disorder can understand.

On the plus side though, my little problem has been of enormous benefit in information technology because of the strange way my cognitive processes work. I've been a systems programmer, a sysadmin, and a systems analyst. I can debug and fix anything, hardware or software, and fast. I just can't pass those fscking tests.

Re: a small problem (none / 0) (#97)
by e7 on Fri Jul 07, 2000 at 02:11:43 AM EST

I wonder if local hiring laws might come into play here (thinking of the Americans with Disabilities Act in the U.S.). You might have the equivalent of a doctor's note stating that you can't take the tests in that format, and they'd have to provide an alternative for you. Or something.

I'll check into this with our HR people at work ... may be a false hope, but it would be worth knowing what your rights are. What country are you writing from?

* e7


[ Parent ]
Re: a small problem (none / 0) (#116)
by e7 on Fri Jul 07, 2000 at 05:50:14 PM EST

My sources tell me that interviewers in the U.S. must take into account obvious disabilities -- a deaf person might process written language differently than a hearing person, for example, and accomodations must be made for that.

An 'invisible disability' is less clear-cut. For one thing, you'd have to bring it to their attention, which obviously could bias the interviewer.

Also, A.D.A. has lots of grey areas. Employers usually make their best guess about the appropriate action, and if a lawsuit is filed then they get a chance to justify the decision before the court.


[ Parent ]
Re: a small problem (none / 0) (#130)
by Anonymous Hero on Mon Jul 10, 2000 at 01:29:38 PM EST

The ADA is useless for invisible disabilities like this. Besides, it doesn't matter to me personally anymore, I've managed to acquire an impressive work record over the years. Since I can't bring test scores (certs) to an interview, I've put together a portfolio of prior work. I've been interviewed for magazine articles, I've been quoted on web pages, and that usually wows the HR folks enough to get to the next interview level. Besides, I've been compensating for my disability for four decades, so it rarely gets in my way anymore. BTW, I have Asperger's syndrome, in other words high-functioning autism.

[ Parent ]
Hiring practices. (none / 0) (#91)
by static on Thu Jul 06, 2000 at 08:25:29 PM EST

I can identify with being the interviewee because I went through a number of interviews just recently. It was very quickly apparant that I was more experienced at interviews than they were!

But I also noticed that the interviews were mostly of the "how well will you fit in here" type. Although the technical knowledge was also important, it had largely been checked already by the recruiter. Perhaps this reflects the IT Recruitment industry in Australia...

For my current job, I first applied by filling in small quiz on their web site. The questions were ... intruiging. And not written by someone who necessarily knew the answers! Without initially saying so, they were looking for someone who appeared to know what they were talking about.

Finally, the quality of an applicant's resume needs to be stressed. Those looking for a job must always be prepared to accept constructive criticism of it in order to improve it. And they must be prepared to ask recruiters how it could be improved. Many people cannot write decent resumes and it is astonishing how much of an advantage this gives to those who can.

Wade.



Questions I ask (none / 0) (#92)
by Buck Satan on Thu Jul 06, 2000 at 08:39:03 PM EST

One of the questions I ask is "what computer related publications do you read?"

If they say "PC World" or "Family Computing" or something that my mom would read, I immediately write them off.

I myself hate taking those fscking tests. The last time I took one of those they had me take it in the office where they were doing telephone support. I got up and walked out after five minutes because of all the damn phones.



Technique on college fresh-outs. (none / 0) (#113)
by Anonymous Hero on Fri Jul 07, 2000 at 01:42:18 PM EST

This used to work pretty well a few years ago.

At the time, most college students used their school e-mail account for everything, including Usenet posting. So, I would hit up Deja (or the like) and search for questions an interviewee had posted to comp.* (or similar) news groups. Usually I would find one or two questions relating to a class project, sometimes a very important one -- like a senior project or thesis. During the course of the interview, I would ask the interviewee to describe the project, problems, solutions, etc. and try to sneak in something related to the question they posted in the newsgroup.

Depending on the content of the post, I could usually tell if they were just asking for an answer that could be copied w/o having to learn anything new, or if they were trying to learn something new and interesting in the process. In the latter, asking about it during the interview could help assess how well they learned and their ability to retain the salient points.



Joel Spolsky's 'guerrila guide' (none / 0) (#118)
by Anonymous Hero on Fri Jul 07, 2000 at 08:47:11 PM EST

Joel Spolsky at joel.editthispage.com has an article on how to do interviews.

Puzzles bad
Problem-solving good
Passion good
etc....
It's quite good and I suggest you read it. Look in the "Site Index." I like all of his writings.

-SH

Look for quick learners! (none / 0) (#121)
by DigDug on Sat Jul 08, 2000 at 03:36:38 AM EST

Technology stuff is changing like nothing else before it. The most important quality I can think of in a job candidate (besides the obvious basics) is that (s)he is a quick learner.

A terribly rough and unfair example: I'd rather hire a person who knows 50% of Perl but is a quick learner than someone who has 100% of today's Perl hard-coded into their brain, without being able to learn anything else. Why? Because if the first person finds a new library that (s)he needs to learn how to use, they'll be able to just read the docs and learn how to use it within 15 minutes. This may be a problem with the second candidate.

The second most imporant quality (well, it goes together with the first) is an above-average IQ. Very often, this can be (sometimes falsely) deduced from just talking to a person, but if a test is required, a test it is.

--
Yavista - if you haven't found a nice homepage yet.

Certs are more of a pain then help (none / 0) (#126)
by smcavoy on Sun Jul 09, 2000 at 11:14:30 PM EST

I really can't stand MCSE's. I worked at a computer chain as a technician for just over a year, and met quite a few MCSE's. Most had no concept of networking and/or computers. They simply read their book, remembered it, then wrote the test. I have since quit the tech job, and started doing consulting for small businesses. I have not got any certs except my A+ which was a requirement of the old job (which I wrote in 15min for the first test, and 9min the second). I do have a great deal of expereince with networking, NT, Linux, and hardware diag. I am finding it hard to get work without a cert, and have acutally thought of getting a MCSE cert to get a regualr paying job (I've thought of RHCE but that would still make it hard). I don't have any college diploma's either, but I have 0 problems learning anything new, which is next to impossible to explain to anyone. So certs do nothing really except help morons (of course there are exceptions) get the jobs, while thouse who really know what they are doing get passed by.

The price of "paper" certifications | 137 comments (135 topical, 2 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest © 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!