The problem with the entire article is that it tries to hand-wave over the real issue: user experience. The article basically suggests, "If we put lipstick on a pig, the pig will look much better." This kind of thought process is exactly why there is so much shitty software in the real world... suddenly, everyone is a designer, everyone knows how to do information architecture, interaction design, and visual design. Wikipedia has a problem, agreed, but it is not just a "superficial" one. If Wikipedia was designed with a different kind of end-user in mind (average Joe vs. geeks), it really could enable folks to contribute on a larger scale. However, there's a lot more involved in improving user experience than just making it look nicer.

So, what you're really saying is that spectrum is the cowbell of medical response systems. (Oblig. SNL reference)

This summary is incorrect. At least, according to the articles, Venema doesn't deny the idea of original sin. That was John Schneider, who taught theology at Calvin College. The summary is lumping all of these Christian scholars into the exact same theology.

Too bad I read the title as "Stock Exchange Website Profiler" and thought that someone had open sourced years of mining and forecasting stock information for the public to collaborate on.

Back on topic, this is quite interesting as well.

I had a similar problem from a person who registered an Xbox Live account under my email address. I couldn't find a way to contact the person, so after several months of getting his Xbox ads, I decided to log into his account and contact his friends to tell him to change his email address. (Probably not a smart idea.) That didn't work. After over a year, I finally got frustrated enough that on April Fools' Day (2011), I logged into his account and changed his avatar from a "lolz im so cool look at ma slipknot shrt n shades" to "elf fairy toddler wearing flower bell-bottoms". You would have thought that he would have got the hint after having to reset his avatar 3 times in the middle of his game, but he still didn't change his info.

In the end, I just set a filter to trash his messages. Moral of the story, it's not worth tracking down 95% of these problems.

Moonlight is far from Silverlight in terms of compatibility. About half of the webpages I visit (for example Ryanair's route map, which used to be cross-platform) nag me to install the Silverlight (aka Moonlight) plugin. Even after installing the plugin and restarting the entire system, the Silverlight control still nags me to install Silverlight.

I have worked with Silverlight before, so I can say that it has some neat features, but its cross-platform compatibility is sad. If Microsoft really wanted to push such a tool, they would have invested more in the Moonlight to ensure that the versions of Silverlight and Moonlight matched in capability. However, any further involvement would probably have undermined their current business direction at the time.

This is less of an inconvenience as more people are using smart phones with more functional address books. The right app could just export the address into an SMS or an encoding of the address that's understandable by the postal service. Alternately, a user could enter a postal code (and possibly a contact name) to isolate the range of the address. In the US, that could be a Zip+4, but Sweden does a similar thing with their 5 digit codes.

Until e-Cards become popular for birthdays (fat chance) and contracts are completely digitized, snail mail isn't going anywhere and thus this isn't a bad idea.

If I were designing such as system, I would encode the string with information about the end address, the timestamp of the text message, the identity of the sender (based on text messaging registration and phone number), and the ID of the transaction. If the "stamp" is unique to the transaction and sender/recipient, then I think that it would be fairly difficult to spoof (unless you had the person's SIM card).

Overall, I think it's an interesting alternative to stamps.

I am doing research in this kind of machine translation. Unfortunately, it is not ready for widespread use in medical offices at this point in time and can invite lawsuits. Usually translation systems need to be trained to a specific domain (medical, business, etc), because there are many ambiguities in translation. The typical (and state-of-the-art) approach is statistical machine translation, but it still suffers some flaws, because decoding has a running time in NP. There are many heuristics that are used now to speed up translation, but they don't guarantee a perfect (or correct) translation. While MT has made a lot of progress over the past few years, it's going to take some time to prove its accuracy before it is used in a place where a misdiagnosis (or a misinterpreted diagnosis) can invite a lawsuit.

Agreed. Universities are usually willing to sponsor students who submit their work to conferences if the work is of exceptional value because they improve the reputation of the university in the research world. So if you can get a free (or cheap) way to go to Utah and represent your university, you'll also get the chance to network with companies or research institutions that could benefit your career. It never hurts to have a publication on your CV -- even in the business world.

Telling your symptoms is probably the easiest part of communicating with a doctor in a foreign language. The hard part is understanding their reply. Machine translation in the medical field is intended to handle the more difficult parts of translation, such as explaining to the patient what's wrong. No "tourist" can be expected to know the names of all of the viruses or sicknesses they could possibly get when they go to another country. Additionally, patients need to know what their treatment options are without ambiguity so that they can make an informed decision. As an American researcher who has been studying in other countries, I know how difficult it could be when my wife had to go to the hospital.

Generally, you should learn languages that are most useful in regular day-to-day multicultural interaction. Seeing that you don't plan to lave the USA, perhaps it's not necessary for you to learn other languages. But I'm constantly impressed by Europeans I meet that speak 3-4 languages with ease and don't complain about my beginner's Italian or Dutch level.

While I think that this product is still a far cry away from what it needs to be (and far too expensive), this is a move in the right direction.

"Brains!"

There are plenty of machine translation conferences/shared tasks going on that are targeting patent translation (e.g. PatentMT - http://ntcir.nii.ac.jp/PatentMT/). Instead of just handing this patent task to Google, why didn't the European Commission host an MT shared task and give a prize to the winner? There are a lot of decent systems being designed by universities and research laboratories, especially in Europe. Oh, well, at least the European Commission is starting to adopt machine translation.

But what if your fridge could eventually adjust its temperature, based on the contents inside? Assuming that the fridge could have regulated compartments, you would no longer have to worry about your vegetables freezing while you're trying to keep your beer cold. Or, alternatively, maybe your beer doesn't have to be so cold when you're on vacation. One more way to help the average consumer save on energy costs.