High-tech Cheating on Exams

India is cracking down on people who use technology to cheat on exams:

Candidates have been told to wear light clothes with half-sleeves, and shirts that do not have big buttons.

They cannot wear earrings and carry calculators, pens, handbags and wallets.

Shoes have also been discarded in favour of open slippers.

In India students cheating in exams have been often found concealing Bluetooth devices and mobile SIM cards that have been stitched to their shirts.

I haven’t heard much about this sort of thing in the US or Europe, but I assume it’s happening there too.

Posted on July 10, 2015 at 12:44 PM57 Comments


Bruce July 10, 2015 1:32 PM

A while back, I remember this guy, Bruce Schneier, had a comment. He said that the real problem is that society values a degree more than it values the learning.

Things haven’t changed.

Spaceman Spiff July 10, 2015 1:59 PM

My father was a professor of physics. All of his exams were “open book”. If you didn’t know the subject, the book wasn’t going to help…

winter July 10, 2015 2:02 PM

Some school systems are more prone to this kind of cheating than others.

If testing is done for root learning and retrieving factoids, preverably on multiple choice questions, then a simple connection to a (conspirator with a) search machine can do the trick. India might be in that camp.

If you get essay questions or (math) derivations, you need a much higher bandwidth and probably an informant who actually can solve the questions.

Spaceman Spiff July 10, 2015 2:03 PM

Re. my father, the physics professor. Once he caught a student cheating because he kept looking at his slide rule (before electronic calculators). The thing was, was that the exam didn’t require any calculations. The student had his crib sheets attached to the slide!

albert July 10, 2015 2:31 PM


It sounds like your dad was a real teacher. Droves of new PhDs are churned out every year. Most know their field of study (often laughably narrow), but don’t know shit about anything else. Education is about rote memorization. Tests are created for computerized grading. That Watson could do so well in Jeopardy should tell us something.
Maybe this is what the system wants; legions of automatons unable to think for themselves. Unable to question anything. Lacking any creativity or original thought. It’s going to be interesting to see what kind of people the Internet Generation grow up to be.

Right now, it’s not looking good…

effres July 10, 2015 2:55 PM

@albert, Don’t take this as a personal attack, but I have been a teacher for several decades and here in the USA rota learning has been off the agenda since the 1960s.

A Nonny Bunny July 10, 2015 2:59 PM


A while back, I remember this guy, Bruce Schneier, had a comment. He said that the real problem is that society values a degree more than it values the learning.

I don’t think that’s quite true. Society values a degree as proxy/signal for learning. It’s just much easier to establish whether someone has a degree than that someone has the learning for which it’s a signal.

As for these cheaters. Well, I’m a bit ambiguous about it. For example, for a programmer knowing how/where to find solutions on the internet is actually a very useful and productive skill. And getting other people to do work for you is a very useful managerial skill.
So using technology or getting answers from other people isn’t necessarily a bad thing; as long as you can do that at your job (at no extra cost to the company, like foisting your work on your coworkers). My main worry is that these cheaters probably only aim for whatever metric signifies success, rather than actually getting a job done properly.

fooc July 10, 2015 3:06 PM

Exam cheating is more prevalent in India because of the sheer competitive pressures involved, which no one in a Western country/system can even understand (you have to live there to really “get” it).

So Harvard has, what, a 5% acceptance rate ? Maybe even 1%. And a few hundred thousand apply to Harvard every year ?

Indian IIT’s have like a 0.01% acceptance rate and maybe 10 million apply every year. There are many fewer quality universities in India (developing country and what not), and many many more people, whose only way out is a good-college (remember, a lot fewer) degree…

J. Peterson July 10, 2015 3:10 PM

This reminds me a little of a (probably apocryphal) story I heard about an illustration class: The final exam was to forge a postal stamp on an envelope. If the envelope successfully went through the mail, you passed the exam.

fooc July 10, 2015 3:12 PM

..as a quick follow on point..

This exam only orientation is massively bad practice in general. If the whole point of all your education is to be at the top end of exam-taking, because that’s the path to a good college and then to a good job..then thought, classroom discussion, etc., all become secondary.

From what I’ve read, Asian countries like Japan and China have a similar dynamic.

Wael July 10, 2015 3:25 PM


What good is a SIM card without a device to put it in?

Virtual moral support. Or to show off. But for sure, the subject they were (attempting to) cheating on was mobile communications – lol

Sam Agnew July 10, 2015 4:21 PM

You know, I really like the way Redhat does exams. What you always seem to get is a VM. The questions on the exam describe desired states that the VM should have. At completion, your grade is based upon how many of the required states your VM satisfies. No one cares how you acheive the end goals. No one cares if you read all the man pages (if you don’t know your stuff it won’t help since the exam is timed). No one cares if you have a different way of doing it than what was taught. If you blow up your VM or make it unbootable you fail. Just like life. I wish all tests were like that.

Bob S. July 10, 2015 4:25 PM

I didn’t cheat in college, but others sure did. I was afraid for one thing, but the other part was, why pay for a college degree based on fraud?

I think the big issue is not that cheating occurred or occurs, there are always a few cheats in any game.

The real problem is when cheating appears to have become widespread…ubiquitous…found everywhere….even organized, maybe encouraged.

I suppose it’s embarrassing to admit your bright stars are cheats, but they doing the right thing trying to stop it. Severe consequences seem in order

yo July 10, 2015 4:56 PM

What the point of studying, they may say.
Google’s wearable glass, Apple’s Smart watch… they provide the
“answer” the student want, but they never teach the way of thinking.

Universities teach you something by force you to “remember” it.
In 2020, people won’t have to “remember” it. Their wearable device
remember eternally for you. All you have to do is think, which is still
not reproduce-able in AI study.

ianf July 10, 2015 5:53 PM

@J. Peterson

The illustrator-exam forged mail stamp story is probably apocryphal, but I have seen a couple of miniature watercolor “stamps” that passed the postal test – they were handpainted during WWII by future master stamp engraver Czeslaw Slania, ultimately author of more than a 1000 stamps done for several national postal issuing authorities (mainly Poland, then Sweden)… a legend among philatelists.

albert July 10, 2015 6:13 PM

I’m not offended. Maybe you can explain why kids with college degrees just don’t seem very well informed. Maybe the kids still have the same potential; maybe it’s the gradual loss of good teachers. Teaching is an art. Some folks have it and some don’t. You don’t give music scholarships to tone-deaf students. Just because a kid wants a certain career, doesn’t mean he’s got the tools for it. If kids can’t construct a meaningful sentence, what kind of journalist or lawyer* will they be? That old saw about ‘those who can do, do; those who can’t do, teach’ is exactly backwards. I’m not saying that because you were a teacher, either. A good teacher can have a life-long influence on a kid.

* you should read some of the drivel I’ve seen from actual lawyers! It’s mind-boggling! (wondering how they got their degrees, it may, in fact, have something to do the the subject of this article:)

Bruce wrote about online university exams and remote proctoring on this blog. Cheating has come a long way since crib sheets and writing on your arms. (Ball-point pens were a marvelous** invention)

** not in spel-chek, I’m past my expiration date.

Justin July 10, 2015 6:24 PM

Cheating. Yeah, it goes both ways. Make yourself look good and make others look bad. The only class I remember cheating in was typing class in middle school. I won’t get into that here, but I also remember in that class a kid had hacked one of the computers. He opened up the file C:\AUTOEXEC.BAT, edited it in WordPerfect, and saved it as a WordPerfect document. I rebooted his computer for him and fixed it so he didn’t get in trouble, (because the computers had a menu system installed and we weren’t supposed to be able to access the DOS prompt.)

I felt guilty about cheating, and for some odd reason I ended up learning to type on a Dvorak keyboard with at least equal ease as the Qwerty keyboard. My typing teacher had actually said that the world record for typing speed was set on a Dvorak keyboard.

I’ve only ever seen one other person type on a Dvorak keyboard, and he was some guy from the Army I happened to be taking a college class with.

Back to cheating. It’s bad. And it’s rampant out there. Even in the U.S. All kinds of excuses and justifications for it, too.

zj July 10, 2015 6:28 PM

Once I was on the hiring panel for an entry level computing job. One of the applicants was really good. He had good marks from university, and he could answer the questions we asked, he was personable and interesting and eager and seemingly a very good chance.

We didn’t find the problem until we hired him.

The problem was that we hired the skill he presented: excellent exam passing. In an exam you have all the information you need to answer the question and little or no information that isn’t relevant especially in these days of multiple choice and short answer. In real life, especially front line computing, much less so.

He was basically a pattern matching algorithm. As soon as whatever he saw matched a pattern in his database he performed the matching procedure. Which in a helpdesk and basic troubleshooting role is absolute disaster.

As long as you gave him precise instructions you could rely on him producing what you asked for. Other than that he needed close supervision.

We did change the interview method for entry level jobs after that, as clearly the mistake was focusing only on knowledge not procedure. Easy in hindsight… The modern school system selects for people like our interviewee and traditional knowledge review hiring does too.

Problem is that procedure and way of thinking interviews take a long time and a chunk of effort and are hard to keep consistent and fair. Yes you can start each one with the basic scenario but after that it’s difficult and very subject to unconscious bias.

Impossibly Stupid July 10, 2015 9:12 PM

@A Nonny Bunny
“It’s just much easier to establish whether someone has a degree than that someone has the learning for which it’s a signal.”

Not in my experience. It might take days or weeks to verify the education credentials of someone, but a casual 10 minute conversation will often be enough to find out if they’re just a poser resting on those claimed laurels.

tyr July 10, 2015 10:40 PM


I love the moderne concept of an education as something
which has an end, a piece of paper for the wall and you
don’t have to study anything anymore. The idea you can
discard your memory because Google knows everything is
another wonderful idea. Once they discarded the connect
between sounds and written language here I’ve noticed
no one can spell anymore and a spellchecker only can
catch incorrectly spelled words. If you want to see
some horror read a few college student papers. The
horror part is that some one taught them to be that
way. You might also try a little Derrida to see what
the avante garde of post modernism are up to.

Terence McKenna used to make good money selling his
lectures as the most far out fringey stuff in the new
age movement and he said it was just a decent liberal
college education of the 1960s. Then the schools were
turned into trade schools for MBAs.

We hired a guy once and I showed him how a data analyzer
could be hooked up and told him he would have to play
with it to become proficient. It was the latest state of
the art instrument. He said he’d wait until they were
offering a college class in how to use it. I explained
very unkindly that accademe had a forty year time lag
behind industry and he might try using his own head for
something besides a hat rest.

An education that deliberately cripples the mind is a
lot worse than just letting them figure things out for
themselves. A little class in logic will convince you
that the person who cheats is really cheating themself.

I like the idea of cyborgs if they add the mechanisms
to being human not as a substitute for parts we already

rgaff July 10, 2015 11:01 PM

@Impossibly Stupid

Yeah, but it takes someone who knows what they’re talking about to evaluate via a conversation (the horrors!! talking??) 🙂 whereas any old paper pusher can check the paper.


Regarding the spelling: all it takes is reading to master spelling… and I mean reading properly spelled stuff, of course. College age kids who can’t spell is an indication they don’t read, just watch tv and play video games.

Regarding the crippling of the mind: I agree, it’s a big failing of a lot of academics, to not teach people how to figure things out on their own… but then again… best way to learn that is to actually figure that out on your own. People just need to be curious about the world around them and feed their brains, that’s all.

Destination Infinity July 10, 2015 11:13 PM

Rote-learning, cheating, degree-orientation, etc. are all unconscious tactics applied to prevent people from learning more than they ought to? Maybe some societies believe education, beyond a level, is/can be a double edge sword? I am not trying to justify malpractices, but since it is happening seemingly in a large-scale, there should be some basis?

Historically, education/knowledge/research has been confined to a small group of people in India and even they are taught beyond a point only if the Guru deems them to be ready/mature enough to handle the responsibility. Schools focused more on teaching self-discipline initially (they still do) instead of teaching more and more knowledge, because knowledge in the wrong hands can be lethal. This blog wouldn’t even exist if that wasn’t the case.

Destination Infinity

mfp July 11, 2015 5:39 AM

It is. About 10 years ago was on all the italian IT magazines, about US and UK. Expensive services offered by specialised companies. Watch John Oliver on ‘Standardised Tests’ (YouTube, an HBO show called ‘Last Week Tonight’).
And I’ve personally seen it in AU (enrolled in that IT school): electronic locks for every room, metal curtains at the windows, high end routers in every room, every room in its isolated collision domain, complex ACLs to get out, DPI on exit from the school, etc.
In Italy (up to 2002) was not so techy; some teachers in the IT department were barely able to use the email… but yes… some were paranoid too. One asked to us to re-authenticate (ie: photocopy of ID and uni enrollment letter, at the entrance of the test room), then choosed our seat one by one, etc. But we already started to import that testing method thanks to private companies; I did them with Sylvan Prometrics (around year 2000) to certificate with Cisco and Microsoft.
What they have in common is that they all use standardised tests (ie: the teacher don’t need/can/want know the students, so need to use something else to evaluate).

The point is that whoever uses ‘standardised tests’ MUST do security enforcement because are easier to cheat, and because there are some details to meet in the composition of the test to make it reliable (ex: special questions to calibrate the test; if at the end of the day 90% didn’t answer properly, it means that the test itself is wrong or the teacher didn’t work properly). And whatever country got cybernetics from the wrong side… ie: applying rationality to education, among other things… is destroying the most brilliant ones.

One of those tests in that AU school once was asking to me: “Which of these machines can be a proxy?”, “A. Router, B. Switch, C. Server”. I barred them all with a single line, and wrote “D. My digital clock”. I’ve been lucky that the teacher start to laugh, and ask me for disambiguation, my answer: “Well, my digital clock have a cpu, memory, and a light interface; maybe can proxy 1 bit a time only, and takes too long for rx-tx by light … it’s a slow transceiver… to be useful… but it can be a proxy”. And he gave me the answer for good. But not all the teachers can and want to give you extra time, extra care, extra love.
In that school there was wonderful people castrated by policies and regulations; I applied for Recognition of Prior Learning, and followed the procedures to get it … up to the general manager. The teachers couldn’t stop me (by law) because I had evidences of prior learning; but that asshole gave for granted that between my MS cert (around 2000) and 2011 microsoft updated their systems (“your cert expired!”), and, that AU would continue to use MS crap (“look at the context you’re in!”). I even didn’t try to explain to him that the childish MS lullaby about Forests and Domains, inhabited by Users, Algorithm and B.O.F.H.s, was just a subset of LDAP and x509 in use since ever in the Unix world. Just a bad rework like Tron3D and Total Recall without Schwarzy. So after 9 months I escaped from that cage.

Whoever is doing this kind of tests, and enforcement, miss completely the point: with the pressure produced by that kind of exams, kids don’t need to study. They ends up memorizing the braindumps: the pool of questions reconstructed (in very short time, by xor based human DHT) outside of the examination room. They don’t need radios or whatever technology to cheat; they just memorize the questions, the answers, and pass the exam. Instead of studying. Instead of understanding where those questions (and answers) come from. Without asking themselves why things are made that way… you produce human hard disks instead of quantum computers (‘Beautiful Minds’, glasses up).

If you ask me for solutions… well… different story. But that crap is a problem.

Dirk Praet July 11, 2015 6:14 AM

@ Sam Agnew

You know, I really like the way Redhat does exams.

Same feeling here. RHCE is probably one of the few (IT) exams I ever sat that had real value. You first got a hands-on troubleshooting round. If you got stuck on a particular problem, you couldn’t move on to the next. If you survived that, you were given a list of stuff you had to set up and make it work according to specifications. I actually did enjoy that exam and passed it with flying colours.

It really made the typical Prometric/Vue multiple choice, point-and-click exams look totally stupid. Many people I know just crammed transcenders and left the exam centres as paper MCSE’s, officially certified but perfectly useless in any real world environment.

mfp July 11, 2015 7:58 AM

@ Dirk Praet
@ Sam Agnew

Cisco used to do the same at the advanced level. First exam just braindumps, then hands-on as well; some of them were a week long (!!!) training-selecting courses (succeed on the spot or you’re out before the cert).
But there’s not much difference between a paper test and an hands-on one; both are seeking for ‘real results in a fake env’. Of course, an hands-on one is ‘more real’ than a piece of paper but… again… you’re still killing the Schrödinger’s cat.
And that’s the issue: you (teacher) are searching for ‘real result’ in a ‘fake env’. Mmm… are you dumb or what? (leave the kids alone)
Any student can, any time, realise the fake env, and head toward the Kobayashi Maru Side Band Job, full throttle.

I can understand in life saving conditions. The army. The doctors working at the emergency department. Building trucks’ drivers and other power tools operators. And many others. But IT, lawyers, are already in the middle land; arguable. There’s no need of many IT guys in a life saving environment; there’s the need of finely crafted IT guys everywhere (laugh).

albert July 11, 2015 1:20 PM

Your point about sound (conversational language) and spelling made me think about something related: spelling and visual word recognition. I was a champion speller in grade school. I can spell most words, but sometimes I need to see them, to check the spelling. This works about 90% of the time. (it’s not a useful skill in spelling bees:) I’ve heard lots of folks mention this. Spelling doesn’t correlate with intelligence or creativity. One of the best programmers I ever met was a terrible speller, yet he could intelligently discuss any topic. Nowadays, there’s really no excuse for misspellings*. it takes all of 2 seconds to check a word, and that includes getting the definition (if it’s not in your spel-cheker, put it there!). Here’s the thing: I worked with this guy for years. We had printed dictionaries. I knew he was smart. Now, when I see common words misspelled, what am I to think? Is the writer smart, stupid, lazy, dyslexic, what? It’s disconcerting. It detracts from the ideas presented. I give folks for whom English is a second language a lot of slack, although they often write better than native English speakers. Bad grammar is even worse. Interestingly, I don’t see much of that in the tech fields. I may be that fields requiring a lot of reading improve grammar skills.
Scanning for the red squigglies isn’t proofreading (proof reading?) I read lots of fiction, and today I notice errors that I’ve never seen 20, 30, 40 years ago.
Regarding sound and words, English (the bastard son of French and Olde Germanic English) is a terrible language for spellers, because there is no logic in spelling; it must be handled on a word-by-word basis. When you have exceptions to every rule, you end up with a bunch more rules. That said, non-English speakers seem to be able to express their ideas more easily in English, than in other languages they understand at the same level.
So, that’s it for print. Well, how about conversation? Keep count of how many answers you hear begin with ‘well’, or ‘so’. Lots, right? ‘Well’ goes back many years, but ‘so’ seems relatively new. I remember from public speaking classes the expression ‘work whiskers’: um, uh, and so on. They haven’t disappeared, but are often replaced with ‘well’ and ‘so’. These are air-fillers. Their overuse reflects the apparently uncontrollable urge to eliminate silence from our environment. These word whiskers most often occur in answers to questions. They should always be avoided, especially in court, in debates, and speeches. If you need to ‘buy time’, there are more clever ways to do it. Silence ‘dead air’ was anathema to the radio business (it still is). You didn’t want folks trying to tune in, and not find your station. Don’t even get me started on movies and TV. Not only is silence MIA, but the sound levels are always the maximum allowed by the FCC, or for music, the recording medium. Truly, even folks who like ‘in your face’ entertainment, must want some respite sometimes. I used to perform with a jazz musician who was a good soloist, but, man, he played a lot of notes. I used to say “relax, you’re not getting paid by the note, and no one’s gonna leave if you stop playing for a beat or two”. Silence is a powerful tool if used properly.
A manager wanted to contract a software engineer for a particularly challenging project. The candidates had similar educational backgrounds and experience. He saved his best question for last. Here’s what he asked: “Why do you want this job?”. The answers were stock; career advancement, challenge, salary, company reputation, etc. Exasperated, he finally got to the last person and asked, “Why do you want this job?”, The answer was: “Because it’ll be fun”. He found his engineer.
Teaching Concepts Instead Of ‘Rules’
This method always worked best for me. If rules are derived from concepts, just learn the concept and the rules will follow. Rules are shortcuts that bypass the derivation process, but someday, you find a situation where your rule set doesn’t apply, then what? You forgot the concept, you never learned it, or (gasp!) were never taught it…Now you gotta step aside and let the one who understands the concept take over. That person may say: “I have no concept that covers this situation.”. Then the conceptual folks get together and say “There are no known concepts that cover this situation. We must force fit the situation into an existing concept, or find a new one.” This is where (despite the denials of scientists), the paradigm is usually preserved, and new concepts discarded, or disregarded.
There seems to be lots of problems with education today, but there are lots of problems everywhere. Technology is a two-edged sword, can help (cut through the forest) or hurt (cut down the whole forest). We can’t blame technology. Where there are problems, you will find monetization, always in the background. The Blood Sucking Parasites contribute nothing to society, only to their own propagation.
I gotta go…

* I had a teacher in grade school called Miss Spelling. She taught Civics.

Bystander July 11, 2015 2:28 PM

Re: I haven’t heard much about this sort of thing in the US or Europe, but I assume > it’s happening there too.

Don’t know if it is an urban legend or not, but there was talk about an interesting case on an university in southern Germany in the late 80s/early 90s involving the use of HP41 calculators used for communication though the IR interface.

Gerard van Vooren July 11, 2015 2:51 PM

@ Bystander

“Don’t know if it is an urban legend or not, but there was talk about an interesting case on an university in southern Germany in the late 80s/early 90s involving the use of HP41 calculators used for communication though the IR interface.”

It worked for the HP48G as well. The only problem is that the visible range was rather low and the IR was at the left side so one of the two calculators needed to be upside down…

Gerard van Vooren July 11, 2015 3:16 PM

And with upside down I mean

LOL + LOL = hihi

or in RPN


A Nonny Bunny July 11, 2015 3:27 PM

@Impossibly Stupid

Not in my experience. It might take days or weeks to verify the education credentials of someone, but a casual 10 minute conversation will often be enough to find out if they’re just a poser resting on those claimed laurels.

You can throw out any application without the right qualifications in half a minute. Most people don’t lie about their degrees on their resume, because it’s something you can easily check.
And it’s not like it takes days or weeks of continued effort to verify someone’s educational credentials; it’s asynchronous IO, not blocking, you can do other work in the meanwhile. So unless you’re in dire need of filling a vacancy ASAP it’s not likely to be an issue that you have to wait for a response.

For companies that have to field hundreds of applications a week, it’s just not doable to spend ten minutes talking to each applicant. It only makes sense to cull the herd based on their resumes, and then fact-check the remaining resumes and interview applicants that look like good hires.

Bystander July 11, 2015 3:43 PM

@Gerard van Vooren


The IR-LEDs were tweaked for better range as far as I heard…

Sounds like fun.

mb July 11, 2015 5:34 PM


A while back, I remember this guy, Bruce Schneier, had a comment. He said that the real problem is that society values a degree more than it values the learning.

It’s actually worse than that. A consequence of that is that this translates directly to assumed skills. Under perfect conditions a Master or PhD should mean that you actually know what you’re doing. But since the degree itself is financially much more valuable than the knowledge obtained in achieving it there is a high risk of cheating. And cheaters generally have no relevant knowledge. The degree creates a rather shady aura of awesomeness and that’s not always justified even if you know your shit.

And what does ‘knowing your shit’ even mean? If we look at CS: Half of the world still doesn’t have a faintest clue what we are doing even though we influence the average guy significantly more than a brain surgeon. And the other half is usually fairly certain it’s Excel.

A ‘fake’ degree doesn’t help much inside the community. It doesn’t take that long to figure out that someone is a moron. But it’s extremely useful outside. It’s the paper of awesomeness that makes you right even though a good share of the audience doesn’t know in what.

And strangely enough a substantial number of these ‘experts’ seem to be experts in being experts. They lack any expertise for being anything else.

Your ‘fake degree’ might not get you far inside the community. But it will get your very far in being the ‘to go to’ expert if some other morons have an answer that needs to be sold. And you have the papers of awesomeness to sell it. It’s pretty much your selling point.

tyr July 11, 2015 6:10 PM

IT education, There was a time when it was not a matter of
life and death importance. That’s not true anymore and it
gets worse everyday. Most of the human race wasn’t given
any input or choice in deciding to put all of societies
aggs into one basket but the mad rush to computerize all
hasn’t slowed yet. We haven’t reached the Butlerian Jihad
stage of destroying all the comps yet either. look at SCADA
for classic examples of how IT threatens us by ubiquity.

The reason people on this blog are banging on about security
is that there’s a lot that gets overlooked in the mad rush
to computerize the world. The idea that you can make this
work without improving the educational system is a sure recipe
for disaster.

Here’s a real life example. I found a problem once in a real
time control system. The programmers looked at it multiple
times and kept reading the source printouts but never could
find a coding mistake. The problem continued as an intermittent.
I finally videotaped it and sent them the tape of the machine
doing the impossible. Then they got serious about finding it.
It turned out that the linker they were using was overlaying
two code modules into the same address space when doing the
assembly code that got burned into the PROMs. That had exposed
thousands of people a day to an unnecessary hazard. No one
wants to distrust their tools but if you assume it doesn’t
add a hazard to the world when your code is broken because a
linker has no real world effects you’ve been lied to by your
comp education system.

That’s the hazard of compartmentalization in education. It’s
the hazard of specialization to too great a degree. it is also
a hazard to make assumptions about the real world based on a
limited viewpoint of consequences.

To sum it up educate as if lives depend on it because it might
be your life you save from the invisible around you.

albert July 11, 2015 7:23 PM

The linker. Whoda thunk?

The Therac 25 radiation ‘therapy’ machine killed several people due to a race condition not duplicated during testing, because operators eventually learned to enter parameters much faster than the testers could. Ironically, the earlier model used hardware interlocks, and was fine.

We used to have problems like this in the early days. Eventually, we came up with a rule: if the code looks good, the problem is elsewhere. Nothing’s more fruitless than staring at source code for days, checking every possibility, and not finding anything. I found a problem on a VAX/VMS C compiler that failed to correctly read files that were exact multiples of 512 bytes! And I didn’t have source code. Crazy!

Chris July 12, 2015 1:16 PM

This emphasis on rote memorization explains a lot about why a lot of job interviews have switched to more of a tricky essay type format. They need to check to see if you can really think on your feet. I also understand why interns would congregate whenever I began to explain basic concepts that they should have already known. They learned to memorize. I learned to think conceptually.

Dongxi July 13, 2015 3:56 AM

It’s quite a bit lower tech, but recently I heard about a popular technique here in China for “improving” your GMAT score.

Apparently the GMAT questions are changed at the beginning of every month. By the third week of the month many of the questions have already been posted online by previous test takers, so people try hard to get scheduled for a date at the end of the month and try to avoid dates at the beginning of the month.

Sam July 13, 2015 6:47 AM

“When Students cheat on exams it’s because our School System values grades more than Students value learning” – Niel deGrasse Tyson says it better and more concisely than I could.

Donald July 13, 2015 10:16 AM

I do NOT know if this story is true, but it always amused me.
First day of a class, the students all file in and sit quietly. The professor shows up and says, “Welcome to Advanced Computer Security 401. This morning I entered an ‘F’ for each and every one of you into the university’s online grade book. There will be no home work assignments, no quizzes, and no exams in this course. I do not care if you attend the lectures or not. I will not touch the grade book again until the end of the year when I will submit whatever grades are in there at that time as your end of course grade. Good luck!”

albert July 13, 2015 11:13 AM

IIRC, doctors need to be board certified. This includes oral examinations*, and ‘essay’ questions**. If anyone needs to ‘think on his feet’, it’s a doctor. BTW, med students do lots of rote memorization. In a complicated field, every teaching, learning, and testing tool is used.

Unfortunately, the ‘diploma mill’ approach (made worse by cash-strapped public schools, and even ‘for profit’ ones), cutting corners (‘cost cutting’) is often the rule. Essays need to be graded, and oral exams need expert questioners and happen in realtime. This is expensive. We can’t afford to do this for everyone, so we let computers do it. So we’re gonna have cheating, until the day when Watson can administer oral exams and grade essay questions. Sorry, you did not format your response as a question

Imagine public school classrooms with computer terminals, and minimum-wage ‘proctors’ (TSA-trained, armed in some areas). Students swept for bugs. ECM systems. Don’t need no teachers, they’re only for the folks who can afford ’em. Of course, students will be able to hack into the systems, and up their grades. They’ll get CS scholarships, and jobs in the NSA.

The final frontier is direct brain-to-machine interface. Programmable humans. Scientists are already working on this, bless their hearts.

HAH! You laugh, well….

You’ve been warned….

* Don’t say it, don’t even think about it:)
** Contrary to popular belief, this isn’t where doctors learn to write illegible prescriptions.

nycman July 13, 2015 11:30 AM

The problem: Overemphasis of paper, which itself tends to overemphasize a single test. If students lose faith in their education system and think their degree is “just a piece of paper” used to get to the next level, their singular goal will be to attain that piece of paper. In many cases it is that piece of paper that is rewarded. Not just the next educational institution or employer, but family and society in general wants to see that piece of paper. That may be fine, but if that piece of paper depends on performance on a few tests, students will quickly figure out that it’s really test performance, or more accurately, test score, that the degree represents, not “learning” or problem solving, etc. This is a failure of the educators who setup the system, not the student’s rational response to it. If attainment of a credential, or admittance to a program depended on a number of factors, there wouldn’t be such a motivation to cheat. Classroom participation, projects, papers, practical work, or just better designed tests that are given more often are some ideas.

Badly designed tests don’t help here either. I’m sure there are ways to make tests less “cheat-able”. It doesn’t help that some professors tend to reuse test questions year to year. This just measures the student’s ability to acquire a previous year’s test from a friend, or now-a-days, off the internet. Poorly written tests that measure the ability to recall trivial facts is another one.

Dumb_and_Dumber July 13, 2015 6:29 PM

If you really want to understand the tragedy that is the current U.S. educational system, you should read the book “The Underground History of American Education” by John Taylor Gatto. It may come with a bit of a built-in slant, but its certainly thought-provoking. IIRC it explains the roots of the current Indian system as well (under British colonialism, it was the prototype for the system now widely employed in western democracies).


pdf’s are available on the internet if you search around.

idea July 13, 2015 8:36 PM

@tyr (we once hired a pattern matcher)
@albert (teaching concepts instead of rules)

There’s a body of work out there called “Programmer’s Stone”. I’ve forgotten most of the ideas involved, but one part of it that struck seems pretty useful is the idea of “Mappers vs. Packers”: http://c2.com/cgi/wiki?MappersVsPackers

They posit that there’s two very different types of thinkers–or rather, a spectrum between two extremes. At one end of the spectrum is the “Packer”, who learns by memorizing collections of facts, “packing” them all into memory. They solve problems by pattern-matching to a previously memorized fact pattern. In contrast, “Mappers” have a complex mental map of abstract concepts and how they are all related to each other. Mappers spend a lot of time re-organizing their mental map to make it fit better, and to simplify recently-learned fact patterns (i.e. they abstract out the common concept and add that concept to their map, and once they are pretty sure they’ve got the concept mapped properly they can forget the details of the memorized facts.)

This paradigm can explain a lot of things about programmers in particular (and probably some other related careers):

(1) The art of programming involves a lot of “mapping”. Virtually all good programmers are “mapper”-style thinkers.
(2) Two programmers who have a conversation with each other, basically “build a map” connecting the concepts in one’s head with the concepts in the other’s–they each quickly learn the terminology that the other is using, to describe the same concept in their mental map. They can then rapidly and accurately exchange complex ideas that involve many interrelated concepts, by exploiting the existing mental maps.
(3) “Mapper”-type thinkers will often instantly see the implications of a statement because of how the concepts involved are stored in their mental map. “Packer”-type thinkers might take longer to reason out the implications, if they notice them at all. (Cognitive dissonance seems to be easier to tolerate for “Packer”-type thinkers.)
(4) When programmers talk to non-programmers, they need to realize they are probably talking to a non-mapper: someone whose brain is organized very differently from their own. So they sometimes need to explicitly spell out conclusions or consequences that they would not need to spell out for another mapper. They sometimes need to explain their reasoning very differently, because the listener won’t be able to follow reasoning that was mostly an unconscious tracing of the connections of the speaker’s mental map.
(5) “packer” thinkers are more common than “mappers”. Mostly because society discourages mapping–it looks like daydreaming, and gets scolded or slapped out of most kids at a young age.

tyr July 14, 2015 1:45 AM


Thats an interesting set of ways of thinking. I’ve also
found that there is a lot of bias in understanding how
we think. This was never mentioned anywhere in academe
until I started reading about mathematicians. You feed
a problem into your consciousness and then go off to
do other things, within a reasonable length of time an
answer will magically appear. No one every said this
is the way a brain does things in school. They also
did not say that the answer you got back might be wrong.

Instead it was assumed you had a magic consciousness
that could figure things out by rote repetition if you
ground away at it long enough. Gatto is quite a guy
I first ran into him on-line in a fan group of HPL,
then learned he was an innovative teacher with good
opinions and a real grasp of what you should do to
help students become useful to themselves and others.

The best teachers I ever had were a working scientist
and a math teacher who was a student when he was not
being a teacher. The math guy would have you explain
your work and at the point of your mistake he would
say your mistake is right there. Then all you had to
do was find it. The same applies to programs, usually
most of it is right and all you have to do is find the
one part where you went wrong. Divide and conquer is
a rational strategy when you get into complexity.
That keeps you isolating the problem in steps down to
a solution. It also helps to check the easy and obvious
first before you dive into some arcane sphagetti nest.

We know that certain methods can program the underlying
modules of the mind to supply the needs of an individual
that is basically what learning math is all about and it
is also why some people don’t get past some levels of it.
The unfortunate thing is that schools are a hammer and
anvil for beating humans into an ideological shape that
even the teachers may not be aware of. That process also
beats out of most the chance at true completion as a
person. You won’t find them reading this blog even though
it might help.

There’s a hilariously fallacious movie of Scarlett Js
that posits humans only use some small percentage of their
brain. If you check with a biologist or doctor they will
tell you that anything that isn’t used atrophies and
goes away. There’s a big difference between inaccessible
to ordinary consciousness and unused.

She turns into a nanotech computer once she reaches 100%
brain usage. That’s the hilarious part.

albert July 14, 2015 12:45 PM

“… This is a failure of the educators who setup the system, not the student’s rational response to it. …”. Yes. Furthermore, that ‘piece of paper’ represents a job (even though it may be a barrista at a Seattle Starbucks[Masters required]). Ya can’t get in the door without it.
@idea, @tyr,
Mapper vs. Packer, eh? Guess I’ll have to think about that. Since those concepts weren’t already packed in my brain, I’ll have to use the mapper approach. I submit the idea that we all use both approaches. Packers would do well at jobs dealing with rules, regulations, and laws, provided they are not totally dehumanized in the process. Mappers would be better at ‘creative’ thinking; coming up with novel ways to do things, or applying old concepts in new ways.

Intuition is the driving force behind creativity. Neither concept is easy to define, explain, or study, yet some of our greatest technological (and artistic) advances came through intuition. Regarding programming, assuming you have a clear and complete specification, the more thinking you do beforehand, the better code you’ll have at the end. Yeah, everybody knows that. “What were you thinking when you wrote this?” is a rhetorical question; the answer is always: “I wasn’t thinking”. I’m not talking about bugs, I’m talking about overall structure and function interactions.

“…Mostly because society discourages mapping–it looks like daydreaming…” A young programmer was added to my group. I gave him a relatively straight forward project; basic stuff. Every day I walked by his cube, I saw him reading, as far as I could tell, product specs. This went on for a about a week, but I said nothing, except a daily “how are you doing?”. Finally, after about 2 days of coding, he presented his program. It worked perfectly; his code was clean, clear, and concise, and we sent it off to beta testing. Never had a problem. I rest my case. There’s a lesson here, for managers. If that had been a ‘we need it yesterday’ project, what would we have gained by pressuring him? We might have actually lost quality. (if I had pressured him, I would have expected (and respected) a polite “get off my back; I’ll get it done.” response.

tyr July 14, 2015 6:33 PM

It is quite interesting to see how people view those
who think or are working that requires thinking to
do the job. If you’re expecting some external signs
akin to a stage presentation of Hamlet acting you’ll
be disappointed. The externals aren’t necessary for
thought. Pressure is useful some of the time but it
is usually just unwanted interference. Bottom line
is always performance and people in tech like to
show results and do the work. The tendency to move
the non-productive into management for non-performing
is a horrible way to do business.

A leader points the way and helps if you need help
after that he should be as invisible as thought
until you look at performance.

Catherine Punjani July 16, 2015 11:40 AM

The article on India – Cheating on Exams is very informative. Hi-tech cheating? I’m not a bit surprised.

TRX July 16, 2015 5:18 PM

They need to check to see if you can really think on your feet.

Everyone thinks they want a code wizard. What they really need are pit bulls who can stick with a problem until they solve it.

A buddy of mine is one of those pit bulls. His employer has flown him all over the world, and rented him out to other multinationals. Because he will find the problem and fix it, long after the code wizard responsible for the problem sloped off to some other employer…

TRX July 16, 2015 5:26 PM

This morning I entered an ‘F’ for each and every one of you into the
university’s online grade book.

My Calculus 201 instructor said something very similar. Then he laid out his after-hours tutoring rates.

The dean of students didn’t seem to see anything wrong with that, even after I explained that since I was paying for instruction I expected to get it. On the other hand, I wish I had thought to bring a camera, given the look on his face when I told him I wasn’t some kid there on a scholarship, and that I would call VISA and have them reverse the charge due to failure to deliver service. (and I did, too!)

Donald July 17, 2015 9:21 AM

My Calculus 201 instructor said something very similar.
Then he laid out his after-hours tutoring rates.

Ha! But a little different. The Computer Security professor, as I heard the story, fully intended to teach the class, but if you didn’t learn enough to hack the school computer system, you flunked. 🙂

Synonymous July 18, 2015 6:11 PM

The cheating scandal is growing more serious in India, based on an article in Los Angeles Times yesterday.

India test cheating stirs outrage — then people start dying

Looks like students there have been using high-tech vests to get answers; geek.com has a demo on these:

India’s supreme court demands students retake exams because of widespread high-tech cheating vests

Leave a comment


Allowed HTML <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre> Markdown Extra syntax via https://michelf.ca/projects/php-markdown/extra/

Sidebar photo of Bruce Schneier by Joe MacInnis.