What the sales tax pays for probably varies depending on where you are, but one of the things I keep seeing in these sorts of stories are the assumption that it pays for infrastructure used by businesses not paying the tax. Is that really the case? It didn't match with my recollection on the matter so I did a bit of research. Anybody interested in doing similar should be able to find the information in the comprehensive annual financial reports from their state department of revenue.

In Wisconsin, sales tax makes up about 30% of the general purpose revenue for the state (about 18% of total state government revenue). This is used to fund several programs, most notably school aids, shared revenue paid to municipalities, medical assistance programs, the University of Wisconsin system, prisons, property tax credits, community aids, tax relief to individuals, other public assistance, and WI technical college system aids.

Since a big chunk of that is money paid to local governments, I then went to the budget for my city to see where that was going. It's paying for public safety (fire and police department), public works (planning, surveying, mapping, city engineering, emergency management, building inspection, trash removal, snow removal, bridge and street maintenance, weed cutting, street lighting, traffic signs), parks, recreation, and cultural services (several area parks, community centers, a museum, the zoo), and general administration. That public works section is about 16% of the city budget and street maintenance is about 23% of that, or about 3% of the total city budget. Looking at revenue for the city, shared revenue from the state would fall under intergovernmental revenue, making up a portion of 42% of city revenues.

So, while it's certainly possible that some sales tax money ends up funding local roads, the vast majority of that is paid for from a different fund through, for example, fuel taxes. Don't misunderstand me. There's a lot of good stuff that sales tax money goes to, but the suggestion that it's largely going to infrastructure that out of state businesses rely on doesn't seem to be true, at least where I am. YMMV.

I wouldn't hold it up as any kind of example of great code (so there are certainly opportunities to improve it and it is still under active development), but another open source (MIT licensed) literate C++ program you can add to the list is some software I've written for data logging/record keeping in commercial coffee roasting facilities. I've tried to keep the generated source documentation reasonably easy to read and understand and the program is used daily at several coffee firms throughout the world.

http://www.randomfield.com/programs/typica/index.html

It seems as though at least some airports are moving away from these things anyway. (warning, anecdotal evidence coming up) A few weeks ago I flew from Chicago to Houston. When going through security on the way out I didn't so much as see a porn scanner (there was one for the flight out of that airport I had taken before, but I went through a different security line this time). I also kept an eye out when leaving the secure area on the way back and again failed to notice one. Flying out of Houston, the security line split in two and I was free to choose either the long slow line with the porn scanner at the end of it (which was scanning everybody who chose that line) or the short fast line without it. I chose the short fast line, didn't get a groping, and was rather amazed at how many people chose the longer, slower line (I suspect most didn't notice the other line was available).

The sooner we stop wasting money on these things (and better yet, get them out of the airports so we can have room for more fast lines to get through security) the better.

To be fair, it does make a lot of sense to do something like that if you're going to be getting a large number of small dollar amount donations. There are costs associated with processing a donation and it's a lot cheaper for the charity to process a single $2M donation than 1M $2 donations. A retailer collecting that along with a purchase will have most of those costs to make the sale anyway. Doing it this way also provides a way to do that charitable donation as something like an impulse purchase, probably resulting in a lot of people donating who might not have considered doing it otherwise.

Now, I'd also say no because I'm not a huge fan of * awareness charities and the dirty look is bad service, but I fail to see the problem with the rest of it.

I'm sure that's on the roadmap, right after proper Unicode support.

What I've read on the matter is that the astrologers agree with your assessment. Nobody changed signs. Still sucks for the newborns who will be unable to record a high score in Gradius.

The economic realities of the global coffee market are not as simple as you are making them out to be. Some historical perspective:

There was once a time when coffee was trading at high prices. Demand was increasing, supplies were not increasing as much, and there were countries capable of producing coffees that weren't. Getting people to plant coffee in these non-traditional producing areas was seen as a way to develop economies. This idea was not without precedent. Brazil was largely built on coffee money, after all. Now, it takes a few years from the time coffee is planted until it produces its first crop. Coffee isn't a crop where you plant it, harvest it a few months later, and know what you've got. It takes about 10 years before you really know what you have with coffee. All of the sudden you have a huge increase in global coffee production. Making the problem worse, these new coffee producers were exporting coffees of such poor quality that more established producers never allowed on the global market. Now, coffee does not really fit the definition of a commodity, but it was usually sold with contracts that specify a differential above or below the NYBOT C price with those differentials based on several factors such as the quality of the coffee (coffee better than exchange grade selling for more, worse coffee selling for less) and the producing country (a comparative advantage model doesn't really work because a specialty grade coffee from Panama isn't going to taste anything like a specialty grade coffee from Tanzania).

Up until this point, the NYBOT C was cyclical. Sometimes it would go bust, but it would recover. This time it was different. There had been a fundamental change in the market which kept prices depressed for a long period of time and the differentials weren't keeping up. Simply letting quality producers go out of business wasn't going to be good for the producers and it wasn't going to be good for buyers either because someone who wanted to buy a nice coffee from Mexico wasn't going to be interested in some garbage from Vietnam. By setting a floor price for some coffees, Fair Trade did a lot of good in keeping those producers in business. You might find it interesting to look into the details of some Fair Trade producers and see what they're doing with those premiums. In many cases they're using them to improve the quality of the coffee and diversify the local economy, both things which reduce long term reliance on the Fair Trade premium, which seems to be exactly what you're advocating (though you suggest doing that before getting the capital needed to undertake such projects).

Now we're back into a period of higher prices for coffee (NYBOT C is over 200 as I write this and the differentials aren't dropping quickly which is a big part of why many coffee firms either have or soon will be announcing price increases) and this is due to many factors. Looking at the fundamentals, this is cyclical and prices will drop again, but probably not substantially within the next 18 months. We also have a lot more diversity in price discovery mechanisms for coffee which is a positive development, particularly for producers of high quality coffees. Any Fair Trade cooperative ought to be looking into getting out of Fair Trade in the long term exactly because that's not a long term sustainable model. Some cooperatives have already made that jump, others will follow, but to argue that Fair Trade was not beneficial in the long term simply is not supported by fact.

I only need two keystrokes to get a division sign (option/) so I'm not really sure what you're going on about there. That said, I don't think it matters so much in terms of writing the code. As others have pointed out, we've been down that road with APL. Where this sort of idea really shines, however, is in reading the code. One of the comments on the article touched on this, so I'll quote the relevant bit:

Robert Melton | Mon, 01 Nov 2010 02:11:13 UTC
What an odd combination of criticisms... first and foremost, I think you already hit on the correct solution... custom syntax and creation mechanisms are best explored as layers on top of existing tools, not a fundamental part of a new tool. I believe to try to integrate your ideas would have crippled Go, and given it nearly no advantages, at the cost of a huge degree of developer mind share. Bootstrapping a language is hard enough without giving yourself new disadvantages. I have never seen Guido van Rossum claim anything other the "readability" and that it was the natural flow as foreseen by Knuth (1974)

The problem here, however, is a cultural one. I suspect that most of the people who write software have never read through a non-trivial program and come out of it with an understanding of the program (contrast this with novelists reading novels) and most software is written in such a way that reading the code for understanding is more like assembling a puzzle from diverse bits scattered all about.

We already have pretty much all of these presentation niceties with CWEB. I frequently write my programs in literate C++ and can use goofy characters with subscripts if I want. TeX markup in comments is very nice. Especially useful is having something a bit more visually distinctive separating assignment from equality testing, but the big gain here is that it makes it easier to write programs as code narratives which humans can read to gain an understanding of the program. Once you have someone reading the program, how the code is represented starts to matter. Granted, this is another example from the pile of ideas that never really caught on.

Obligatory xkcd

My experience does not agree with your hunch. We've been running an open hot spot on a DSL connection for years. The phone company has known that we are doing that for a long time and their only concern is that they don't want to provide tech support our network (fair enough, I don't want them supporting my network either). The only thing that has happened with that connection over the years is that it has gotten faster and cheaper.

Interestingly enough, that's why I make a point of buying furniture that's slightly uncomfortable rather than the cushy sofas, to get people to stand up now and then and maybe wander back to the counter for another drink or eye the bakery case. Some other shops do the same thing with music that's slightly too loud, bad lighting, or keeping the place too cold. Once upon a time there was a coffee shop not that far from where I live that I'd go to. It was definitely a local youth hangout, but the owner kept the heat off, bad lighting, uncomfortable furniture. He didn't buy coffee from me, but he got good stuff and his staff knew what it was doing behind the bar. The place was always packed and did good business, but one day the owner decided to move back to his home town and sold the place to his employees. They sat down and rather than leave a good thing alone, thought it would be cool if they made it like the coffee house on Friends. They put light bulbs in the fixtures, turned on the heat, brought in the cushy furniture. The place was still always packed, but nobody ever got up to buy another drink and they were out of business in less than a year. I still miss them. None of the other half dozen or so coffee shops that have gone into that location since have been anywhere near as good in terms of serving a delicious drink. That said, I love my customers who stick around all day. They bring their out of town guests, have their meetings at my shop, and are really some of my best word of mouth advertisers.

You must work for a phone company as they have tried to tell me the same thing, but the typical high volume use here is a guy doing video conferencing with his girlfriend, a girl playing some game on Facebook, someone else on youtube, a real estate agent uploading photos, others doing light web browsing/email/IM, in addition to the business use traffic. DSL plans around here are more than enough to handle peak usage and most of the time it's much lower. I just don't see a dozen people all trying to torrent half of the Pirate Bay. A DSL connection is all my shop has and nobody has yet complained about that. It works just fine. Then again, for places with heavier users, going with lower bandwidth could be a gentle way of making the problem self-correcting as the people who are only there for the network will find it not very useful and move on while it doesn't really affect the people who use it as the nice convenience that it's intended as.

Right. I provided a breakdown in the first paragraph of what they're paying and it's mainly a one time expense that's insignificant when split among purchases (the AP, though some ISPs are starting to include that with service now) and an ongoing expense that the business would be paying if they offered the service or not (Internet access). There's nothing magical about the accounting, it just really doesn't cost that much. Now, if you start adding fancy features like receipt access codes, registration before login, and the like, the costs get larger but a shop that's just using a COTS wireless router and throwing it out there, my breakdown is accurate. Here's the real test for you. Have prices gone down at the shops that are no longer offering the service? I doubt it, just as there was generally no price bump to go along with introducing the service in the first place.