Just about all of my engineering exams used option (D): let the students make their own. Teach them the material, tell them that they can bring a sheet of notes (or more), and let them figure out how to go from point a to point b. If they can't handle that, then what are their chances of figuring it out during the limited time of an exam? They have people, books, and the internet available to them well in advance of the need date, just like they would in a real job. At some point, they will be stuck in a room for a short period of time with no outside help available and people asking them questions that they had better know how to answer - just like a real job. You will not always have unlimited time and resources at your disposal; the students might as well learn this while they are still taking classes.

Been there, done that, never used the thing. About five years ago I bought a compact camera using this reasoning, but I rarely found myself in a situation where I felt like bringing a camera with me but didn't want the DSLR. I took maybe three pictures with it back in 2007 and haven't even gotten them off the card yet. The convenience seems like it should be worth it, but the compromises are just too great for me; I can't even go back to my old DSLR because of the big drop in image quality and capability. If I want to take pictures, it's no big deal to toss the camera and two or three lenses in a bag. If I don't, then I still have a cell phone for random stuff of little consequence - my pockets are full enough as it is, so even adding a compact camera is extra stuff to carry, plus different memory cards and batteries to deal with, different controls to figure out, and a big handicap in performance. I just don't see the middle ground, but that's just me. The best camera for me is the one that can get the shot. If I don't have it with me, then I just look and get on with my life.

Well, if that was the plan, then it worked too well; most of the cables and chargers that I've had for more than two years are completely useless due to wear. At first I thought it was the port, but a rarely-used cable worked perfectly while the others couldn't even get a momentary connection with any amount of wiggling. So much for not needing to replace all of your cables when you get a new phone...

And there's the problem - forget camera phones, DSLRs are becoming the new low end. DSLRs are the new big thing, but people want to use them like they use their cell phone cameras (point in the general direction, mash a button, post on Facebook). As a result, there's a big race to see who can make the cheapest DSLR, at the expense of quality of course (the reputation for higher quality is based partly on reputation from older models, so the perceived quality will remain as actual quality falls). Better quality is available, but the gap between the top of the low end and the bottom of the high end will grow, just like it did in the case of stereo equipment and every other piece of commodity hardware that started as a niche product. Most people who purchase a DSLR these days will seldom, if ever, change the lens. Few will ever use a mode other than Auto. And composition? Even among professionals, composition skills are far from a given. But megapixels? Everyone loves more megapixels, so just cram them in, image quality be damned! Manufacturers will design their products with this in mind, trying to woo customers with fancy gizmos, big numbers, and low prices. This will erode sales in the mid-range and stagnation in that product class (upgrading to the fancy doo-dads of the low end without actually making any improvements) will drive customers to the next tier up (typically full-frame for the prosumer crowd), further reducing the market. Mix in how the current economic conditions result in price increases at the high end (due to the strength of the Yen against the US Dollar) and you get the same scenario described in the article - crap quality for the masses and a high price tag for anything better. This has happened countless times before and will happen again many more times. The moral of this story is simple - buy quality while you still can.

While I am tempted to agree with this, I fear that reality isn't on our side here. There are two separate factors at work here clouding the issue. First, people who had this technology 20+ years ago were likely to be hobbyists and enthusiasts and not simple users, who make up the vast majority today. You can't expect everyone to be an expert in everything, so any technology that becomes mainstream will be dominated by users who treat it as a mysterious black box (this trend didn't start with computers and won't end with them either). Brace yourself though, computers are increasingly being designed to function accordingly, making it more difficult to be a casual hobbyist in order to simplify the user experience. Hardcore hobbyists will always find a way, but it will be more difficult to take advantage of features beyond the advertised mass-appeal feature set (particularly in areas that conflict with the desires of our corporate masters).

Second, the depression-era mentality of reusing and repairing everything until there's nothing left to work with is likely to die with us. I don't really know why this happened, but I would guess that cheap replacements, fast-evolving product lines, pop culture distractions, and flimsy plastic are key factors. Looking back at my own childhood, I would also point a finger at today's overly risk-averse parenting methods that restrict kids from learning about the world in an attempt to keep them safe. Everything real that we played with in our childhoods is being replaced by a fiercely-marketed "safe" alternative. Even many of the toys from that era have since been deemed unsafe and have been replaced. The problem with this mentality of raising kids in a safer alternate reality is that it leaves them unprepared for the challenges they will face in the real reality. Chemistry sets no longer contain chemicals. Woodworking kits no longer contain wood. Do they even allow kids to use electricity, fire, metal tools, or sharp objects anymore? Sure, there are safety concerns to be aware of (hence the need for parental supervision, which should always be a given with anything), and doing something can still be beneficial without any real-world applications, but replacing something real with a "safe" alternative adds another layer between doing and understanding. All of this extra safety only serves to delay the introduction of key life lessons while protecting against the occasional splinter, cut, scrape, etc. (which kids will still find a way to cause). If you don't learn to interact with real things early, you might not get used to thinking inside the black box, instead forever treating it as an unchangeable entity that either works or is trash.

4. Generate parity files so you can detect and correct data corruption; an extra 5-10% of data can quickly find and fix the occasional accidental overwrite or bit error. The crap-ton of DVD-Rs (or some other optical media, since DVD-Rs are a bit small for large quantities of photos these days) comes in handy here if the corruption was passed along through the backup chain and is too extensive to recover from the available parity files (in this case, the parity files tell you what you need to find and let you know when you have a good version). It's an extra step, but you can just do it once per set of photos and forget about it until you have a problem or want to feel proactive about data protection.

Did you read the rest of the post? Maybe YOU should read more carefully, this had nothing to do with international providers, it was a case of the domestic carrier changing the plan after the fact against the wishes of the customer. I'm pretty sure I mentioned that...

Yeah, there it is. I agreed to unlimited data for the days specified and still had this unlimited data plan active for a period of time afterward. There could be no data charges involving international providers until the global data plan was deactivated, which happened long after I had returned. The charges appeared because they retroactively canceled my global data plan two days before I requested it to be canceled. I'm not sure how much clearer I can make this. When a charge can show up has nothing to do with whether a charge should be possible in the first place.

Sometimes this excessive billing happens retroactively, making it impossible for the customer to keep track of charges or receive notification of reaching limits. I recently added a global data plan (at a rate of about $1 per day over my regular data plan) for an overseas trip and was shocked to see a $130 data charge that didn't show up until a couple of weeks after the trip. I had told the carrier beforehand the days on which I would need the global data plan (with an extra day on either end of the trip just in case), thinking that they would set it up to be active on those days. In reality, they just left the service on until well after I had returned, then retroactively changed my plan back to my regular data plan for the days I hadn't requested it. And a couple that I had. In order to save me a couple of dollars, they stuck me with $130 in data charges. Luckily, they fixed it without much trouble, but it shouldn't have gotten this far. These plan changes had to be made over the phone with no written confirmation of what had been requested and what was to be provided and no notification of the retroactive changes was given until it showed up on my bill (which was a tangled mess of charges and refunds). But hey, they might have lucked out and gotten an extra $130 for their incompetence.

This is basically how it works in engineering exams - either one sheet of notes is allowed or the exams are open note / open book. Hilarity ensues when students who didn't bother to learn the material try to cram all of the course material onto one sheet (usually by scanning multiple pages and printing them out together at such a small size that you would need a magnifying glass to read anything). I would just add key concepts to a running list as the course went on, usually only adding three or four lines per exam. And then I would go through the exam without referring to my sheet (because the process of isolating the key concepts forces you to learn them).

High-tech anti-cheating systems just aren't particularly useful for most engineering exams. Most methods of cheating are either easy to spot by a live proctor ("I wonder why that guy keeps looking at his left shoe...") or take too much time to be effective on a fast-paced exam. In the end, the results often don't even justify punishing the cheater. The worst case I ever saw was a guy who had his eyes practically glued to the exam next to him and did nothing to hide it. I considered reporting the incident, but realized that it didn't matter after I finished grading the exam; his efforts only got him 13 points (out of 100). If you don't know the material, cheating at the last minute is pure desperation, not a recipe for success.

Homeworks are another matter, but with group collaboration encouraged, cheating just has to be accepted and marginalized by minimalizing the impact of homework on the overall grade in favor of exams and projects. At worst, bad cheating on routine homework assignments is insulting to the grader and singles you out as a target. If you're too lazy to even cheat properly, you're just wasting everyone's time.

Of course, this viewpoint is really limited to engineering, and even then probably only certain disciplines (and I've been out of the loop for a decade or so). Written assignments are always risky, but in-class essays and presentations are an easy low-tech solution. Even just the occasional Google search can have significant benefit; the professor in one of my grad classes tended to assign essays as homework assignments and would routinely run web searches on key pieces of the answers that were turned in. Inevitably, he would find something copied verbatim without citation, even though it was stated up front that he would be checking the web (and it was a small class of less than 20 people). Though I suppose that doesn't make a strong case for these detection systems preventing cheating... Maybe instead of focusing on analyzing the assignments, teachers should work on getting to know their students. That's not exactly practical in many situations (especially in larger classes and with overworked teachers), but the alternative of a never-ending cat-and-mouse game doesn't look much more promising. Encouraging an adversarial relationship already makes people distrust other authority figures, I don't see it helping education outside of impersonal exam-processing facilities (which seems to be the conclusion of the article as well).

I spent two weeks in New Zealand earlier this year, and the countryside is lucky to have wired telephones. The North Island wasn't too bad (cell phone service available except where terrain blocked, somewhat slow but reliable free wireless at one hotel), but the South Island felt like a completely different country. If it was available, internet access at hotels tended to be slow and pricey (browsing web sites was often difficult or impossible and 10 cents per megabyte was typical) and prices were similar at internet cafes. Many hotels were stuck with satellite internet, so this is understandable. Cellular service (my unlimited data plan only cost about a dollar a day above what I paid in the US) fell off to nothing as soon as you left any densely populated area but was very good in any of the major cities. I can't speak to residential connectivity, but the image presented to visitors is not a pretty one, especially when you get out of the cities; it's so bad that you have to actually go outside to entertain yourself. They're just lucky that there's so much to see and do over there...