The market for dev kits can't expand in time to meet consumer demand, nor would it be cost-effective to try to do so. It takes a lot of capital to ramp up to full consumer production capacities. And, any dev kit taken out of the hands of actual developers will tend to limit eventual dev support at launch time. It's crucial to get those devices into the hands of actual developers in order to ensure there is actual support for the product at launch time. There's no need to expand access to this particular product, because it's not a consumer product.

Very true, I didn't mean to imply they don't. All I'm saying is that blowing up a shopping mall is a page three story, while blowing up a 747 is a page one story - nothing more than that. I'd guess that airline bombings just play into an already existing fear of flight for many, so the psychological impact tends to be magnified beyond the simple fact of the incident.

As far as why the US hasn't been hit (I assume you're talking about shopping malls, because we've certainly been hit plenty of other ways) - I'd guess it's perhaps because most people that *would* like to do so would logistically have problems getting to the US along with their bombs. It could also be that any major terror campaign against the US would be likely to bring down the wrath of the US military / law enforcement against said organization and any supporters. It's hard to say, really.

Honestly, I find it hard to get into the mind of people who randomly kill and maim civilians, including innocent children, so who knows what they're thinking. I'm glad the US is relatively free of terrorism, but I wish the rest of the world didn't have to suffer at the hands of those animals either.

Change that to "major catastrophic accidents of large commercial planes". Keep in mind that those statistics you linked to include passenger vehicles with greater than six vehicles (i.e. short-hop puddle jumpers or private aircraft). If you only include larger passenger aircraft, the number of incidents would probably dramatically. Note that higher on the page is a summary of *major* incidents of note. As the article itself points out, the Malaysian Boeing 777 is the only major large airliner accident this year, and similar incidents occur only once every few years.

You neglect the primary reason for taking a plane down when there are bigger and softer targets available (shopping malls, theaters, huge lines at the airport) - the psychological impact. People are already, whether they want to admit it or not, slightly terrified of flying at a subconscious level. Hurtling through the air in an aluminum tube at 30,000 feet is insanely unnatural act, and everyone knows it, but we've made it as safe as reasonably possible. People want to feel as though everything that can be reasonably done is being done to prevent guns or bombs from being carried on board.

Terrorism is a political statement. You want your message to be heard as loudly as possible, and there's nothing like taking down a plane for doing that, since a plane going down unexpectedly is guaranteed to generate world-wide headlines.

Some people criticize any language that doesn't meet some sort of "purity" standard, but languages and programming standards evolve over time, but perhaps they haven't clearly thought through the alternatives. There are only three possibilities that I can see.

1) The language stagnates, stubbornly refusing to fix flaws or grow over time.
2) The language rejects old "mistakes" and breaks backward compatibility in order to change or add new features, fragmenting the community.
3) The language grows to incorporate new features while retaining backward compatibility, thus becoming "bloated" or "a mess".

Personally, I think option 3 is the most attractive option of the bunch, and, if I'm not mistaken, this seems to describe Perl fairly well (I'm more familiar with C++, which has a lot of similarities in the "old cruft and baggage" department).

Obviously, many newer languages have the benefit of design-wise hindsight, but new languages also have an uphill battle to carve out a proper ecosystem and prove to be a long-term viable programming candidate. Some programmers are attracted to new and shiny things (we all are to some degree), but for serious work, devs need a language that can be counted on as a workhorse.

That's been my impression of Python as well. Fragmentation and stagnation, despite some assurances to the contrary. Granted, impressions don't necessarily equate to reality, but there you go.

It feels like some language devs tend to discount the importance of backward-compatibility for developers, or the longevity of code in general. People complain about all the cruft of C++ (which it does have in spades, of course), but part of the enormous success of C++ was undoubtedly due to the refusal of the standards committee to impose wholesale incompatibilities upon the language, even when that would have made serious improvements, and of course, the original decision to make C++ compatible with C as a starting point. Had they done otherwise, the result likely would have been the stillbirth of the language or the later fragmentation or damn near stagnation you see in Python, where years after the release of 3.x, developers *still* are stuck on 2.x because of the compatibility issues.

There are some cases where community forums make a lot of sense, of course. On occasion, you genuinely have a "community" of users that would be willing to help others answer questions with both technical and creative problems to solve, so why not take advantage of that? Some products are extremely high-volume and low-margin, and a single support call from even a small percentage of users would kill profits. Most users are smart enough to figure out that the quality of support should, by nature, tend to go up with the cost of the product. For instance, I wouldn't expect anything beyond community forums type support for Mozilla products, since I paid nothing for them in the first place.

Unfortunately, I feel too many companies simply try to use the "customer support forum" as a cheap alternative to real support. And it's frustrating to users when the only answer they can get from other users is "sorry, it's working great for me", and aren't even sure that the company in question is aware they're having problems, and can't even provide any sort of feedback.

Delegating that responsibility to a community forum (at least, without an alternative for "official" support) says to me, as a customer "I really can't be bothered about answering your questions. Here, you guys can collectively figure it out". Frankly, if a bunch of customers are asking the same question over and over, maybe the company who created the product should pay attention to what those questions are and try to design their product to be a bit more intuitive. Yes, good customer service is likely expensive, but the smart response is to design your products to minimize user questions and problems, not to cut out call centers.

There have been some companies I do business with which have surprised me with excellent one-on-one support, and it makes me much more willing to do business with them in the future, since I know I always have the option of working with a live and helpful human being to resolve any issues. Perhaps as importantly, I know they will work with me in an official capacity as a company representative, and they'll stick with the problem until it's actually resolved to my satisfaction. No matter how nice the people on the community forum are, they *are* only volunteering their time, and can't be expected to demonstrate that level of commitment unless you happen to get *extremely* lucky. It's one of those intangible expenses that is likely tempting for bean counter mentalities to cut back on, but which smart companies realize is an important investment in keeping customers satisfied.

1-2 times a year? That's certainly not typical for game development, at least in my experience. We're in our debuggers all the time - in fact, it's considered good practice to step through new code you've just written to make sure everything seems to be looking and working as you'd expect it to. With the development we do, there's not always an obvious correct or known solution to the problem you're trying to solve. Or, even if it is a problem with a well-known solution, it can involve writing rather complex algorithms and data structures that very few programmers could get perfectly the first time (i.e. the A* pathfinding algorithm). I can't imagine writing that sort of code without having a debugger available to watch the internals of your code along with development.

What do you do that only requires running the debugger once or twice a year? I'm guessing you're not using an integrated environment, or have deployment consideration which make using a debugger somewhat non-trivial.

The problem, I guess, is that could encourage patent trolls to simply hold down their own expenses while attempting to incur massive expenses on the plaintif. The idea is to discourage the "lottery mentality" of our current system, where patent trolls are attacking giant corporations in the hopes of a massive settlement or patent licensing fee. They're currently safe because their own expenses are relatively fixed (with retained lawyers on staff, or lawyers paid on commission - only if they win), while the defendant may be forced to spent a lot on their defense simply because of what's at state, or because of what they're being asked to prove.

Dryeo's answer probably makes the most sense here - leave it to the discretion of the judge. If they felt that the suit was brought without merit, they can order the plaintiff to pay 100% of the defendant's bills. If the suit was felt to be merited, then perhaps set at some minimum threshold, like what happens with criminal sentences.

It's not perfect, since it's still subject to the same sort of abuses by "specialty" judges. That, incidentally, is something that's may need to change as well - plaintiffs shouldn't be allowed to pick their private little courtroom to sue from. We see the abuse there as well. If you forced plaintiffs to sue in the home courts of the defendants, it might not be fair, as the judge may well be influenced by a powerful industry and act as its protector. Well, as it turns out, it's not fucking fair the other way either, as patent troll have built a nice little industry around judges that tend to favor their practices. Maybe for interstate disputes we should have to use some sort of lottery system, so these cases are randomly assigned to the territory of a disinterested 3rd party.

I don't know... things just get messy whenever lawyers are involved, because it's their damned jobs to find the maximum advantage within the framework of rules, and they'll keep doing so even if the rules change. It's sort of hard to fix a system like that, but for all we rant at lawyers, it's still better than the alternative. That is, it means that we're a society that's based on the idea, if not always the perfect execution, of the rule of law, and not the simple exercising of power.

Interesting, thanks for the link. According to the article at least, it sounds like the bill has been just delayed a bit longer rather than stalling outright. Also, the writer of that article appears optimistic that it could actually pass, which makes sense, given that it's apparently a bi-partisan bill.

Hopefully they'll hammer out a reasonable compromise. I can see the merits of both arguments here. The big issue is whether there have been any significant number of legitimate patent suits which would otherwise have been discouraged. Call me cynical, but I'd sort of be surprised if there actually were that many, given how badly the system is being abused right now. At this point, I'm probably willing to swing the pendulum a bit to the other side and go with "loser pays" penalties, since for so long there's been a massive financial incentive for actually abusing the system. I think it would help to balance the equation a bit. The law can always be amended in the future if the "loser pays" option turns out to be too harsh.

Should be interesting to see what happens with this.

Unless Texas is on that list, I'll give the states an "atta-boy", but it's not as though it will make a serious difference except for smaller business that can't afford regional offices. And let's face it, they're not the serious problem, since it's really only larger firms and patent trolls who go out of their way to set up offices in East Texas that are the problem.

Well, here's hoping that the federal government does the right thing here eventually. I'm typically not one to jump with knee-jerk reactions in favor of government regulation, as over-regulation gone overboard can have a stifling effect on business. Any reasonable analysis demonstrates massive and obvious problems with the current patent system, and the private sector has absolutely demonstrated an inability to handle this problem in a sane solution. In fact, the private sector has gleefully demonstrated that it's perfectly willing to exploit the situation and actually make a fucking business out of the problem. That's about the time for the government to step in and put the hammer down.

The only danger is that whenever the government steps in, there's a very real danger of making a problem worse despite all the best intentions. The individual state's efforts real legacy may be of giving some real, working examples of how to potentially fix the issue before it's tried out on a national level - that's certainly not unprecedented.

BTW, does anyone know why, in fact, the senate's patent reform bill is actually being held up (other than "politics" or "lobbyist")?

When gamers complain about unnecessary voice acting, I often wonder if they're reacting more to the poor writing and storytelling within a game. There's no need for voice acting if the dialogue is written poorly - it will only exacerbate the issue. Likewise, there's no need even for good writing if there's no interesting or coherent narrative to hang it off of, or gameplay that actually supports that narrative. This makes the dialogue, however well written, feel like a separate and therefore unnecessary part of the experience - more of a chore for the player to get past rather than something interesting for them to listen to and enjoy. I've seen a number of modern MMOs that suffer from this especially.

However, when gameplay, writing, and voice-acting combine into a seamless and coherent experience, the result is fantastic. There are plenty of modern examples of games that do this very well, like Mass Effect (ignoring the pooched ending of ME3), The Last of Us, the Uncharted, and many others. I haven't heard much in the way of complaints about the writing / voice-acting in those games because it's relatively seamless part of the gaming experience.

There are also players who are much more interested in the mechanics of games than the narratives of games, often heard in arguments against cinematics as well ("I want to play a game, not watch a damn movie!") I happen to love a well-told story integrated within my games. For me, a cutscene always feels like a reward after reaching some point in the game, but some players just see it as an interruption in the "fun" part of the game (i.e. gameplay). That's fine for those players, but many people actually do care about lore and story. These two sets of gamers will probably never agree on whether adding voice-acting is beneficial or a waste of money, because they're seeking fundamentally different things from the game experience.

Not every game needs cinematics or voice, of course. I think Limbo is a great example of a very minimalist game that does well without it. Likewise, it's hard to see how Minecraft would be enhanced by fancy cinematics and voice acting.

Having written several technical books and contributed to many others, I can't imagine why an author should have a problem using MS Word or the OpenOffice equivalent for writing - that's what the program was designed for, after all. If you want to just "focus on the text", it's easy enough to turn off the rules you don't like and pick a simple, plain font to work in. In fact, that's what my publisher actually required of me - they took care of all the formatting, of course, and I just turned in a plain, double-spaced manuscript.

I never really felt distracted by Word, and in fact, it was especially helpful when it came to editing / review time. For example, you can easily track changes you make within the document, then reject or approve those changed. This would be much more difficult to do without built-in support within the word processor. Likewise, it's easy for an editor to highlight a section and add a "sticky-note" like comment or question about it. Again, none of this is impossible without a word processor, but why would you want to throw that functionality away?

If someone wants to use a text editor instead of a word processor, that's fine. A few authors still use typewriters, or write longhand even. Don't pretend it will someone make someone a better or more efficient writer, though, or that it's even a good idea for most authors. This article only makes news because it's unusual. No one cares about the other 95% of authors who happily use Word or some equivalent on a modern computer.

My advice would be to just use the tools you're most comfortable with. If it's MS-Word, that's fine (that's what most authors I know use). If you want to use a text editor, that's fine too. Personally, I think someone who doesn't use a modern word processor is just making more unnecessarily work for themselves, but it's no skin off my nose. Ultimately, though, I don't think the tools matter a whole lot. I have a suspicion that the most successful authors/writers don't fret about about their tools nearly so much as they fret about their writing/stories.

Software does, in fact, age. It ages against changing conditions, demands, and threat vectors. Semantically, it doesn't really matter if the software "ages" or the threats simply become more advanced. The result is the same. That's fine if you're not exposed to changing conditions or threat vectors. Don't fool yourself though - modern OS's are harder for malware to penetrate than XP is because of address-space layout randomization, enhanced security models, etc. The big advantage older software has is that it's more battle-hardened than newer software, and security can only every be "proven" through actual, real-world use. If you're in a perfectly static environment (at home off the net, or an embedded XP machine, etc), then obviously there's no need to worry. Or in the author's case, if he's only using that old machine to write novels, then obviously it's as safe as a typewriter and perfectly fine for him.

I'd argue that most slashdotters probably are expert-enough users to safely use Windows XP. Use Firefox or Chrome instead of IE. Install no-script and ad-blocker. If possible, remove Flash and Java. Don't browse sketchy sites or open e-mail attachments or docs from *anyone* unless you're expecting them. Blah, blah. Unfortunately, average users will not do this. They'll download all sorts of malware or spyware and install it themselves. They'll open any attachment from anyone they get, even if it's named infect_your_computer.exe. For the "average" computer user, they probably need a more modern OS to help protect them. Hell, that's why tablets and phones with high-security models are better for the average user anyhow.

I was curious how you arrived at the conclusion that fiber would only take a year to pay off. I'd have guessed, given the amount of work I saw as fiber was being buried beneath my own street a number of years ago, that the costs probably ran a couple thousand dollars per house.

A quick search turned up this paper to give me an estimate of per-mile costs. Depending on the type of installation, laying fiber can range from $50K per mile (aerial lines) to $400,000 per mile max (installed underground via boring). My neighborhood was likely on the middle to high end, as they did indeed appear to be using boring techniques, and the neighborhood isn't super-dense.

Let's see... I'd guess that an average home in my neighborhood maybe takes up about 75 ft or so of street space on average, so fiber is reaching about 140 homes per linear mile (70 houses per linear mile x 2 sides of the street). Assuming a total installation cost of about $250K per mile, that would means each home cost a bit under $1800 to connect. With these costs, I think you're looking at more of a 10-15 years to recoup that investment.

Anyone with more real-world knowledge about this know if that's anywhere close to a reasonable estimate?