World's Youngest Microsoft Certificated Professional Is Five Years Old

HughPickens.com writes Gurvinder Gill writes at BBC that Ayan Qureshi is the world's youngest Microsoft Certified Professional after passing the tech giant's exam when he was just five years old. Qureshi's father introduced his son to computers when he was three years old. He let him play with his old computers, so he could understand hard drives and motherboards. "I found whatever I was telling him, the next day he'd remember everything I said, so I started to feed him more information," Qureshi explained. "Too much computing at this age can cause a negative effect, but in Ayan's case he has cached this opportunity." Ayan has his own computer lab at his home in Coventry, containing a computer network which he built and spends around two hours a day learning about the operating system, how to install programs, and has his own web site.

Microsoft Certified Professional (MCP) is a certification that validates IT professional and developer technical expertise through rigorous, industry-proven, and industry-recognized exams. MCP exams cover a wide range of Microsoft products, technologies, and solutions. When the boy arrived to take the Microsoft exam, the invigilators were concerned that he was too young to be a candidate. His father reassured them that Ayan would be all right on his own. "There were multiple choice questions, drag and drop questions, hotspot questions and scenario-based questions," Ayan's father told the BBC Asian Network. "The hardest challenge was explaining the language of the test to a five-year-old. But he seemed to pick it up and has a very good memory."

Comments

But?

[Anonymous Coward]

But can he write a device driver in assembly language?

Re:But?

[thaiceman]

Unfortunately I have, used to work for a company that custom made their own CNC controllers and software to work with their custom built machines. Wire wrapped cards & technology that was almost as old as I am, the hardest part was finding enough 486/P1 machines & industrial monochrome monitors to work with it.

Re:But?

[Anonymous Coward]

Yes.

Goddamn it!

[Anonymous Coward]

It took me 3 attempts to pass that exam and now there are 5 year olds who can pass it?

Re:Goddamn it!

[hcs_$reboot]

It took me 3 attempts to pass that exam and now there are 5 year olds who can pass it?

Indeed. You need a mental age of 5 years to pass.

Re:Goddamn it!

[pegdhcp]

This is unfortunately true for most certification programs. In Cisco SupportPro training (1996 I guess) I remember some of my classmates' points was better in the exam performed before training than the certification exam. That was because, in the first exam they used a simple criteria that "which answer is most costly and Cisco friendly", but in the actual exam they tried to use the training material they were given...

Re:Goddamn it!

[LordLimecat]

This is unfortunately true for most certification programs.

Thats really not true. Actual Cisco certs like the CCNA / CCNP / etc are about Cisco equipment, but they test pretty heavily your understanding of basic networking protocols.

And keep in mind that a lot of these certs ARE centered around a particular vendor; a RHCP or MCSE are going to focus on Red Hat and Microsoft-- but that doesnt make them pushovers.

Id like to see you go into a VCP cert and BS your way through by picking the answer that favors VMWare, for example.

Re:Goddamn it!

[Marginal Coward]

It took me 3 attempts to pass that exam and now there are 5 year olds who can pass it?

This reminds me of a line from the song New Math by Tom Lehrer:

    Hooray for New Math,
    New-hoo-hoo Math,
    It won't do you a bit of good to review math.
    It's so simple,
    So very simple,
    That only a child can do it!

For those of you who are too young to have experienced New Math, Professor Lehrer introduces it by saying, "Some of you who have small children may have perhaps been put in the embarrassing position of being unable to do your child's arithmetic homework because of the current revolution in mathematics teaching known as the New Math....In the new approach, as you know, the important thing is to understand what you're doing, rather than to get the right answer."

Re:Goddamn it!

[Anonymous Coward]

And that's why I love and respect the *nix

Re:Goddamn it!

[MachineShedFred]

MCP, however, is a pushover. You can get that by simply passing one of the MCSE tests, usually the one centered around the workstation OS.

If you can install it and do very rudimentary administration, you can get an MCP.

Still impressive at the age of 5 though.

Re:Goddamn it!

[Anonymous Coward]

Aye, MCSE is a much larger bitch to get. I'm working on mine right now (BI track), and it requires 5 tests. Passing any one of them gets you the MCP title.

But I agree, passing any of them at 5 is impressive.

Re:Goddamn it!

[Penguinisto]

1) He said most, not all.

2) It's beyond-belief true for Microsoft exams. Unless you think like a MSFT marketing manager (and not, you know, an actual sysadmin), the MCSE tests will be impossible to pass. I've gotten near-perfect scores on those things simply by suspending disbelief and thinking like a Redmond Marketing droid.

3) I actually agree with you about the Cisco tests - or at least concerning the older ones (not sure about more recent ones, as I haven't had to touch one in years), since they did probe the protocols pretty deeply.

Re:Goddamn it!

[pegdhcp]

Unfortunately that was the case in my time. I should know that, as I was on the list of Cisco Certified SalesPro and SupoortPro trainers, with permission to establish a lab for certification of said *Pro people. I am sure whatever certifications you have are of far more better quality, this is the basis of Cisco training program.

Cisco certification is better than MS and Check Point is better than Cisco AFAIK, with a serious margin of error, based on territory and certification authority. Again this is my (unfortunately true and based on facts) view. Your take naturally can be different, but BS is one of the most effective tools of trade in commercial certification. That is why we still have universities...

Re:Goddamn it!

[ArmoredDragon]

You don't even have to do that. You can just take the MTA, which is an exam that is so easy they permit high school teachers to serve as proctors for it.

Right now I'm trying to get the MCSA for Windows Server 2012 R2, and it's a PITA. The biggest problem with it though is that there are NO training manuals for it that are sufficient. The Microsoft Official so called study guides sold at book stores are a joke. CBT nuggets/trainsignal are also insufficient. Microsoft's reasoning apparently is that you have to be very seasoned with Windows Server 2012 R2 before you can pass the exam. Problem is NO shops run Windows Server 2012 R2 yet. That won't happen until Server 2016 or whatever it ends up being comes out.

Which is annoying because I bought a voucher pack and I just can't find the right study materials for it, and it's about to expire in a few months.

Re:Goddamn it!

[Dishevel]

Seems legit.

Re:Goddamn it!

[CheshireDragon]

The CCNA exam was like this when they first changed it back in October of 2013. It USED to be heavy on, CLI commands, network knowledge, etc. Now it's heavy on troubleshooting and essential to have a lab at home for break/fix.

Re:Goddamn it!

[o_ferguson]

Base eight is just like base ten... if you're missing two fingers.

Re:Goddamn it!

[Richard Dick Head]

It took me 3 attempts to pass that exam and now there are 5 year olds who can pass it?

Hey, buying a recent brain dump and using it to cheat on a cert is a great job skill FYI. There is no way a 5 year old actually passed legitimately, he's just being taught by pops how to win at the corporate game

Re:Goddamn it!

[AK Marc]

I got my CCNA in 1999, and the test had nothing Cisco specific I can remember on it. CCNP had lots of Cisco specific CLI knowledge needed. (Brocade) BCNE had a class that was "CCNA uplift", but the test was interchangeable with the CCNA. Perhaps it's because they, like so many, just ripped off Cisco's CLI.

Re:Goddamn it!

[davydagger]

implying there is legitimacy at all in corporate standards

srsly bro?

Re:Goddamn it!

[doccus]

I justa bout choked on my sandwich thinking a 5 year old had passed MCSE. I guess I didn't read carefuly enough. Yeah, still im,pressive for ANY 5 year old.. when I was 5 I was the only one I knew who could even read or write..

Re:Goddamn it!

[Wolfrider]

It's impressive on one level, but then you realize:

o The kid has no pressing social-life responsibilities

o The kid doesn't -need- to pass the exam to put food on the table

o The kid hasn't been dragged down by the general pain,angst,disappointment everyone goes through in life yet, so he can think clearly without being worried and still sees it as a great opportunity.

Yeah he's bright I'll give him that, but the above factors make it easier for him to pass as well.
 
/Wolfrider

Which says what?

[Loki_666]

Either the kid is pretty damn smart, or else the quality of the MCP exam has become so easy even children can pass it.

Ill go for a little from column A and a little from column B. Bright kid probably (and coaching from Dad helped for sure) but MCP probably isn't worth jack shit.

I remember years ago being asked by an MCSE for help... installing Windows 2000 Server. I was a Novell certified engineer and could do it in my sleep.

Re:Which says what?

[beheaderaswp]

I agree with you. While he's probably a very bright kid and I would not seek to take anything away from him..... it bears comment.

My team used to refer to "MCSE" as "Make Coffee Send Errand". Mostly because of the issue you are pointing out: these guys had no skills whatsoever.

Re:Which says what?

[Thanshin]

installing Windows 2000 Server. I was a Novell certified engineer and could do it in my sleep.

You woke up and discovered you had installed Windows 2000?

Quite the scary illness you've got there. I'd rather find my horse's cut head.

Re:Which says what?

[dissy]

You woke up and discovered you had installed Windows 2000?
Quite the scary illness you've got there. I'd rather find my horse's cut head.

I don't think they added the horse head option to the installer until Windows XP...

Re:Which says what?

[ruir]

I used to refer to it as Minesweeper Certified Solitaire Expert...

Re:Which says what?

[Anonymous Coward]

Obligatory: "Must Consult Someone Else".

Re:Which says what?

[hcs_$reboot]

There are many changes happening at MS. Since Nadella has been appointed CEO a huge PR operation is in progress. Whatever brings publicity to MS affairs is a good thing, e.g. a 5 yo passing a MS exam.

Re:Which says what?

[Anonymous Coward]

You only saw the horse head option if your machine crashed and you had to use "Unsafe Boot".

Re:Which says what?

[hey!]

Add column C: children are geniuses... at certain things. One of those things is acquiring language, something that adults struggle with but which is literally child's play to them. That doesn't mean children -- even precocious ones -- can reason like adults.

The linguistic genius of very young children might well help one pass a test a standardized test with a simple scalar score which depends in part on whether you can talk the lingo the way the vendor wants you to talk. I'm assuming the tests are devised by marketing people rather than people in actually evaluating human capabilities, like people education or research psychology PhDs.

Re:Which says what?

[fuzzyfuzzyfungus]

The part that worries me was: "The hardest challenge was explaining the language of the test to a five-year-old. But he seemed to pick it up and has a very good memory."

Sounds like the kid is pretty bright, might well be pretty impressive in a few years; but 'explaining the language of the test' is pretty much a (much easier) equivalent to 'identifying the problem to be solved'.

As an exercise in mental capacity, I'm definitely not going to knock the kid, I certainly wouldn't have managed it at 5, and those capabilities will likely come in handy, I hope for him that they do.

For the MCP, on the other hand, it seems pretty dire that it can be passed by somebody with an excellent memory; but a need to be coached on what the questions mean. Real life is an open book (and/or google) test; but it is notably unsympathetic about telling you what the questions mean, what sort of answer a given question requires, which questions are actually on the test, which answers trigger a surprise exam about disaster recovery 18 months from now...

If somebody is a 'Certified Professional' I'd much rather seem them have an elegant grasp of what the problem is and what the solution should look like; but check the manual for some registry settings, than be conceptually befuddled but have a perfect grasp of the details.

Re:Which says what?

[Anonymous Coward]

I did the exam for windows NT 4.0, it was a while ago.

But knowing about windows hampers your ability to pass the exam. If you only read the book and never touched a computer that will make that exam easy.

The questions are in multiple choice, but you need to give the 'best' answer. Most of the time three of the four answers will solve the problem.

Just keep in mind: if one of the answers is "Reinstall windows" than that is the correct 'best' answer (this was the 'best' answer of 5 of the questions on the exam), if one of the answers is 'edit the registry' than this is the 'worst' answer (even if it solved everything and is the quickest/easiest way of doing it).

Re:Which says what?

[drinkypoo]

Obligatory: "Must Consult Someone Else".

I guess this is the age of Oracle but we used to say "Moron Confused by Sun Equipment"

Can we do Novell and Cisco certs next?

Re:Which says what?

[Anonymous Coward]

installing Windows 2000 Server. I was a Novell certified engineer and could do it in my sleep.

You woke up and discovered you had installed Windows 2000?

Quite the scary illness you've got there. I'd rather find my horse's cut head.

How does a horse head and torso compare to Windows ME?
What about Windows Vista? Is 8 equivalent to two horse limbs?

Re:Which says what?

[WalrusSlayer]

You woke up and discovered you had installed Windows 2000?

Scary what one can accomplish while on Ambien and not remember a thing the next morning.

Re:Which says what?

[Anonymous Coward]

Why does either one have to be the case? I really wasn't much older than him when I began to dig into Red Hat Linux and set up mock servers for fun, nor was my brother or my best friend. Neither of them want a career in this kind of stuff, nor do I, was just something we played with on rainy days. This stuff really isn't as hard as everyone makes it out to be, just a lot of shit to remember.

Re:Which says what?

[Anonymous Coward]

Nonono, MCSE stands for Microsoft Certified Shutdown Engineer.

Re:Which says what?

[damien_kane]

How does a horse head and torso compare to Windows ME?

WinME was a horse's ass, not head;
As opposed to a llama's ass, something of which Winamp really whipped.

Re:Which says what?

[MachineShedFred]

The days of the NT 4 paper MCSE are over - Microsoft fixed that by making the infrastructure exam a complete bitch several years back.

MCP is only one of the MCSE tests though, so pick the easiest (workstation cert) and get yourself a certificate in about an hour of brain dump reading.

Re:Which says what?

[GoJays]

MCSE = Must Call Someone Experienced

Re:Which says what?

[MachineShedFred]

The NT 4 MCSE exams were a joke, because Microsoft was still trying to get people in the door against Novell.

Once word got back to Microsoft that nobody cared about the MCSE battery of certifications because they were viewed as being slightly tougher than CompTIA certs, they revamped the entire thing for Windows 2000, and made it actually something you needed to study for.

But it was too late - even 14 years later everyone views the MCSE as a joke.

Re:Which says what?

[Anonymous Coward]

I believe the issue at hand was the five year old part. Think about it, if you start at 10, you had (up to) 7 years of reading experience. If you start at 5, we are talking about 2 years. It's a huge leap.

Re:Which says what?

[Anonymous Coward]

That's what Facebook is for.

Re:Which says what?

[Anonymous Coward]

So what does it say about universities when fourteen year olds can get a degree?

Re:Which says what?

[Anonymous Coward]

Microsoft does this PR stunt where a child pass a cert every year.

Re:Which says what?

[geekmux]

Either the kid is pretty damn smart, or else the quality of the MCP exam has become so easy even children can pass it.

I'd say it was a little more of the latter shining like a polished turd here, since a 5-year old no matter how bright he or she is, is still a child.

Ill go for a little from column A and a little from column B. Bright kid probably (and coaching from Dad helped for sure) but MCP probably isn't worth jack shit.

Well... now it's not worth jack shit. How the hell am I supposed to compete with a 5-year old's salary requirements? And you thought we had a problem before with H-1B holders impacting the job market? Pfft.

I remember years ago being asked by an MCSE for help... installing Windows 2000 Server. I was a Novell certified engineer and could do it in my sleep.

There were plenty of paper MCSEs back in the day. Most likely you did run into someone who Must Consult Someone Experienced.

Re:Which says what?

[Hognoxious]

That rather depends on how many fourteen year olds are getting them, doesn't it?

Re:Which says what?

[Anonymous Coward]

If you only read the book and never touched a computer that will make that exam easy.

Sadly this is still true. I took the 2008 BI exams, and failed the first time because I actually use the fucking technology in an enterprise setting.

Once I realized they wanted the book answers, and not the actual real world answers, it made it much easier the second time round.

Re: Which says what?

[Anonymous Coward]

If it's too difficult, less people will try, I assume. It's Microsoft. Honesty is not really their speciality.

Re:Which says what?

[fustakrakich]

Yep, The grasp of fundamentals isn't a priority anymore. They teach pilots how to work an autopilot, but not how to fly the plane.

Re:Which says what?

[Anonymous Coward]

Before AOL bought it

Re:Which says what?

[Anonymous Coward]

McDonald's Certified Sandwich Engineer

Re:Which says what?

[pr0fessor]

Well the thing about certifications are memorization, some people can memorize a large volume of information for a short period of time with no real understanding of the subject. I aced history in high school, I would read chapters from the textbook the night before and ace tests but not retain any information. The teacher accused me of cheating but was even more disappointed when he found out I wasn't actually cheating and how I did it.

I've worked with plenty of fresh out school kids with degrees and certs that they got mostly by memorization. They end up learning on the job.

Re:Which says what?

[Anonymous Coward]

The part that worries me was: "The hardest challenge was explaining the language of the test to a five-year-old. But he seemed to pick it up and has a very good memory."
Sounds like the kid is pretty bright, might well be pretty impressive in a few years; but 'explaining the language of the test' is pretty much a (much easier) equivalent to 'identifying the problem to be solved'.

It's not like his dad could sit in on the test with him to actually explain each question. He no doubt spent a lot of time explaining sample questions, but to me the sound of the statement is that the kid needed to understand a specific vocabulary (which certainly those tests use to the fullest, i.e. the difference between "shall" and "may") so for that I give the kid a lot of credit. Even if he only takes away the vocabulary skills of a professional exam he has gained something that will go a long way in any field he wants to go in to.

Re:Which says what?

[Anonymous Coward]

The part that worries me was: "The hardest challenge was explaining the language of the test to a five-year-old. But he seemed to pick it up and has a very good memory."

Funny, that's what I consider the impressive part. Because it means that the answers have not just been matched in Chinese-Room manner to the questions.

In college, I had one final examination where I did not bother attending class (too early) and instead just wanted to get the script about a month before the exam. There wasn't one in this course. And nobody I knew had taken it. So I prepared a month long real hard just based on the title of the course ("speech processing").

It was an oral exam where one had something like half an hour to prepare after being given the questions, and there was at least one question where I said upfront in the exam: "Ok, I don't know the terms in the question. What do they mean?". The professor told me, and then I was able to derive what it took to answer the question.

The professor was actually enjoying this exam since there was no need to figure out real knowledge from rote repetition of coursework. Ithink I was really lucky to have a professor without the typical balloon-sized ego that would have just have shut down upon seeing a student without class attendance.

At any rate, I consider it a good sign that the kid answered the questions after having them explained rather than without having them explained.

Re:Which says what?

[CheshireDragon]

Indeed...I still have version 2.95. Was the last best version they had, IMO.

Re:Which says what?

[Anonymous Coward]

Does that mean his dad was with him during the test explaining the questions? Since when is that allowed? I'd like to have that too.

Re:Which says what?

[MildlyTangy]

Serious question here, but how did you determine that you were able to install Windows 2000 Server whilst asleep?

Was it due to sleepwalking?

Re:Which says what?

[Half-pint HAL]

I heard the horses' heads were installed by Microsoft sales agents when public bodies announced plans to use Linux and OOo/LibreOffice....

Re:Which says what?

[Darinbob]

It's a certificate. Does any employer actually treat these as important for anything other than grunt jobs?

Re:Which says what?

[Darinbob]

I remember back in the 90s I asked the only IT guy at our small company who was Novell certified a very basic question about netware protocol as I had never used PCs before, and his response was "sorry, we didn't cover that in class."

Re:Which says what?

[Darinbob]

I worry that this poor kid is being funneled into a bleak and dreary career path too soon in life.

Re:Which says what?

[Anonymous Coward]

So what does it say about universities when fourteen year olds can get a degree?

Even more importantly, what does it say about universities when a person with Down Syndrome can get a degree.

Re:Which says what?

[Ravaldy]

That was always the problem with the MCSE. It didn't actually test your troubleshooting skills like the CISCO hands on tests.

Re:Which says what?

[Dr_Barnowl]

I'm really not sure it puts out a good image of your product if a 5 year old can pass your professional certifications.

It either means your certs are weak sauce, or your product is lamentably simple.

Re:Which says what?

[Anonymous Coward]

I would say the same about all tech certifications, including Cisco's and Red Hat's.

Re:Which says what?

[Anonymous Coward]

The days of the NT 4 paper MCSE are over - Microsoft fixed that by making the infrastructure exam a complete bitch several years back.

MCP is only one of the MCSE tests though, so pick the easiest (workstation cert) and get yourself a certificate in about an hour of brain dump reading.

Hahahahahahahahahahaha
When you can download the question pool for pretty much every Microsoft exam with answers and explanations he days of the paper MCSE are still here.

Re:Which says what?

[David_Hart]

I'm guessing that you are trolling... I too took the NT 4.0 exams and not one of the answers were ever "reinstall Windows". Microsoft also never had answers that ever involved editing the registry.

I do agree that many of the questions could be solved in multiple ways, if you knew how, and that the exams expected the answers from the book or exam guide. That part was very annoying, especially when they listed both correct answers.

Re:Which says what?

[AK Marc]

They even changed what MCSE stand for. Microsoft Certified Solutions Expert (MCSE) now, was Systems Engineer.

And another reason for the name change was because they "withdrew" the old cert.

Re:Which says what?

[Half-pint HAL]

I really wasn't much older than him when I began to dig into Red Hat Linux and set up mock servers for fun,

Get off my lawn!

Re:Which says what?

[shocking]

I remember years ago being asked by an MCSE for help... installing Windows 2000 Server. I was a Novell certified engineer and could do it in my sleep.

Must Consult Someone Experienced

Re:Which says what?

[davydagger]

the MCSE was always the running joke of the internet. It never really required talent. It was for suits though, as opposed to geeks. Woe is me if I lived long enough to see slashdot forget about suits and geeks.

Re:Which says what?

[davydagger]

There are many changes happening at MS. Since Nadella has been appointed CEO a huge PR operation is in progress.

you mean the public relations isn't smearing open source software, making all kinds of empty threats, trying to get FOSS devs classified as terrorists. Use of PR in combination with legal team to strike terror and fear into the computer industry as a whole. Or mabey he means they might remain more positive, and send less unknowledged paid employees to heckle at train shows so vicously, slashdot used to call them "brownshirts" after the Nazi SA. They stood out because they were overly aggressive, had "facts" that didn't jive with reality, and routinely embarresed themselves in front of the people who actually developed software.

Whatever brings publicity to MS affairs is a good thing

and why is that? so they can continue to rake in money from an operating system they didn't contribute to, one bit(android), or had anything to do with, because they were able to legally wrangle the profits away from the company, using byzantine law they helped get passed with campaign donations they made, using money they earned winning market share with anti-competative practices?

a 5 yo passing a MS exam.

Once you get past the "awww cute" factor, you realize that microsoft exams are so easy a 5 year old can do it, so don't expect more out of anyone with that kind of cert.

I think we get to the next question, is since the sole purpose of the microsoft cert is not to learn about computers or make oneself a better operator,programmer,or engineer, but to go to get a job and go to work.(and mabey indoctrinate people on microsoft products) This is promoting child labor almost.

Re:Which says what?

[Anonymous Coward]

I agree with you. While he's probably a very bright kid and I would not seek to take anything away from him..... it bears comment.

My team used to refer to "MCSE" as "Make Coffee Send Errand". Mostly because of the issue you are pointing out: these guys had no skills whatsoever.

Young kids have wicks for brains. They soak up everything. Eventually the wicks get saturated and the soaking up slows down to adult speed.

Re:Which says what?

[Anonymous Coward]

There's two different MCSEs now. There's the old MC Systems Engineer, and the newer MC Solutions Expert. They are totally different things.

Re:Which says what?

[cwsumner]

I'm really not sure it puts out a good image of your product if a 5 year old can pass your professional certifications.

It either means your certs are weak sauce, or your product is lamentably simple.

I think it was a -very- unusual 5-year old.
You might want to remember his name, for later times... 8-)

Microsoft...

[Anonymous Coward]

MSCP: So easy a 5 year old can do it.

Or alternatively:
Microsoft: Engineered for ages 5+.

slow news day /.?

[Anonymous Coward]

We had this a few weeks ago on here!

Note to IT recruiters.

[BeCre8iv]

A MS cert does not trump a computing degree.

Re:Note to IT recruiters.

[Lumpy]

An MS cert does not trump someone with 2 years of real experience. MS certs are only for people that have zero experience hoping to work somewhere other than best buy.

Re:Note to IT recruiters.

[Anonymous Coward]

You have to pass many exams and coursework to obtain a degree. It's value is far higher, provided your university has a good reputation (ie not an online 'university' but a real one that actually does rigorous testing and ranks highly compared to others).

Re:Note to IT recruiters.

[Anonymous Coward]

Funny how both of you fail basic grammar. Even this non-native speaker spotted it.

Re:Note to IT recruiters.

[OneSmartFellow]

Grammar Nazi strikes again.

it's/its

Re:Note to IT recruiters.

[Anonymous Coward]

I find that this is often the case. As a native British English speaker, I rank suitably well on a variety of standardised tests in the language -- however, I frequently ask competent "English as a second language" speakers to proof my work.

You see, at least in the UK, most English language education omits discussion of the rules and functional components of the language so diligently taught abroad. Rather than teach the correct approaches to writing the errors are simply corrected and some students never recognise the patterns, even subconsciously. I sometimes wonder if the educators themselves know the actual rules, rather than relying on what "sounds right?"

English language teaching for native speakers (again, in the UK) is simply used to provide a basic, practical, working understanding and use of the language, rather than transferable understanding and linguistic properties. In particular I hear this complaint from many French language instructors who struggle to explain the many pattens in their language to their students who have no fundamental knowledge of how language is constructed.

Of course, the English language is a living thing and will undoubtedly change over time. I know it's fun (for some) to ridicule those who have only focused on practical usage, with little consideration for what is "proper". But surely the purpose of language is to communicate, and to some extent the fact that you understood enough to identify fault suggests that, on the whole, the author was successful.

Now, have at it. I'm sure there are more than a smattering of errors here, but c'est la vie.

Re:Note to IT recruiters.

[wed128]

This depends highly on
1) the reqiurements of the position
and
2) your hiring budget.

Do you need someone to run msi installers and register users on a *windows* system? you'll probably save some money hiring an MS cert. do you need database design or some complicated distributed server farm built? and IST degree is probably what you're looking for? Need some application code written? You probably want someone with a CS degree. Need device drivers/firmware/hardware designed? Go for someone with a Computer Engineering or even an Electrical Engineering degree.

All of these degrees/certifications don't fit the same niche...

Re:Note to IT recruiters.

[flappinbooger]

An MS cert does not trump someone with 2 years of real experience. MS certs are only for people that have zero experience hoping to work somewhere other than best buy.

I once had to help a MCSE guy configure an email account. In outlook. They have no value if there is no experience backing up the ability to pass a test. I've been working with computers for over 30 years, and I'm 40. No official certified training, just a lifetime of being a geek. They still call me for the hard problems and never complain about the bill when it comes.

My understanding is there are "boot camp" test prep courses that are worthless then there are (apparently) some real cert programs.

Re:Note to IT recruiters.

[frank_adrian314159]

A MS cert does not trump a computing degree.

It depends on how much the five year-old costs compared to someone with a computing degree.

Re:Note to IT recruiters.

[cyberchondriac]

Funny how both of you fail basic grammar. Even this non-native speaker spotted it.

Thats redeculous. Your such a looser. I english just find.

Re:Note to IT recruiters.

[Required Snark]

When will the 5 year old be replaced by a 1H-B because it's cheaper?

Re:Note to IT recruiters.

[Anonymous Coward]

And passing the Bar exam or having a PhD in astrophysics doesn't trump being MS certified in a lot of cases.
 
Your post has no real value as it has no real context.

Re:Note to IT recruiters.

[Anonymous Coward]

Depends on the person more then the degree. We have had absolute idiots hired in that had computer degrees. Could do the book work but had no real world experience. Anyone that has either been through the certification process or went toward a degree in the computer field knows there is the way these programs teach and then there the way it is actually done in the real world. If you followed how you were taught you'd have a costly ineffective network.
 
Another example of this is why a lot of times artistic people are better at programing then people that have specialized degrees in the field. They tend to write cleaner, more flowing code.

Re:Note to IT recruiters.

[AqD]

MS certs are only for people that have zero experience hoping to work somewhere other than best buy.

And zero enthusiasm plus zero creativity at work, making them rather useless.

Still it's better than MVP which is a sign of serious brain damage - you could typically find these guys at M$ forums, giving pointless answers saying you cannot do this or that, or you must do certain things in specific order without knowing why, like mindless robots.

Re:Note to IT recruiters.

[Anonymous Coward]

A MS cert does not trump a computing degree.

An MS cert is at least a card.

Story time.

A dancing partner of mine was into rowing in her youth. She happened to be in the same rowing club that the eventually gold medal winning olympic team for the eight was training at the moment. One of the rowers fell ill, and since this was a sweep-oars boat class, he would be required for balance. So they grabbed her (she was in pretty good shape). Quite traumatic for her, being 15 or 16 at the time, seated between all those giants and tasked with "just" moving the oar along in time, never mind actually pulling, let alone pulling her weight. I think it turned out that she could not sustain a sprint, and they ended the training session early anyway.

But she probably was still doing better than someone whose proudest qualification in rowing would have been the figurative equivalent of an MS cert. That's probably more akin to "get into the boat facing in the right direction without falling in the water while a client is watching".

Re:Note to IT recruiters.

[__aaclcg7560]

A MS cert does not trump a computing degree.

A computing degree doesn't indicate that a person knows jack about computers. I once had to explain to a fresh out of college software engineer at Google that he needed to press the power button to turn on his computer. Unlike a school lab environoment, no one else was going to turn it on for him.

Re:Note to IT recruiters.

[frank_adrian314159]

This week on AskHR!

When will the 5 year old be replaced by a 1H-B because it's cheaper?

This is a simple question for any HR professional! It's very obvious when you think about it. You're assuming that all engineers are equivalent - they're not! Each of them have different cost functions and must be handled as an individual on that basis! As such, the H1-B will replace the five year-old when the H1-B cheaper than a five year-old. And the easiest way for that to happen is if the H1-B in question is a five year-old! As, I said, simple... At least for an HR Professional like me!

Remember to send your questions to AskHR! Everyone who asks a question gets entered into a drawing for a free pink slip! Not that you weren't already in that lottery, but... May the odds be ever in your favor.

Re:Note to IT recruiters.

[Anonymous Coward]

It's not an either/or scenario.

What about when the local 5 year old is replaced by an H1-B 5 year old?

Re:Note to IT recruiters.

[Anonymous Coward]

CS grads don't write applications. They talk about how other people ought to write applications.

And your shift key appears to be sticking.

Re:Note to IT recruiters.

[Darinbob]

An MS cert does not trump a good attendance star from grammar school.

Re:Note to IT recruiters.

[Darinbob]

5 year olds are expensive. They have much higher medical expenses, as their parents are taking them too the doctor every few weeks for vaccinations, ear infections, sniffles. A twenty year old who never goes to the doctor and who is satisfied sharing a studio with three roommates and eating every night at Taco Bell is the best bargain.

Re:Note to IT recruiters.

[AK Marc]

And how does a computing degree relate to server admin?

Re:Note to IT recruiters.

[AK Marc]

Depends on the local laws in the country you are outsourcing to.

Re:Note to IT recruiters.

[AK Marc]

I don't see a grammar error.

http://www.gpuss.co.uk/english...

And the OP is strictly correct if MS is pronounced "Microsoft" and the abbreviation is only for saving typing letters.

For me, I always sound out "Microsoft" because one wouldn't say "MS" in mixed company in the spoken sentence, because the meaning would be so unclear.

Re:Note to IT recruiters.

[Half-pint HAL]

The problem is that they spent ages trying to teach artificial "rules" of English and when kids just stuck with natural English, they concluded that "grammar teaching doesn't work". No, trying to teach kids to use "may" instead of "can", "whom" instead of "who" etc doesn't work. Focusing on awareness of genuine patterns does work, eg "he's, she's, it's... apostrophe. his, hers its... no apostrophe". Grammar is easy to teach to natives, because they already know it. If they don't know it, then it isn't the grammar of the language.

(ESOL teacher and language major.)

Re:Note to IT recruiters.

[Anonymous Coward]

Computing degrees from over priced daycares.

I am enrolled at one at the moment, and I am so unimpressed with the teaching with this world classed University that I dropped out and got a job doing the very thing I was studying.

It is moronic HR with their text book cookbooks that say employable people need degree X to do a job. BULLSHIT! With the right on the job training and guidance, as demonstrated by the source article anyone can become an expert in the field.

Re:Note to IT recruiters.

[davydagger]

lets put this to rest right now. In English, there is no authorative definitions(real rules) on either spelling, or grammar. Just conventions, and they are diffrent in diffrent places in the world.

There litterally exists no such thing as correct spelling and grammar in the English Language.

This differs from Spanish and French, which there are official bodies governing them, and actual real rules.

Re:Note to IT recruiters.

[davydagger]

the 5 year old is the H1B. the days of sweatshops will come back folks.

Re:Note to IT recruiters.

[cwsumner]

The problem is that they spent ages trying to teach artificial "rules" of English and when kids just stuck with natural English, they concluded that "grammar teaching doesn't work". ...

Not knowing the differences may be fine ... Until you need to communicate that difference to someone else, accurately.
Or, you need to write something that people expect to read very fast, without any stumbling over things. If you fail, they don't read it and you fail.

Too bad for MCPs?

[Anonymous Coward]

Well, either the boy is genius or the Microsoft Certified Professionals (MCPs) are retards... or even both!

Re:Too bad for MCPs?

[camperdave]

There is a third option: The boy is a "paper" MCP. He knows the right answer to the questions, but doesn't understand the reasoning behind it.

Re:Too bad for MCPs?

[Lumpy]

This is 95% of all MCP holders.

Re:Too bad for MCPs?

[140Mandak262Jamuna]

There is a third option: The boy is a "paper" MCP. He knows the right answer to the questions, but doesn't understand the reasoning behind it.

What part of Microsoft in MCP you did not understand? There is no reasoning behind it. Other than, it looked like a great way to screw some competition way back when we could do it. The only other reasoning other than that is, "the newbie code monkey hacked it this way and his/her manager was too stupid to catch it code review. Now it is carved in stone".

In other words, the reasons are either malice or incompetence.

Re:Too bad for MCPs?

[DigiShaman]

Honestly, the certification should back up field experience, not the other way around.

Been working IT since 1996. MCITP, MCDST, MCTS, MSCE (2000).

Re:Too bad for MCPs?

[Anonymous Coward]

This is 95% of all MCP holders.

Are you sure the percentage is that low?

Re:Too bad for MCPs?

[Anonymous Coward]

BAH! Working IT since 1985... 16mm, 1 and 2 inch, U-matic 3/4 inch, Beta, M-format, LaCart, Betacart....

Oh, wait... I was doing infotainment technology...

Re:Too bad for MCPs?

[Lumpy]

Yes there is 5% error :-)

Honestly some people that get suckered into the Microsoft Kool-Aid really do know their stuff. And unfortunately a lot of HR departments are far too stupid to understand that certifications mean nothing so they require a billion certs thinking they will get someone that is perfect for the job.

So I can see why youngsters (Under 30) really think they need certifications and they listen to the siren song that Microsoft Bleats as the holy grail to high paying jobs.

A friend of mine is a fantastic thinker and tech, sadly he drank so deep of the MSFT refreshment that he will not even look at a Unix of any type so he is locked into one small skillset that when he has to deal with something outside of that world, he will have no idea what to do.

Exactly why we test all candidates.

[mrthoughtful]

The only way that we have found for being able to assess a candidate's suitability for work at our company is to write tests that suit the job, and then ask the candidates to demonstrate their skills. We've had people with all sorts of qualifications relevant to the LAMP architecture not know the basics of regex, sql, bash, etc. Let alone what ARP is.

IMO qualifications in IT aren't really very relevant, other than showing the intent/interests of the individual. Also, as IT is changing so rapidly, by the time a (non-theoretical) qualification has been published, it is pretty much out of date.

My response, as an employer, to this news could be summarised as: 'We never had much credence to the MS qualification in the first place - and now we have none.

Re:Exactly why we test all candidates.

[goarilla]

The only way that we have found for being able to assess a candidate's suitability for work at our company is to write tests that suit the job, and then ask the candidates to demonstrate their skills. We've had people with all sorts of qualifications relevant to the LAMP architecture not know the basics of regex, sql, bash, etc. Let alone what ARP is.

ARP as in ethernet ip-mac mapping ? How exactly is that relevant to a LAMP job ?

Re:Exactly why we test all candidates.

[OneSmartFellow]

Presumably OP is interviewing for sysadmin positions maintaining LAMP installations, not developing LAMP based services.

Either that or OP is just as clueless as his candidates

Re:Exactly why we test all candidates.

[plopez]

"the basics of regex, sql, bash, etc. Let alone what ARP is."

never ask a question that can be answered by the web,,,

Re:Exactly why we test all candidates.

[Anonymous Coward]

never ask a question that can be answered by the web,,,

Actually, if I wanted to know something about something like ARP, I'd head to the web first. Anything I know on the topic may very well be out of date. Such is modern computer technology.

On the other hand, someone who doesn't even know what ARP (or whatever) is may well be ignorant of other critical things. And/or may not be able to read the web documentation and make proper sense of it in the necessary amount of time.

Sherlock Holmes and the library versus the lumber room...

Re:Exactly why we test all candidates.

[Charliemopps]

Testing during an interview is not fair to the candidate. Rarely do the tests reflect the reality of the job and for some insane reason it's seen as cheating to use the Internet to solve the problem when, in the real world, that's exactly how they should be solving it. The knowledge they have in their head is useless and can be wrong. Doing research before proceeding is always preferred. The real skill is knowing what to search for and what information to trust.

To be honest, if an employer asks me to take a test, I usually just say "No thanks" because it suggests the employer is fearful of hiring the wrong person... usually because the business has policies that make it hard to terminate bad employees. If that's the case, it's doubtful the test actually did much good and I'll end up working with a bunch of people my boss wishes he could fire and I'm likely to soon agree with him.

Re:Exactly why we test all candidates.

[mrthoughtful]

Well, it depends upon the job. As OneSmartFellow correctly divines, a recent post was for a sysadmin / sysops post. We don't require other devs to know what ARP is, but it's always good if they have some idea about the network stack.

We have been repeatedly amazed by the levels of ignorance that IT-qualified candidates have had. One of the most disappointing finds is that very few who have come from university have any substantial programming experience. Likewise, 'hack-a-day' php coders and sql-ers about, but most of them do not know when to apply a left join, some of them don't even know what a key is used for (just think of all that wasted cpu time due to ridiculously poor sql implementations. It makes me shudder).

Regarding the idea of methods for developing a re-usable, maintainable codebase for our work (primarily webwork) - seems to be beyond everyone that we recruit. The team that we have right now is second to none - but we have found that a well-written test reduces the initial number of applicants from about 700 to 800 down to about 10, most of whom we will interview.

Re:Exactly why we test all candidates.

[Charliemopps]

Oh, and I do agree with you. MS certs are completely worthless.

Re:Exactly why we test all candidates.

[Anonymous Coward]

Air Raid Precautions were essential during WW2, but I'm not sure what their relevance is to IT workers in 2014.

Re:Exactly why we test all candidates.

[BlackHawk-666]

People laugh at that comment but he's quite right. It is literally impossible to work today in the programming field with just the information you can remember in your head. You should instead focus on getting people with good reasoning and research skills. People who can learn fast, apply past lessons and derive a correct solution based on your situation.

There's still a need for quite a large amount of information to be remembered by team members, but that is definitely trumped by how fast they can acquire it from resources they will have on hand at their work. For programmers they will have a syntax highlighting, auto-completing editor, project and make file management tools, language references, API references, and of course, vast realms of information on the net.

I was a little surprised the other day when I tried to remember the old definition for an OOP language. Inheritance and encapsulation jumped straight to mind but the third was just out of reach of my memory - despite actually using it every day for almost 30 years; polymorphism. I'd had to think I wouldn't pass some recruiter's idea of an test simply because my memory isn't what it used to be.

There's some skills like regex that I need to look up every time I use them. Mostly because they are used infrequently, have an arcane syntax, an API that varies from one language / environment to the next and every implementation seems to use different syntax. I've written heaps of SQL, I know where to place a key when to leave it down to a table scan. I can find my way through a query plan and figure out which part of a query is nonce...and yet I might still have trouble remembering some vague syntax thing. I have existing code to remind me and the web for when I don't have a snippet.

There's too much emphasis on rote learning of information that often is not even that useful, and not enough on developing analysis, research and planning skills.

Re:Exactly why we test all candidates.

[mrthoughtful]

We do allow our candidates to use the web. We also write the tests specifically for the job at hand. We do NOT penalise for mashing, copying, or even asking for help. What we do penalise is for when someone grabs something and doesn't understand what it is, how it works, and cannot make it 'theirs'.

We also penalise people who copy code from the net and then attempt to pass it off as their own.
We don't monitor the test - we allow the candidate to work against their own clock.

We aren't fearful of hiring wrong people - but we don't have time for them either. We also find it's an extremely good means of filtering out what can be up to 1,000 applications. Those who apply for the job are those who really want to work for us, and are willing to show us their skills.

Our questions tend to be qualitative, which means that it's very hard to 'find the answer on the net'. They will include questions such as (eg for a web designer) - "In what ways could you significantly improve the BBC news website, and why do you think the BBC have not made those improvements already?"

For a (S)CSS engineer, we will be asking questions to demonstrate approaches to carving and presenting a responsive page, based upon a simple flat visual.

For all of these things, there are no right answers, but there are good answers.

The funniest response we once got from a programmer, to about 9 out of 10 of the questions we had on the test for the position he was applying for, was "It's not my department." - needless to say he wasn't shortlisted.

Re:Exactly why we test all candidates.

[Anonymous Coward]

The only way that we have found for being able to assess a candidate's suitability for work at our company is to write tests that suit the job, and then ask the candidates to demonstrate their skills.

Even managers? Directors? Whenever I'm asked to take a test, I ask to see the test my manager took. I have yet to see one yet it makes a huge difference. Ever work for a manager who never had anyone who worked for him advance in company?

My experience with tests is that you tend to fail a test if you don't answer the way the author answered it. "What's wrong with C?" If I think highly of C, I wouldn't never come up with the answer someone who hates C would expect. I might not even come up with the answer if both the author of the test and I agreed on the merits of the C language.

I generally don't accept positions from testers unless the pay is really, really good.

Re:Exactly why we test all candidates.

[Anne Thwacks]

Air Raid Precautions were essential during WW2, but I'm not sure what their relevance is to IT workers in 2014.

Think of the drone strikes! Tin foil is not enough. You need real tin hats.

Put that light out!

Re:Exactly why we test all candidates.

[luis_a_espinal]

The only way that we have found for being able to assess a candidate's suitability for work at our company is to write tests that suit the job, and then ask the candidates to demonstrate their skills. We've had people with all sorts of qualifications relevant to the LAMP architecture not know the basics of regex, sql, bash, etc. Let alone what ARP is.

ARP as in ethernet ip-mac mapping ? How exactly is that relevant to a LAMP job ?

If you do LAMP jobs, chances are you will be doing some of the installation and configuration of infrastructure yourself. Meaning, some sysadmin work will be involved.

Meaning, there is an expectation of being aware of problems that can occur - by accident or stupidity (3rd party or yours) - with any non-trivial setup, including troubleshooting network problems that causes Apache to proxy among several PHP boxes (if you distribute horizontally) or why or database connections get truncated/closed prematurely (damned unknown firewall) or why your system is so slow (until you fire up snoop or something like that and you detect a shit-ton of re-transmitted packages because the NIC on your database server is running full duplex while the NIC on your PHP box is running half-duplex).

Developing is not just writing code. It is deployment. It is testing, and ZOMG, support of the shit we write and deploy.

I for one harbor suspicions of any person that claims web/enterprise development experience (Java, .NET, RoR, PHP, whatever) and yet doesn't know (or at least show some awareness of) the basic protocols that make up the interweebz.

Because if they do, then they are not the type of people one can rely on to make shit work when shit inevitably breaks.

Be very afraid of coding without deployment awareness.

Re:Exactly why we test all candidates.

[luis_a_espinal]

Presumably OP is interviewing for sysadmin positions maintaining LAMP installations, not developing LAMP based services. Either that or OP is just as clueless as his candidates

One cannot make non-trivial LAMP based service development without having at least an awareness of deployment and installation issues. Hell, that holds for any stack (RoR, JEE, .NET).

Re:Exactly why we test all candidates.

[goarilla]

Meaning, there is an expectation of being aware of problems that can occur - by accident or stupidity (3rd party or yours) - with any non-trivial setup, including troubleshooting network problems that causes Apache to proxy among several PHP boxes (if you distribute horizontally) or why or database connections get truncated/closed prematurely (damned unknown firewall) or why your system is so slow (until you fire up snoop or something like that and you detect a shit-ton of re-transmitted packages because the NIC on your database server is running full duplex while the NIC on your PHP box is running half-duplex).

Shouldn't the switch rectify that by buffering ?

Re:Exactly why we test all candidates.

[__aaclcg7560]

Air Raid Precautions were essential during WW2, but I'm not sure what their relevance is to IT workers in 2014.

You don't want your hard drives overheating from a lack of air circulation inside the RAID box.

Re:Exactly why we test all candidates.

[mrthoughtful]

We only recruit managers from our existing workforce, and we only choose directors from our existing managers.
There are only two people who haven't taken a test to be employed - and both are the founders of the company.

Re:Exactly why we test all candidates.

[plopez]

If you can find it on the web so can the interviewee. I always try to ask questions where basic learning must be extended to new situation which will NOT be easy to Google up. I am looking for abstract operational thinking and above.

Re:Exactly why we test all candidates.

[luis_a_espinal]

Meaning, there is an expectation of being aware of problems that can occur - by accident or stupidity (3rd party or yours) - with any non-trivial setup, including troubleshooting network problems that causes Apache to proxy among several PHP boxes (if you distribute horizontally) or why or database connections get truncated/closed prematurely (damned unknown firewall) or why your system is so slow (until you fire up snoop or something like that and you detect a shit-ton of re-transmitted packages because the NIC on your database server is running full duplex while the NIC on your PHP box is running half-duplex).

Shouldn't the switch rectify that by buffering ?

Yes, but when it does, what does the "dev" do? Would he/she even know where/what to look for? Would he/she even had a notion of something going wrong, a notion good enough with which to ask the appropriate questions to appropriate liaisons?

Or would he/she sit helplessly waiting for a "guru" to do some magic (sometimes forever... I've seen it happen)?

Re:Exactly why we test all candidates.

[luis_a_espinal]

Testing during an interview is not fair to the candidate.

Testing during an interview is the norm with good tech employers.

Rarely do the tests reflect the reality of the job and for some insane reason it's seen as cheating to use the Internet to solve the problem when, in the real world, that's exactly how they should be solving it.

OTH, testing to see if a developer actually knows how to write simple programs without the use of the interweebz is perfectly reasonable. Seeing how so many "developers" flunk to write a semi-complex for loop or flunk at explaining why a SQL cartesian join is typically a very bad idea completely legitimizes (in my eyes) the practice of testing.

Re:Exactly why we test all candidates.

[luis_a_espinal]

Yes, but when it does,

Re-quoting myself. I meant to say "Yes, but when it does not".

Re:Exactly why we test all candidates.

[AK Marc]

When troubleshooting your load-balancer and/or reverse proxy to your web server. If you can't keep your server running LAMP up, what good does all the LAMP training do?

Re:Exactly why we test all candidates.

[AK Marc]

You can't buffer your way out of duplex mismatches. And basics, like the L2 issues (including ARP) cover why you want everything set to auto/auto, or fixed/fixed.

People still don't believe that auto/auto on one side and 100/full on the other should have the auto/auto set to 100/half under proper (Standards based) operation. Yes, even on 100FX (two-strand fiber, only one strand used on one side, causing massive issues, didn't help that they exceeded the distance for Ethernet, increasing the number of collisions).

So long as the admins think they need to know layer 7 only, the networking gurus will make a comfortable living.

Re:Exactly why we test all candidates.

[Half-pint HAL]

One of the most disappointing finds is that very few who have come from university have any substantial programming experience.

I don't think most engineering firms complain that engineering graduates don't have any substantial programming experience, or that law graduates don't have any substantial legal experience. Why does the computing world expect to be any different?

How long is a real-world programming project? 2, 3 years? Try to implement that in a university setting and you're basically starting on it while you're still dealing with the basics, and you end up with a programming degree (as opposed to CS) that really doesn't equip the students for deep reasoning about software engineering.

The other complaint about CS is they teach idealised notions that aren't applicable to the real world.

But these are two facets of the same problem, and it comes down to two different (non-exclusive) notions of "expertise" (equally applicable in fields as diverse as medicine, flower arranging and bridge building): the academic expert knows most the options, and can spend a long time choosing between them, even researching the options he doesn't know; the professional expert knows enough options that he can build something that works and get the job done in a short time.

I say that the job of the CS degree isn't to make an academic expert, but to equip the learner with a broad toolbox of concepts so that they can become a better professional expert once they become professionals. A professional without broad knowledge ends up becoming a hacker, kludging up inelegant solutions and producing a codebase that is esoteric and hard to maintain.

It is that breadth that sets the CS student apart from both the programming student and the self-trained coder.

Re:Exactly why we test all candidates.

[Half-pint HAL]

You should instead focus on getting people with good reasoning and research skills.

If you'll indulge me in a little reductio ad absurdum... you mean people with a joint degree in Philosophy and History?

The "knowledge is dead, long live data" philosophy is appealing, but it's easy to overstate it. Fundamental concepts have to be learned, only superficial detail and facts can be looked up. Changing from programming Lisp to Java isn't just a matter of looking up a language description on the net, there's a lot of concepts that need to be learned. I think it's fair to want people with a basic understanding of the full network stack when working with a technology... but not for a grad. If we add together all the computer knowledge that every computer recruiter thinks is "missing" from CS degrees, we'd end up designing a 20 year long programme....

Re:Exactly why we test all candidates.

[Half-pint HAL]

Testing during an interview is not fair to the candidate.

Only if the test is unfair. Many interviewers are happy to hear "I'd pick up the phone" in answer to multiple questions -- it shows you accept that your knowledge is limited (which is particularly appreciated by graduate recruiters). A wrong answer can also be perfectly acceptable if you make it clear that you're not sure and you highlight the assumptions you make (eg "assuming Visual BASIC arrays are 1-indexed" (last I knew VB arrays were 0-indexed, however they had inherited BASIC's unusual n+1 length -- ie dim arrayname 3 in BASIC gave a four-element array: 0, 1, 2, 3. This is for a weird jumble of historical reasons, so you wouldn't expect everyone to understand this.))

Re:Exactly why we test all candidates.

[BlackHawk-666]

"If you'll indulge me in a little reductio ad absurdum... you mean people with a joint degree in Philosophy and History?"

Heh.

I'm not a huge fan of typing, since my fingers ache most the time from a life at the keyboard. So long as I managed to convey the gist, I'm reasonably happy to let the reader follow the train of thought on their own.

Anyway, good points there, all stuff I agree with. I went to uni in '87, and I can still remember to this day the difficulty of the switch from BASIC (my first language) to Pascal (where's the line numbers, no goto, what!), the introduction of C, and then...WTF...Lisp! What the hell is going on here? You definitely need a different state of mind and a different set of skills to use Lisp instead of most modern imperative languages.

You can jump from Java to C# and back pretty easy. Just learn the libraries and a few foibles and their "preferred style of coding" and you should be set. You can pick up a C based language pretty fast too...just learnt the ways it's different to C and hit those libraries again.

Ideally, you'd try and test a candidate on their ability to absorb knowledge, comprehend, reason, design, prototype and their understanding of the fundamentals. Most everything else is fru fru.

"If we add together all the computer knowledge that every computer recruiter thinks is "missing" from CS degrees, we'd end up designing a 20 year long programme...."

I've been at this 35 years now and all I know is that there is a vast ocean of things I still do not know.

Re:Exactly why we test all candidates.

[goarilla]

You can't buffer your way out of duplex mismatches. And basics, like the L2 issues (including ARP) cover why you want everything set to auto/auto, or fixed/fixed.

With modern gigabit switches why would you meddle with manual duplex settings anyway. I haven't had to
deal with these issues for some time now and have forgotten almost everything about it.
The only thing which I do still remember is that ICMP isn't to be relied upon when troubleshooting these issues.

But this goes back to the job recruiter wanting an Allround "specialist". Which is contradictionary.
Either you have someone with broad knowledge or a specialist.
Hopefully you have a specialist with a passion who is willing to acquire broad skills on a ad-hoc basis.
But you will not know this unless you let them try out.

That you will get a lot of people who oversell their skillset is a symptom of the bombastic job descriptions of today.
I don't dare to apply for a job these days, I feel completely unqualified and unaccomplished.
This in contrast to a big part of the market which just bombards recruiters with sollicitations and hope that something sticks.

Re:Exactly why we test all candidates.

[AK Marc]

With modern gigabit switches why would you meddle with manual duplex settings anyway.

Because, as the topic is about, a server admin who is good at maintaining the programs, but has no knowledge of basic networking will want to ensure that the server doesn't run at "half" speed, so will set the server manually to 1000/full, when the networking guy set the switch to auto/auto.

Then, after the server guy screws up basic config, he complains to everyone in the IT department (and every other department that doesn't kick him out) about the poor network, making his server run slow.

I've seen it many times.

I'm afraid that he will be "no life" in the future

[Anonymous Coward]

Sad thing.

The problem with certifications

[Tyr07]

The problem is that you only have to remember specific information to pass a certification.
This doesn't mean you know /why/ it is the way it is. Only that it is.

This makes the difference between IT professionals who love their work, and will find new solutions, versus the sys admin who can follow the instructions laid out before him.

You might be able to say 'FTP uses TCP' 'Voice uses UDP' Oh, you passed.
This doesn't mean that they understand that TCP sends acks for each received packet to insure each one is received in order, and if it isn't, it resends the package.
That UDP sends it blindly hoping it worked.

Or why. E.G FTP uses TCP because you're transferring files, if part of the file is missing, it's toast. So you need acknowledgement for each packet sent.
UDP is used for voice. This is because with voice, if part of it is missing you get a slight drop in quality, a blip or blurp in sound, machine like sounds or depending how long, a missed word or two.

You can still understand the conversation for the most part with minor impact. If it was TCP, people would cut out a lot more vs the odd artifact in the sound.

Re:The problem with certifications

[Anonymous Coward]

You might be able to say 'FTP uses TCP' 'Voice uses UDP' Oh, you passed.
This doesn't mean that they understand that TCP sends acks for each received packet to insure each one is received in order, and if it isn't, it resends the package.
That UDP sends it blindly hoping it worked.

Or -why- the latter is better for voice. Answer - the acknowledgement means higher latency and requiring all previous packets have been received means lag keeps increasing when you have packet loss, when really people often don't notice the occasional dropped packet and can just say pardon if they do.

Re:The problem with certifications

[Anonymous Coward]

This makes the difference between IT professionals who love their work, and will find new solutions, versus the sys admin who can follow the instructions laid out before him.

Certifications are thus very useful for a particular kind of hiring manager: It means the candidate has a baseline rote scripts stuffed in his head and successfully replayed them on the exam. That makes for a nice standardised and thereby easily replaced cog for the machine.

IMO sysadmins aren't executors of small scripts, they write them. They mightn't be big on fancy new trickery but they're good at finding root causes and taking out the problem root and branch. In that sense "windows doesn't need administration" -- it cannot stand that sort of administration or it'll fall over. Hiring managers don't want custom solutions, they want easily replaceable ones, just like their workmen. But so windows does require a disproportionally large amount of workmen with digital grease scripts in their head to keep going the creaking things that vendor sells.

Re:The problem with certifications

[camperdave]

In other words, the error correction process is moved up to layer eight.

Re:The problem with certifications

[Luthe_Faydwire]

A number of applications to transfer files have moved the control out of the network for various reasons. UDT, TFTP are two UDP only examples. Torrents can be UDP or TCP.

For most the people working in IT the protocol does not matter except to update filters. For those building specific applications or the networks that support those applications the understanding is often critical.

The problem with most certifications is that it is a paper test. This is a compromise to make the system easy to confirm pass / fail. The alternatives are much more expensive from a resource perspective.

Re:The problem with certifications

[dfn5]

The problem is that you only have to remember specific information to pass a certification.

For the most part your right. 99% of certifications are useless and I ignore them when trying to hire someone. However, I kind of like the Cisco certifications (the advanced ones anyway), where you are asked to build and configure something in a lab, then they come in and screw with it, and you have to figure out what is wrong and fix it. Being able to fix something that is broken demonstrates a deeper understanding than mere memorization of facts.

Re:The problem with certifications

[Anonymous Coward]

All this sex and violence on TCP is going to warp the kid's mind. Better call the FCC and CPS...

Re:The problem with certifications

[HeckRuler]

Oh my god, I don't know why I found this so funny, but "moving the error correction process up to layer eight" is now my favorite euphemism. It's a feature!

Re:The problem with certifications

[Anonymous Coward]

Too much time spent in the OSI model?

It's a pretty good observation that the human is level 8. There is where many issues with software arise (from misuse, ignorance, or normal human mistakes).

Re:The problem with certifications

[AK Marc]

Or why. E.G FTP uses TCP because you're transferring files, if part of the file is missing, it's toast. So you need acknowledgement for each packet sent.

You fail. FTP works fine over UDP. You just call it TFTP then.

. If it was TCP, people would cut out a lot more vs the odd artifact in the sound.

If your transit time is less than your jitter buffer, you'd have better quality with TCP voice over UDP voice. As they aren't, the "lower overhead" (not better quality) is the deciding factor.

Again, I'd mark you as a fail.

Double fail for posting this as an example, and being wrong on all points. Though you are close.

Re:The problem with certifications

[Anonymous Coward]

This sounds like good technical speak for a script I'm working on.
Thanks,
Arash Amel.

Re:The problem with certifications

[Tyr07]

Alright you got me, you can increase latency to compensate for voice quality over TCP but it doesn't work for what we consider natural real time communication.

So if you were failing people for this, you would be a terrible person to be marking scores. TFTP is old as shit and rarely used, so that's ignorance there and often for small amounts of data, throughput suffers for large volumes at high speeds on todays internet. So you would fail as an instructor.
Last time I saw TFTP was to do with network boot protocols, never seen people use it for normal transfers over public internet.
You would have severe issues using udp for FTP over a realistic network distance over the internet.

So what I said was 100% right. Realistic world side of performance, for real time communication, if you switched them over to TCP, you would either have A - longer latency for communication or B - artifacts and jitter.

You sir, are the biggest douche the unnniiiveeerrrseee...

P.S Yes in your own private networks any of these protocols can work fine since you control the network. Talking to someone in europe from Canada does not give you control over the network and where my statement 'the internet' makes them correct. So don't try to get technical to weeny out of what looks like a troll post.

Re:The problem with certifications

[AK Marc]

TFTP is old as shit and rarely used,

Massive fail. It's a (the most?) common boot loader for network gear with config loaded at boot, most commonly used these days for IP telephones, though can be used for larger and more complex networking gear, but the people that work on networks trust them least for such things.

TFTP is newer than FTP, and you used that in your own example. So if old disqualifies something, FTP should be disqualified about 10 years before TFTP. I do more networking than server work, so I use TFTP more than FTP, so again by rarity of use, your example is worse than mine, despite your ill-informed protestations.

You sir, are the biggest douche the unnniiiveeerrrseee...

When some doucehbag tries to lecture others and is obviously (and easily) provably wrong, yes. I'll go douche in response to douchebaggery. You know enough to be dangerous, but I'd never hire you.

, if you switched them over to TCP, you would either have A - longer latency for communication or B - artifacts and jitter.

You'd not have any artifacts. You'd get no more drop with TCP than UDP. Latency, no. The latency would be identical. TCP takes no longer to travel than UDP. Jitter? Maybe, so long as you set poor window sizes.

You ignored my correct reasons, and tried to correct me with more worthless shit designed to attack and distract. You are an uninformed idiot who thinks himself an expert, but knows barely enough to poke fun at others, and not enough to actually do anything.

Talking to someone in europe from Canada does not give you control over the network and where my statement 'the internet' makes them correct. So don't try to get technical to weeny out of what looks like a troll post.

You are 100% wrong. I'm 100% right. I give facts. You give insults only. I gave insults this time because you apparently require them to communicate, and I'm trying to speak ape to you, as that's all you speak. You think it's a troll because you are never wrong, and anyone who disagrees with you must be a troll. I spoke clear English with facts and you ignored it. TCP is superior for all configurations where the jitter buffer exceeds RTT. Whether you have to set your jitter buffer to 2s for that doesn't matter to the statement of fact. I never said TCP was superior in any real world application. You are just moving the goalposts because you realize you are 100% wrong.

So, move them again, insult me. You can't prove me wrong, or convince anyone reading this that you are right. Everyone is entitled to their own opinion, but not their own facts.

Re:The problem with certifications

[Tyr07]

There is absolutely longer latency using TCP as the process does not proceed forward without receiving the next packet.
We're talking real world, internet, not your private lans or internal corp networks.

You did exactly what I thought you would, you tried to get exceptionally technical with ambiguous technical statements about TCP vs UDP.
You are correct that in an environment with no other traffic, UDP and TCP transmission would not introduce additional latency. I wouldn't want to be hired
by someone who couldn't handle real world concepts, you sound like like potentially a middle manager that has no idea of real world concepts.

So I'll explain it. Why would TCP have a higher latency over the internet? As you don't control the network, you also cannot ensure the quality of the networks you are routed through. This means when packets are dropped, you wait to receive another packet, as it sends the packet again In most voice conversations, this causes a delay as it buffers communications before playing them on your end. E.G receives enough packets.

When using UDP it sends the packets without waiting for an acknowledgement of the receipt of the packets, therefore when transmitting packets and some get lost, you lose a little voice quality or get the odd artifact.

Alternatively, you could use TCP, and adjust the buffers so that it doesn't wait for the person transmitting to stop talking before sending the information. If it all reaches there with no problem, no issue, if packets are dropped a long the way, you will get artifacts and jitter.

So yes, adjusting the buffers makes a difference, but can also increase latency in environments where you do not control the internet and travel fair distances network wise.

So the reality is, you're an idiot that's omitting fine details to try to make yourself look correct. This is akin to someone who preys on those less knowledgeable managers to make their co-workers look bad by spouting information that is correct in specific situations, but doesn't apply to what is actually being discussed.

It's a terrible practice, if you know anything about networking at all or are in a position where you can decide if someone is an expert or not, you would know that the scenarios I've described are accurate.

Obviously your natural defense is a strong offense, point out other peoples flaws and discussed things like trying to distract, which is entirely what you're doing. It's called projecting. If you weren't doing it, you wouldn't have put the first post you did.

Now, I don't need to convince anyone reading it's right, as they know it is, which is why they're not commenting, as they're more mature than I am and I shouldn't have responded in the first place. As for insulting you, well, yes I did. I do apologize for it, but even with that, I'm not removing it from this post either, as you're being an idiot.

Re:The problem with certifications

[AK Marc]

There is absolutely longer latency using TCP as the process does not proceed forward without receiving the next packet.

You either don't know what a window is, or are ignoring all basic knowledge to attack. Either way, you have given up trying to discuss, but are so personally offended that it's all worthless anti-technical spew.I didn't get a single sentence in, without finding such a blatant error, so I'm not bothering to read the rest of the "I'm right, you're wrong (even though all the facts are on your side and I'm a bumbling idiot)" post.

But I appreciate the effort. Well, not really, but I'm trying to make you feel better, for having made such a fool of yourself in public.

Waste

[mwvdlee]

The kid apparently has either a talent for computers or for learning.
Either way, it's a waste to train him for an MCP exam when the kid could be learning something actually valuable in the future.
I have no idea whether an MCP exam is easy or difficult, but it'll damn sure be useless by the time he is old enough to get benefits from such certification.
Having the kid get an MCP certification is about the parents' bragging rights rather than actually teaching the kid something valuable.

Re:Waste

[hawkinspeter]

While the qualification is worthless for him, the kid's interested in computers so why not take an exam? He's 5, what else would he be doing with his time?

Re:Waste

[mwvdlee]

Perhaps learn non-MS-specific, generic computer skills that'll still be useful to the kid in ten years time.

Re:Waste

[drinkypoo]

There's no reason he can't move on to bigger and better things from here. Perhaps this has whetted his appetite.

Re:Waste

[Anonymous Coward]

He's 5, what else would he be doing with his time?

Playing with other 5 year olds.

Re:Waste

[Selur]

> He's 5, what else would he be doing with his time?
Obviously, starting to build his own P0rn-Empire!!

Re:Waste

[gtall]

Or distorted his view of computer systems.

Re:Waste

[Karlt1]

Either way, it's a waste to train him for an MCP exam when the kid could be learning something actually valuable in the future.

True. I learned how to program on an Apple //e in the 6th grade in 1986 -- first Applesoft Basic and then assembly language. By the time I got out of college 10 years later and got my first job, there were no Apple //e's anywhere and no one wanted a 65C02 assembly language programmer! Spending six years learning how to program before going to college did me know good. It's like knowing the fundamentals really was a waste of time and was so not transferrable.....

Re:Waste

[deadweight]

I am a MCP. You can pass the test and not know ONE THING about actually using a computer if you have good memorization skills and get one of the test prep guides. Even if you do know what you are doing, the memorization way is better because the common sense answer is not always the correct answer. At least when I took the test you had to parse the sentences to figure out WTF they were even ASKING and then keep in mind the MicsoSpeak. I think "cost effective network" was always 10BaseT back in the day.

Re:Waste

[Kagato]

I think it was in the kid's best interest. Not that the MCP is worth more than an A+ certification these days. Be that as it may the notoriety will likely bring him opportunities. It could certainly open doors for better education and scholarships. I wouldn't be surprised if Microsoft didn't kick some freebies and money his way to capitalize on the publicity.

Re:Waste

[hawkinspeter]

I imagine that would be a bit more challenging for a 5 year old than learning MS-specific exam answers. A lot of generic computer skills require a level of abstract reasoning that wouldn't be easy for a youngster to master. I guess it also depends on what his parents know - if they are MS specialists then I guess the kid's only going to be learning MS stuff.

Re:Waste

[mr_mischief]

Perhaps he's bored by children his own age. MAybe a mix of 5-9 year olds. He needs to stay stimulated but also learn to be pleasant to people his own age.

Re:Waste

[Anonymous Coward]

He's 5, what else would he be doing with his time?

Playing with other 5 year olds.

Nonsense. If my 5 year old got an MCP, he'd be making his way in the world. I'd probably buy him an interview suite, seeing that he's only 5, but there's no reason for him not to work. Unless it's in the USA where we have child labor laws. Damned liberals!

Re:Waste

[Anonymous Coward]

Spending six years learning how to program before going to college did me know good.

A tiny bit of English Lit might have helped :-)

C'mon, man.. it was there... I could not resiiist!

Re:Waste

[Anonymous Coward]

>Spending six years learning how to program before going to college did me know good.

Sound like you should have done less programming and more work in English class.

Re: Waste

[Karlt1]

Yeah. I saw that after I posted....

Re:Waste

[Brett Buck]

Nice parenting skills. What the hell is a 5-year-old doing screwing around with computer code or "skills"? Give him a damn baseball or something.

Re:Waste

[Anonymous Coward]

It would do you no good if you still don't know the difference between those two homonyms and regularly say things will do you, "know good." It also does you no good if you did not and do not differentiate between knowing Assembly and related concepts from the specific information set for the 65C02, and then translate to the next processor family. You know, the way the rest of us did out of college.

Re:Waste

[Rinikusu]

Oh great, now you sound like my friend's dad. "Why are you wasting your time with video games when you could be planning for your future? Why don't you learn how to program spreadsheet?"

Re:Waste

[StormReaver]

Spending six years learning how to program before going to college did me know good. It's like knowing the fundamentals really was a waste of time and was so not transferrable.....

Then you didn't learn the fundamentals. Instead, you learned something very specific to a particluar product. They're not even remotely the same thing, as the fundamentals haven't changed in over 30 years.

Re:Waste

[Anonymous Coward]

He's 5, what else would he be doing with his time?

Playing with other 5 year olds.

A 5 year old playing with other 5 year olds? That would make way to much sense!

Re: Waste

[Karlt1]

Sometimes no matter how hard you try, obvious sarcasm just isn't obvious....

How much more of the fundamentals of computers can you get than assembly language?

I remember speeding a program up by accessing zero page memory because operations on zero page memory took 2 clock cycles instead of three....

STA $20

Took less clock cycles than

STA $2000

Re:Waste

[Half-pint HAL]

I'd probably buy him an interview suite,

...but he's going to have to pay for the rest of corporate headquarters himself...?

Re:Waste

[Anonymous Coward]

The 'Buzzword' of the year when I took mine was RIS (Remote Installation Services), It was all over the Win 2000 Pro. & Server exams.

Guess how many times I have used RIS for deployments: 0 (Learned ghost instead and is way more valuable and way faster for bulk deployment than RIS)

Did learns some useful stuff in the classes, but the material memorized for the exams are worthless except for the piece of paper you get. (at least until the experience years kick in)

Re:Waste

[mwvdlee]

What's a 5 year old screwing around with training for a one-in-a-million sports career?
If the kid likes doing something, you can (and I think, should) encourage him, not force him into doing what you like.

Re:Waste

[Anonymous Coward]

Perhaps learn non-MS-specific, generic computer skills that'll still be useful to the kid in ten years time.

Really? I've been in the biz since before random memory. We had to figure out where the drum storage would be after each instruction so we could fetch the next instruction without rotational delay. Now I'm on Linux writing in Python. I don't see much carryover, except that I know some algorithms that aren't in the library. Nobody does assembler anymore. Learning a new environment or language is another head trip through an arbitrary space. If the stuff actually worked as advertised, they wouldn't need us.
I have written programs in about 20 languages. The new ones are getting easier and more powerful, but a lot of features in their BNF description don't work in real life.

how to fix this:

[Anonymous Coward]

add an arm wrestling component to the test.

Wow! I'm impressed

[Anonymous Coward]

People always say Microsoft certification doesn't mean much, but it took a bright three-year-old a full two years of study to pass!

Must be more challenging than I thought.

"Certificated"?

[Anonymous Coward]

Maybe we could get that 5-year-old on /. to proofread.

Dad sacrifices sons childhood for MCP

[Macfox]

Seriously why fill a such important time in child's development with crap that will be irrelevant before he'll hit puberty. There's many more important lessons in life than this. The dad thinks his kids something special now, but in a few years when those valuable lesson and experiences can't be drawn upon....What then? This dad is a douche-bag.

Re:Dad sacrifices sons childhood for MCP

[Tyr07]

Some kids played with model airplanes and found it fun, grew up to be pilots or aerospace engineers.

This kid, assuming he enjoys computers, more power to him.

Re:Dad sacrifices sons childhood for MCP

[Macfox]

This isn't what happened here. It's one thing for a kid to play and learn. Here we have a father pursuing his sons MCP certification, for his own gratification. If this was about the kid, it wouldn't be in the news.

Re:Dad sacrifices sons childhood for MCP

[Lumpy]

If he enjoys computers, then give him a REAL OS to deal with. Why train him on the DUPLO of the computer world?

Honestly this kid needs a good linux and a good compiler. Any monkey can do IT, geniuses need to learn the inner workings and have the tools to break things or make them different. He needs to be using an OS that lets you interact with the hardware level and have access to compilers.

MCSE is the factory worker of the 21'st century. Do you want to be an assembly line worker or an engineer who designs the machines that others stand at and put caps on the bottle?

Re:Dad sacrifices sons childhood for MCP

[Anonymous Coward]

Classist

Re:Dad sacrifices sons childhood for MCP

[Anonymous Coward]

The kids has a load of machines he pisses around on for two hours a day. It's just LEGO to him. He's already above any Apple genius. Is that why you're so sore about it?

Re:Dad sacrifices sons childhood for MCP

[gewalker]

Are you telling me I could not use a compiler on Windows. Or write device drivers on windows? Granted, I have to download or maybe purchase the compiler for Windows, but there is really no big difference on the fundamental concepts. There are many differences in the details.

E.g., on Windows you don't fork a process. This prevents you from doing some multi-processing things in Windows as easily as the natural way under Unix. But if you are learning the fundamentals, you can happily learn what you are doing with either environment. You can still use multiple processes on Windows, you just don't can't inherit the parent process state so it is more complicated. Way back when, fork/exec was unnecessary overhead in Unix since most of the time all you wanted to do was spawn an unrelated subprocess. Unix fixed this extra overhead long ago, but Windows never added fork to its API -- so Advantage Unix. Potentially a big advantage.

There are also built-in WIndows APIs that have no equivalent on the Unix. -- Advantage Windows.

Price, freedom, source-code, etc. -- Linux

Apps -- Maybe Window, maybe Linux depends which apps aee important to you.

Games -- Windows

In fact, you are better off learning both environments even if you have a strong preference for one of them. It opens your mind to different ways of thinking.

Re:Dad sacrifices sons childhood for MCP

[houghi]

Some kids played with model airplanes and found it fun, grew up to be pilots or aerospace engineers.

Some grew up to be hairdressers or politicians or whatever.
Causation is not correlation.

This kid, assuming he enjoys computers, more power to him.

This. If he likes doing it, let him be, regardsless if he still likes it when he is 10, 15 or 20 and rather hangs out in the garage playing lousy music with his friends and becomes a plumber or a banker.

Re:Dad sacrifices sons childhood for MCP

[Anonymous Coward]

Causation is not correlation.

It most certainly is. I think you meant to say: "Correlation does not imply causation".

Re:Dad sacrifices sons childhood for MCP

[Anonymous Coward]

If he enjoys computers, then give him a REAL OS to deal with. Why train him on the DUPLO of the computer world?

The neckbeard-Linux-sysadmin "get a real OS" hubris is pretty played out at this point. Windows is fine. Linux is fine. The former probably moreso if one chose to pursue an IT career at a non-IT company.

Re:Dad sacrifices sons childhood for MCP

[MattBD]

Exactly. Not only does Linux let you get your hands dirty in a way that Windows simply doesn't, but the fundamentals of Unix haven't actually changed all that much down the years. If the kid learns things like bash, sed and awk now, those are a damn sight more likely to still be useful than any Microsoft certification. Having a kid build his own Linux From Scratch install would be much more impressive than him getting this certification.

"Explaining"

[Kvathe]

"The hardest challenge was explaining the language of the test to a five-year-old." Makes you wonder how thorough this 'explaining' was. I may be wrong, but standard applicants probably don't get much in the way of explanation, and understanding the question is a very important part of any problem solving.

MCP is the entry exam

[Anonymous Coward]

The MCP cert is a way to prove you are more knowledgeable about the Windows client computer than your grandmother that calls you for help. Like the A+ exam, I found it remedial. MCPs are not MSCEs. I enjoy how the people who insult MCSEs either 1) do not have it, or 2) have a different certification (NCE). And yes, your anecdotal stories are very cute.

Re:MCP is the entry exam

[Anonymous Coward]

MCSE's are complete and utter MORONS. I have no certs at all, just 20 years experience and real fucking degrees in EE and CS.

Not one person on this planet holds a MCSE in high regard.

Re: MCP is the entry exam

[Anonymous Coward]

I don't hold an MCSE because I never really saw the point. That being said when they revamped the program a few years ago it got much harder than it used to be. The questions tactually test fundamental understandng vs rote memorization now. It's a certification with some merit and its pretty arrogant to insult everyone who holds it just because you don't. It was a bit easier to break into the industry 20 years ago than today...

Re:MCP is the entry exam

[AK Marc]

EE and CS have *nothing* to do with IT. At least an MCSE shows an interest in IT.

No big deal.

[Qbertino]

I think the mold on the left yogurt in my fridge is an MCSE. ... Yeah, he was bored one afternoon.

Seriously though, if my kid were a computer prodigy, the last thing I would teach it is something proprietary with such a short half-life as MCP. Basic knowledge of a programming language and TCP/IP would've been much better for this kid at that age. What a waste of talent. ... Put him on the kernel team and Linus accept a commit by him - *that* would be news. :-)

I hope this wasn't some nutty dad driving his kid to do something so he could feel great about himself as a dad.

But maybe the kid is happy and loves his dad and dad loves him back. That's the most important think at that age - MCP or not.

My 2 cents.

Re:No big deal.

[Anonymous Coward]

And now my 2 Cents

I keep going back to the last sentence of this article " But he seemed to pick it up and has a very good memory." Kids do have good memories and I think passing this test was not so much about talent but of pure memorization (that is what most Pakistani and Indian kids are good at not so much about applying that knowledge.) This is more like running for the bright kid without learning walking. Pak and Ind tend to exaggerate things a bit. When you say "I hope this wasn't some nutty dad driving his kid to do something so he could feel great about himself as a dad." that might be the case somewhat. Why do i say that because i am one of them :)

Now i will also give Kudos tot he kid for this achievement because it is a big feat for a growing mind. Hopefully he turns out to be a good Techie and not some best cost engineering center junkie.

Re:No big deal.

[flopsquad]

Put him on the kernel team and Linus accept a commit by him - *that* would be news. :-)

This kid could be ready to do just that by the time he's seven. Problem is, he won't be ready to read Linus' comments until he's seventeen.

microsoft OS? really?

[Anonymous Coward]

At least let him learn on a real OS: unix.

New Age of Age Discrimination begins!

[Anonymous Coward]

Great, now by 18 we are going to be too old and too expensive to do the tech jobs!

Old News

[Anonymous Coward]

Saw this several days ago...

If only most American 5-year-olds could do it...

[damn_registrars]

Sure, the MCP test isn't that great. But a 5 year old kid passing them does show a drive for learning that most American 5-year-old kids don't have (for various reasons). Most American kids that age wouldn't even be able to read the questions, let alone answer them. Unless his dad was able to read the questions to him and then put his answers in for him, the kid has exhibited a superior reading and writing ability in comparison to most.

Hopefully he doesn't end up becoming a desk jockey troubleshooting windows PCs. If he keeps up this desire to learn he should be able to go much, much, further. I wish him well (and I wish him a better OS as well!).

It doesn't mention which exam

[Anonymous Coward]

Any Microsoft exam you pass gives you MCP.

Why is everyone surprised?

[__aaclcg7560]

When I was a video game tester at Accolade/Infogrames/Atari (same company, different owners, multiple personality disorder), we kept a five-year-old in the inventory closet in case we ever needed a console button smasher. After you turn 30-years-old, you're likely to smash the console than smash the buttons.

Re:Why is everyone surprised?

[Anonymous Coward]

At the Soviet 30s, the console buttons smash the fingers and wrists.

If only most American 5-year-olds could do it...

[Anonymous Coward]

Sure, the MCP test isn't that great. But a 5 year old kid passing them does show a drive for learning that most American 5-year-old kids don't have (for various reasons). Most American kids that age wouldn't even be able to read the questions, let alone answer them. Unless his dad was able to read the questions to him and then put his answers in for him, the kid has exhibited a superior reading and writing ability in comparison to most.

Hopefully he doesn't end up becoming a desk jockey troubleshooting windows PCs. If he keeps up this desire to learn he should be able to go much, much, further. I wish him well (and I wish him a better OS as well!).

By the time this child is old enough to enter the full-time paid workforce those skills will be worthless. Then again with the push towards cloud services a point-and-click jockey is probably the only technical jobs besides infrastructure engineering and devops which will require 100 years experience in twelve dozen products and technical skills.

Only one exam

[dyeazel]

IIRC, you are an MCP for passing ONE Microsoft test. You need to pass 4-5 (can't remember the exact number) to be an MCSE or MCSP. So the real headline is, "5 year old with photographic memory passes one test with a lot of coaching by his father."

Re:Only one exam

[Anonymous Coward]

And his website is a simple template from a CMS, the kid probably doesn't even know how to properly mark up content, let alone style it.

That explains the Windows 8 interface

[Anonymous Coward]

Well if they're going for 5 year old MCPs that explains the Windows 8 interface nicely.

I would be much more impressed...

[Torp]

To find out who's the youngest linux kernel developer who (to have some objective criteria) had a patch accepted into the mainstream line.
Or you know, anything else besides something based on multiple choice and memorization.

Re:I would be much more impressed...

[jittles]

To find out who's the youngest linux kernel developer who (to have some objective criteria) had a patch accepted into the mainstream line. Or you know, anything else besides something based on multiple choice and memorization.

That's not a good guide at all. When I was 23, I submitted a patch to kernel.org for an issue with the USB HID driver. Some idiot had gone to sleep with a spinlock. I submitted that patch as per all the guidelines on kernel.org at the time. The bug (which would grind your computer to a halt with the correct hardware attached), remained unpatched for 6 kernel versions before some mainstream kernel dev took my exact same patch and submitted it under his own name. If you don't already have a name, or if you don't have the weight of a giant corporation behind you, you'll never get your patch into the kernel, even if its a simple one line fix that is obvious.

Re:I would be much more impressed...

[Anonymous Coward]

"some mainstream kernel dev"
"submitted it under his own name"
" If you don't already have a name, or if you don't have the weight of a giant corporation behind you"

So, does he have a name or not? Who was it?

Re:I would be much more impressed...

[quantaman]

To find out who's the youngest linux kernel developer who (to have some objective criteria) had a patch accepted into the mainstream line.
Or you know, anything else besides something based on multiple choice and memorization.

Not sure when his first patch was accepted but Marcelo Tosatti became the maintainer of the 2.4 stable branch when he was 18.

Re:I would be much more impressed...

[Anonymous Coward]

Lennart Poettering.

Are you more skilled than a five-year old?

[Anonymous Coward]

I have always thought these tests were more a barrier to entry - you have to pay for the test to get work in the field - than anything else. Now this is proof. A test so easy a five-year old can pass it? Makes me glad I didn't get into the whole Microsoft test scam.

A++ & MCP

[slackoon]

I used to work tier 2 tech support for RoadRunner. People, usually from Florida, would call in complain their internet wasn't working. I would always get people trying to let me know just how smart they were.

Customer: "Hi, I'm A++ and MCP certified and I have a certificate from Devry so I know my stuff but I can't get my internet working"

Me: "What makes you say it isn't working Sir, what exactly is going on?"

Customer: "I can't load any websites except ones I've seen before, I tried restarting but it's just doing the same thing"

Me: "...type ipconfig, what do you see?"

Customer: "...static IP..."

Me: "Sir, do you have a kid who uses your computer for gaming? You have a static IP, that's the issue. follow these instructions and it will work..."

Customer: "No, I certainly don't have a static IP. I looked for that. The issue must be on your end"

Me: *FACEPALM!!!!!!!

comclusion: MCP MEANS NOTHING!!!!!!

Re:A++ & MCP

[ArcadeMan]

MCP only means one thing: Master Control Program.

Re:A++ & MCP

[Anonymous Coward]

Can confirm. Road Runner ISP was utter shit in FL. Weekly outages, particularly on Wednesdays for some reason. They lost a huge amount of business when FiOS was rolled out in our county, and the remaining RR users are the elderly that still have AOL email adressess, and the poor. They went from rock solid to utter garbage in the space of about three years.

Re:A++ & MCP

[morgauxo]

Hah! I took pretty much those same calls working for another internet provider. Countless!

Re:A++ & MCP

[BonThomme]

End Of Line.

Re:A++ & MCP

[__aaclcg7560]

I have an A+ and Network+ certifications when I called Comcast to complain about their technician disconnecting my Internet service at the box. The service rep reassured me that my Internet service was working fine despite everything on my end failing. After going a month without Internet service and refusing to pay the current bill, Comcast sent a technician out who discovered that the last technician installed a bypass filter backwards inside the box.

Dump is valid!

[Anonymous Coward]

Passed today with 869. Dump is valid!

Child Abuse!

[morgauxo]

Ugh! Windows indoctrination at such a young age!

The intrinsically racist truth is that

[Anonymous Coward]

he does not have to worry about inane matters such as existentialism and individuality. These matters tie up the minds of those whose cultures are base on the individual.

Achievement at the expense of individuality. It is the very definition of what it means to be Asian.

This child abuse!

[Anonymous Coward]

People, teach you kids Linux or Cisco or Java - anything but MicroShit!

Reminds me of Ham Radio

[EmagGeek]

I remember back after 9/11 when everyone and his uncle got their 5 year old to memorize the FCC question pool answers and get their ham radio license...

see?

[BonThomme]

Hey, this code.org thingy is really working!

Old Windows Installer Messages

[Dareth]

if you ever find yourself installing an old version of Windows and are not paying attention, you can almost get excited by the list of "new features" in the old version of Windows. Their OS may be so so, but their propaganda has always been top notch.

Groucho on Microsoft Testing

[Galaga88]

Why, this Microsoft certification test is so simple a five year old child could pass it.

Run out and find me a five year old child. I can't make head nor tail out of it.

Still won't be able to get a job...

[Andrewkov]

H/R Drone: Do you have 5 years experience in the field?

Kid: I am 5!

Re:Still won't be able to get a job...

[Herkum01]

H/R Drone: That would be a "Yes" then...

Note to College Graduates

[Anonymous Coward]

A "computing degree" does not trump passion, tenacity, and experience sans degree.

Re:Note to College Graduates

[HeckRuler]

Correct, and passion, tenacity, and/or experience sans degree don't trump a college degree.

I don't care how passionate you are if you can't learn anything.
I don't care how long you banged your head against that simple problem.
I know a couple of self-taught programmers who are simply incompatible with any other coder or codebase. They've got their one project at their company, and no-one else can touch it. And frankly they shouldn't touch anything else.

It's as if there is no trump, and there is no silver bullet guaranteed way to make a competent IT worker. Or a developer. Or an engineer. Because if there was a simple guaranteed path to becoming one of those, people wouldn't be bothering with anything else.

Passion is great if you capitalize on it. Tenacity is required to get anything done, but you have to know how to overcome problems. Most experience is good experience. Some experience just teaches you bad habits. Some experience is the same experience over and over again.

And a degree doesn't mean you're all that hot either. It just means you can pass a specific sort of hurdle. A big hurdle, sure, depending on the school and the degree. But there are plenty of grads who can't code.

Re:If only most American 5-year-olds could do it..

[Anonymous Coward]

Sure, the MCP test isn't that great. But a 5 year old kid passing them does show a drive for learning that most American 5-year-old kids don't have

Pure B.S. All this proves is that this father is selfish. This is the equivalent of 5 year old beauty pageants and nothing more.

no mention of linux?

[Anonymous Coward]

More than 100 comments and no snarky remark about how he would have been better off learning Linux?
what a shame.... I'm going back to usenet to talk about FreeBSD's imminently approaching death

Re:no mention of linux?

[rubycodez]

I was sad the child wasn't exposed to one of many real operating systems rather than redmond's glorified program loader.

But

[pak9rabid]

But can he run Linux?

MS certs are a joke

[Anonymous Coward]

You don't pass them by knowing your material (unless you're an idiot savant who has memorized every meaningless facet of the product) or by taking classes. You pass by memorizing the bizarre questions they ask and the answers they expect to see. I was forced to take a few MS exams for work, and I passed them all on the first go-around, but I don't know jack about Windows.

The dad did it...

[nowsharing]

Before accepting that a 5 year old not only passed the exam, but could read and comprehend at a fully adult level, I have another hypothesis: the dad did it... He obviously made the kid's website, so why would it surprise anyone if he "helped" the kid through the test. It's the same way that kids used to win slot-racing competitions.

Re:The dad did it...

[Anonymous Coward]

The article mentions invigilators....

how much help is OK?

[Anonymous Coward]

The hardest challenge was explaining the language of the test to a five-year-old. But he seemed to pick it up and has a very good memory.

While a good exam isn't supposed to test the student's ability to understand the language, it is supposed to test the student's ability to understand the underlying concepts. This means that the exam should, in part, test the student's ability to read a question, identify the key concepts, and figure out what is an appropriate answer based on those key concepts. If you receive help reading the question by converting concepts into something easier to understand, then that would seem to undermine the test result and therefore the certification achievement. Not to the point where it's completely void, but it may not be completely accurate to the kid's skill level

Now, i don't know if the parent acted as a dictionary or actually simplified the language, but the sentence in the summary does seem concerning.

Groucho

[slapout]

"Why it's so simple, a five year-old child could understand it. Now go out and get me a five year old child"

TEST

[westlake]

The MailOnlne described the test as "Supporting Windows 8.1." Schoolboy becomes world's youngest qualified computer specialist after passing Microsoft Windows exam aged just FIVE

Yes, this is technician level. Doesn't claim to be anything else.

But IF the range and depth of the exam is equivalent to the MS Course of the same name, it is far from the trivial achievement that the geek with five to ten years of practical experience likes to pretend. Course 20688D: Supporting Windows 8.1

There is an entry-level technician grade exam Configuring Windows 8.1

which still implies an understanding of concepts and methods that will be quite alien to most five year olds, and every now and again trips up an adult who should know better,

contact family services

[Anonymous Coward]

This is child abuse. The guy should be reported.

wow!

[BenLutgens]

If this kid has a memory like this he should be doing more important things than learning how to be a windows admin.

Says a lot about the MCP Exams

[Kryptonut]

I remember earlier in my career, looking for work with a tertiary qualification and 4 years experience in the IT workforce under my belt (I worked in IT before, during and after tertiary study) and being turned down by potential employers because I wasn't "Microsoft Certified"

Nevermind the fact that at least 2 of the papers I studied toward that tertiary qualification revolved around configuring and supporting Microsoft networks and I'd been working with Microsoft technologies full time for about 2-3 years prior.

I later just got the damn certification anyway, because I needed the job prospects that came with it. I learnt very little by doing it.

Re:Dagnabbit!

[Marginal Coward]

Corollary: Base sixteen is just like base ten...if you've got four extra fingers.

Re:Dagnabbit!

[Marginal Coward]

FTFM: Make that six.

3 R's

[Anonymous Coward]

Reading W(R)iting and A(R)ithmatic. So the kid has these skills does he? or did dad just read them off and help him out a bit?

Ummm. No. I do not believe it.

Why all the hate?

[Anonymous Coward]

Look at all these haters. The kid is 5 years old..that's the whole point of the story. How about being positive and encouraging towards the story. Realistically only a very small percentage of you viewers can probably diss the Microsoft exams due to your high accreditation while the most of you I bet are nowhere near an elite level in this industry to be shaming this story. Your telling me most of you are better than the next MCP employee? I doubt it so shut the hell up.

Re:Why all the hate?

[rubycodez]

Why yes, I am and have been a admin of real operating systems (posix, midrange and mainframe); Microsoft Certified "pros" have little useful skills or understanding, and I've seen more time wasted on a single Windows server than dozens of servers running a real OS.

Bud bud mbud, jolly goodness old boy

[Anonymous Coward]

"Too much computing at this age can cause a negative effect, but in Ayan's case he has cached this opportunity."

That doesn't make any sense at all. Perhaps he should spend some time learning English instead.

Probably Eidetic memory

[Anonymous Coward]

The next Sheldon Cooper.

Re:Dagnabbit!

[AK Marc]

And here I was picturing you counting with your feet and your 7 fingers per hand.

Re:Dagnabbit!

[Zanadou]

And, I can count to 21 if I take all my clothes off!

Re:Dagnabbit!

[AK Marc]

Yes, you have a belly button. Good for you.

Certificated?

[Anonymous Coward]

Is that even a word?

Im skeptical is it a joke ?

[Anonymous Coward]

Can you image 5 years old kid on alone in a chair at prometric/Pearson VUE center taking a multiple choice test exam against the clock alone ?

Ok Now I'll put the game controller down Mom

[Anonymous Coward]

Jeez, way to mess up the CURVE kid! NOW I realize I ain't did !@#$% with my life.

What's the point?

[Anonymous Coward]

What good is an MCP certification to a 5-year old? A pointless exercise that makes a mockery of the whole certification system.
Of course it doesn't help that company HR departments post level I help desk positions as requiring MCSE certifications. :-P

Seriously?

[Stubbyfingers]

Do us 50YO Americans stand a chance?

Which test was it?

[Anonymous Coward]

I swear I read this same story about 5 years ago...

Back when I took my first MS test, you were called an MCP if you passed one. There's a bunch of tests, each of which make you a MCP. Which one did he take?

I took a couple MS tests recently to satisfy a job requirement. The books from MS were crap. Totally unhelpful. I passed based on experience, which I wouldn't have if not for an employer that was willing to take a chance on someone with no real experience, a degree in Network Administration, and a couple MCPs for XP and Server 2003.