I'm not sure I believe any of those numbers. A decade ago, I think I got $3k–$4k (I forget the exact number) for teaching a 10-week CS class as an adjunct. I think they're paying closer to $6k per quarter per class now. The only problem is, most folks only get to teach one class per quarter (at most), and you can't live on $18k per year in the Bay Area, so college adjunct faculty end up taking two or three jobs just to make ends meet. The problem isn't the pay per hour so much as the lack of sufficient hours. That and the fact that most classes involve spending lots of time outside of class grading papers, preparing lectures, etc. Three classes could easily be close to a full 40-hour work week, depending on the class. So even if you could manage three classes at $6k per class, you'd be working full-time for $54k per year (assuming you can't teach summer school). Again, you can't live on that out here.
As for your current salary, that isn't much higher than what I'm making at a startup, but I'm working roughly a 40-hour week, give or take. Where I work, we manage our project goals sensibly, we don't over-commit, and we deliver a quality product that is used by over a million people (and by about a quarter million in any given day). If you're really working 100 hours per week, your management is incompetent. No competent manager would work people 100 hours in a week, or even 60, because the productivity of programmers starts to diminish after barely twenty. By 60 hours, you're getting less done in a week than if you only worked 40 hours. By 100 hours, you'd get more work done by coming in for a single eight-hour day and spending the other four sitting around doing nothing.
You see, 100 hours per week, if it is a 5-day week, means that even if you have a cot to sleep on in the office, you're still getting less than 4 hours sleep. If it is a 7-day week, you would still have to live within a few blocks of the office to get eight hours of sleep, without doing anything else but working. Besides being very psychologically damaging, that is physically damaging, putting workers at dramatically elevated risk of premature death due to heart attacks, strokes, cancer, and pretty much every other disease you can think of. And on top of that, working so close to when you go to sleep would cause severe sleep disturbance, so even if everyone somehow managed to do that for more than two or three weeks without completely burning out and saying "screw this" and spending most of the day goofing off, they would still not be able to function at more than a fraction of their normal mental capacity.
Even menial jobs like factory work require better concentration than can reasonably be achieved with a 60-hour work week, much less a 100-hour work week. There's simply no way anyone remotely competent as a manager could ask for such things of employees, so if your management is doing so, you should start looking for another job right now. Because that's not the way the Silicon Valley normally works—not even at startups... except maybe game companies.
You have a good point regarding precedent. Of course the landmark case specifically ruled "sold at full price", so it makes Lexmark sense that Lexmark is pointing out the discount.
What is "full price"? We're not talking about a product sold directly by the manufacturer here; it's a product sold by a store at an arbitrary price point. The discounted ink cartridge at Office Depot might cost the same as the non-discounted version from another local retailer, at which point even that theoretical consideration goes out the window unless the cheapest local vendor sells both products.
And unless the customer was made clearly aware of the obligation to return the product prior to purchase (rather than tossing it out or returning it via the reseller's normal recycling channel)—and no, mere text on the box probably isn't sufficient, for the same reason that shrinkwrap licenses are dubious—I would argue that the discount likely wouldn't be perceived by the purchaser as consideration in exchange for agreeing to the license in the first place, in which case it isn't a valid contract, and because sale of a product is generally considered to automatically confer the use of all patents as embodied in that product by default (or at least that's my understanding), a preexisting contract between Lexmark and the customer is the only way in which such a license should be binding in the first place.
Besides, given Lexmark's sleazy DMCA abuse just a few years ago, I'd hope that any judge who knows their history would toss about the word "barratry" by the end of Lexmark's opening arguments, but maybe that's just me.:-)
On the other hand, I could see instances where agreements to sell with a condition make sense. Suppose I developed a tool for finding and buying things on eBay, which I successfully use for buying used RAID cards at the best prices. You want to buy baseball cards at the best prices. So I offer to sell you the tool, on the condition that you not use it to buy RAID cards, in competition with me. You can use it to buy anything else. That's fine with you, because you don't WANT to buy RAID cards, you want to buy baseball cards. Should it be illegal for us to make a mutually beneficial deal? (BTW I actually DID build such a tool, and it also benefits the sellers).
That's a little different, as it isn't really a consumer product that you would purchase off the shelf. Presumably that's software, which typically does require someone to agree to licensing terms. Whether those terms are valid or not is... complex, but given that you're talking about specialized software that is presumably sold directly to specific customers, there's nothing preventing you from requiring them to sign a contract that limits their use with a noncompete clause prior to delivering the goods. Either way, it would be contract law and/or copyright law that would be used for enforcing such terms, rather than patent law.
More importantly, there's no consideration. Once you have purchased the product, you own it, and presumptively have a legal right to use it. Lexmark might claim that they're giving consideration with their discount, but that discount was granted at the time of purchase, prior to when you were asked to agree to this so-called license. They have no legitimate legal right to demand further money after purchase in exchange for you not having to return something that you have purchased. Purchases don't work that way. So in the absence of evidence to the contrary, such as a signed agreement made at the time of purchase, that so-called license grants you nothing that you didn't already have. Because there is no consideration, there is no contract. Because there is no contract, you are not bound by the terms of the license. Or at least that's the way I would interpret the law as it pertains to the latest in Lexmark's long string of idiotic schemes to rip off their customers.
I don't know what is more ridiculous: the fact that they keep pulling this crap or the fact that people keep buying their crap knowing what an abusive company Lexmark is.
Was every store required to carry both SKUs? If the answer is "no" (and given how valuable shelf space is, I'd bet that it is), they're effectively forcing customers to buy what is readily available. And even if the answer is "yes", they still probably don't have a case.
Normally, a physical product (as opposed to something that has actual copyright protection) cannot be licensed. It can be sold, or it can be rented. Either the customer owns it or they don't. If they do, then they can do anything they want to with it. Once a product is sold, the original manufacturer has no legal right to limit its use. The SCOTUS has consistently tossed out attempts at post-sale restraint.
If a product is rented, there must be a legal contract in place, which means, among other things, that the customer must clearly understand that he or she is just renting the product, rather than buying it. If it looks too much like a sale, once again, post-sale restraint gets tossed out. And that's true even for cases involving patents.
I'd be utterly shocked if Lexmark won this, assuming it bubbles up to a high enough court. There's way too much case law precedent saying that they should get their a**es handed to them. It would require the courts doing a complete 180 from their consistent position on this issue over the last century.
Actually; at 30 km/h almost no accident is fatal, and at 50 km/h almost all of them are. Doing 5 mph (8 km/h) over a 24 mph (30 km/h) speedlimit will lead to a lot more more deaths statistically. Also getting hit by a car while speeding across and intersection is not a common way to have an accident as a cyclists, fatal accidents being even less common.
First, I should point out that when I said that 5 MPH was unlikely to increase the risk of death in an accident, I was talking about for the driver of the car, not for a cyclist hit by the car. I was comparing the increased risk to a car's driver from driving in an unsafe manner against the increased risk to a cyclist from riding in an unsafe manner. I guess that wasn't quite as clear as I'd intended it to be.:-)
Second, you're presumably basing your assumptions on speeds of roads in Europe, where minor streets often have much lower limits than here in the 'States. With the exception of school zones, which are typically 15 MPH (~24 km/h), very few American roads are posted lower than 25 MPH (~40 km/h), even in residential areas. So even on the safest American roads, you're halfway from your never fatal speed to your usually fatal speed.
Outside of residential neighborhood streets (and sometimes even on residential streets), the vast majority of roads are posted at either 30 or 35 MPH (48 or 56 km/h), which means they're solidly at the "usually fatal" end of your scale.
So basically, speeding by 5 MPH on most streets will make little difference to cyclist fatality rates because almost all wrecks at full speed will be fatal either way. Either the driver sees the bicycle and successfully stops (or nearly stops) or doesn't. In the first case, the cyclist lives. In the second case, the cyclist doesn't. Short of getting the meatbags out from behind the wheel, the only thing that will significantly improve cyclist safety is cyclists avoiding roads with high speed limits and being extremely cautious in their riding behavior.
That's in large part because speed limits are so frequently set 10 MPH or more below the prevailing speed of traffic on the roads. The correct question is what percentage of people travel faster than the maximum safe speed for the roads. That number is much, much less than a third.
The purpose of speed limits is not to minimize the risk to cyclists. If that were the case, all road speed limits would be set at 15 MPH and nobody would get anywhere. The purpose of speed limits is to mitigate the risk of cars pulling out in front of you, based on the number of driveway entrances, the number of them that are blind, etc. And occasionally, speed limits are designed to reduce the risk of death if a kid runs out in front of you near a school, where there is likely to be a high concentration of young children.
Note that most deaths by bicycles are on large urban arteries, which you would typically expect to have speed limits of at least 40 MPH. At that rate, a collision has a 50% fatality rate, give or take. There's only one good way to prevent those accidents: get the bicycles off of those major arteries and onto secondary streets where they won't be consistently traveling at a fraction of the prevailing speed of the road. The problem is not the drivers. The problem is that those bicycles shouldn't have been on those roads. They simply aren't capable of achieving the speed necessary to not pose a serious safety risk when used on those larger streets.
It's not just technically correct, it's actually correct. You're basically blaming murder victims for getting murdered here - "shouldn't be out at night" type of bullshit. No. It's the driver's fault almost all the time and it's the attitudes of drivers that must be changed.
Blame is too strong a word, as that implies primary fault. I'm saying that they bear some responsibility. If I break into your house and steal something, your insurance pays for it. If you leave your house unlocked and I walk in and steal something, your insurance probably doesn't. It is the burglar's fault primarily in either case, but in the second situation, you failed to take the expected basic precautions required to ensure the safety of your property, and as such, the insurance company is within its rights to deny the claim because of contributory negligence on your part.
From what I've seen, drivers' attitudes have absolutely nothing to do with the problem. Drivers are almost invariably courteous around bicycles when they see them in time. However, the reality is that driving can be enormously complex, and it is very easy to make mistakes, whether through fatigue, distraction, or simply having too many pieces of information flying at you all at once.
The only people whose attitudes can change in a way that would have a meaningful effect on bicycling safety are the cyclists themselves. Cyclists must consistently avoid getting too close to cars, avoid taking unnecessary risks, always wear attire that makes them easily visible, avoid making sudden stops that can get them rear-ended by cars that are trying to get through traffic lights, and generally assume that nobody else on the road can see them until proven otherwise. Ride defensively. If everyone did this, we would have a lot fewer cycling deaths.
Have you ever broken a speed limit? Speeding contributes to accidents in far higher proportion than anything else. (Google is your friend). If you've ever exceeded the limit, you're a damned hypocrite if you complain about bikers breaking the rules.
Breaking the speed limit by 5 MPH results in a fairly small increase in the odds of an accident being fatal. Speeding across an intersection on a bicycle without stopping greatly increases the risk of a fatal accident, from zero to 100% if you get hit. And it is fatal accidents we're talking about here. The statistical difference in risk makes that comparison utter crap. Your argument is like calling someone a hypocrite for complaining about the guy doing 120 MPH in a 25 MPH zone simply because that person is known for going two or three MPH over the limit.
2) Why do other countries like Denmark, The Netherlands etc not have these problems, even despite not wearing helmets, cycling like crazy people etc?
I'm guessing their cyclists are more careful, they have fewer drunk drivers and drunk cyclists, they have more bike lanes, and fewer people ride bicycles on major streets there.
American roads have a tendency to be a lot more uneven than roads in Europe, at least from what I've seen. In Europe, most of the roads I saw were almost either highways (no bicycles allowed) or they were two-lane city streets. Here in the U.S., we seem to have a much larger percentage of four-, six-, and eight-lane major city streets, often with no smaller parallel streets that are suitable for bicycles (because they all get cut by highways every two or three miles). The largest percentage of bicycle fatalities happens on those large urban streets.
I blame, at least in part, the "traffic calming" nonsense that has become the norm in lots of cities, where it is impossible to go anywhere on back streets without eventually getting to a point where you're forced out onto one of those major arteries by side streets that don't connect. I also blame the haphazard way in which American highways are allowed to carve a slice right through a city, rather than being elevated or built underground as is common in much of Europe, allowing bicycles to pass over or under it on every existing cross street.
But the biggest problem, IMO, is the steady increase in average commute times. This leads to drivers being more and more fatigued by the time they get off the highway at the end of the day, and surprise, surprise, that's when most of the fatalities happen. We have an epidemic of long commutes in California, brought on in no small part by prop 13 creating a tax disincentive for selling your home and moving closer to your job. This, coupled with poor regional land use planning, has created a situation where about 80% of the roughly 120,000 people living in the greater Morgan Hill/Gilroy/Hollister area commute an hour or more to work in the Silicon Valley because there are almost no tech companies with offices in that area. Same goes for a much smaller (but still sizable) percentage of the greater Santa Cruz area, plus everything from the East Bay all the way up to greater Sacramento.
3) The vast, vast majority (that big G again for those who want the stats) of bike deaths are directly and entirely attributed to driver error, not cyclist error.
That's technically correct—the best kind of correct. However, according to New York's statistics:
20% of cyclist fatalities are directly attributable to the cyclist having a BAC over the legal limit. These are likely 100% avoidable by the cyclist not getting on a bicycle while intoxicated.
40% involve rear-end collisions when the cyclist stopped without getting off the road. These are also likely 100% avoidable by cyclists being consistent about pulling to the side of the road when stopping—particularly if a traffic light is about to turn red.
Well, it doesn't give you the right to hit the cyclist, if that's what you mean. It does give you the right to tap your horn, or maybe even call the police, if you're feeling particularly vindictive that day.:-)
You're an absolute fucking moron if you think that doesn't happen in natural languages.
It is legal in most natural languages, and is quite common in non-imperative sentences. However, for imperative sentences (commands), putting the action before the condition is not the normal way that people speak in most natural languages, precisely because it leads to mistakes. For example, in a high-stress situation, if one person shouts to another "Shoot her if she moves", there's a high probability that the other person will shoot reflexively before even hearing the last half of the sentence.
Similarly, shifting the "if" statement off to the right end of the line makes it much more likely to be missed by someone scanning down the column of text, because people naturally skim down the justified edge of a block of code, rather than the ragged edge. This language design anti-pattern makes Perl code considerably harder to skim than languages that do things in the more typical order of condition followed by action. (To be pedantic, C does provide one statement type—do-while—that puts the action before the condition, but the action is preceded by an additional token that makes it obvious that the action is conditional. This distinction matters.)
But you don't have to parse the regex! You have to parse the delimiters. It's the same three lines of C code, only with a / as delimiter instead of a "
If someone uses brackets as delimiters, your code has to also parse character classes correctly.
There was a study done one this, mostly pointing out that there was a lot of confirmation bias in play - that people who were incensed at bicyclists not obeying traffic laws generally did not noticed automobile drivers committing similar acts. (This is, of course, a general observation, I can't say jack about you in particular.)
Even if true, the risks to a cyclist doing something stupid are much greater than the risks to an automobile driver, statistically speaking, because they are much less visible, and they don't have a giant metal box wrapped around them to offer protection. There should be exactly zero cyclists doing these things. If there are enough bad cyclists for us to even notice their poor habits, that's already way too many bad cyclists. If the percentage of cyclists doing these things is comparable to the percentage of motorists doing so, then that is likely to be the cause of a sizable percentage of those cycling deaths, and a negligible percentage of deaths by motorists.
A bicycle has the same right to use the road that a car, pickup, or tractor trailer has.
Yes, and bicycles traveling in a traffic lane at speeds significantly below the posted limit are required by law to periodically pull off the road to let cars around, just like any other slow-moving vehicle. Neither bicycles nor cars nor any other vehicle are allowed to block traffic for an extended period of time.
Guess what? If the lane is not wide enough for a car to pass me with (in FL) a three foot gap, then, yes, I do have the right to "block" a full lane of traffic...
Not really. You are allowed to do so for a short time, but not indefinitely. In particular, most states have slow vehicle laws that require you to pull off the road if you are unable to drive at a speed that does not impede the flow of traffic. The exact laws vary widely from state to state.
California, for example, requires that you pull off the road if there are five or more cars behind you; the same law also applies to slow-moving cars, trucks, etc. (California Vehicle Code section 21656)
This. I would have put it more tactfully, but a thousand times, this.
Almost without fail, cyclists ignore stop signs, ignore traffic lights, turn without signaling, and generally ignore traffic laws willy nilly. It's one thing to make a right turn on a bicycle without stopping; it is quite another to speed past a stop sign, going straight through an intersection where the opposite direction doesn't have any obligation to stop.
The number of serious mistakes made by cyclists in my presence has been climbing steadily over the past decade or so, which suggests that the level of recklessness exhibited by cyclists is increasing at an alarming rate, or at least the number of idiot cyclists is increasing at an alarming rate. You can't have cyclists ignoring the laws of the road and somehow expect drivers of cars to magically take up the slack. Responsibility for cyclists' safety must be shared.
From my perspective, regular expressions outside the context of a string leads to, among other things, all sorts of fun for IDEs, where they can't properly do brace matching because you have a brace in a regex and they don't know what to do with it. The string syntax is much less problematic because they invariably follow bog-standard language parsing rules, which are much easier to handle than parsing regular expressions (where the rules are inconsistent from one language to the next in subtle ways, with Perl being the biggest mess).
What makes Perl's regular expression engine particularly nasty is that there's more than one way to do it. Don't like slashes as regex relimiters? Why not use brackets? Or parentheses? Or curly braces? The complexity of the parsing rules goes up by a sizable margin every time you add such "handy" features. And the way special characters are handled inside a regular expression depends on context, which makes it a serious pile of hurt.
By contrast, I can parse a regex in a C string in about three lines of code (count the number of double quote marks not preceded by an odd number of backslashes or inside single quotes, and if you get an odd number of double quote marks, you're inside a string), and any brace matching engine that can't handle such simple rules is just sad.
By backwards instructions, I mean that in C, you write this:
if (foo) bar();
whereas in Perl, you write this:
bar() if foo();
That design violates the norms of natural languages, where conditional expressions almost invariably precede the actions that they modify, and as such, leads to many an incorrect assumption when skimming other people's code.
Yeah, I almost made a comment about the vast quantities of PHP code out there, but decided against it because so much of that code probably shouldn't be out there. These days, before I bring in any piece of PHP software, I tend to spend at least a day or two doing a cursory security audit in which I look for SQL injection and shell command argument passing vulnerabilities. I usually find some.
But yeah, at least from a jobs perspective, the community, codebase, support, and ease of hiring are what makes a language valuable. And the more bad code there is out there, the easier it is to get a job fixing all the holes.:-D
Sure. Lots of folks use PHP for writing quick server code, because it makes light work of prototyping things. If you know what you're doing, it is a solid programming language, offering a fairly clean, C-like syntax, without the horror of Perl's backwards instructions and bare regular expressions, the OO bloat of Java, the need to install additional interpreters (problematic on shared hosting services), etc.
You know that Lightroom is still available for sale, right?
When Adobe announced the "Creative Cloud" nonsense, I bought a copy of CS6, upgrading from CS3, knowing that if I didn't act, I would never be able to do so again. At that point, I immediately began phasing out my use of Photoshop. Unfortunately, I haven't created any big new projects since then, and I'm stuck using Photoshop for existing projects because Pixelmator doesn't handle manual text kerning changes correctly on import. But the cover art for my fourth novel will be done entirely in Pixelmator.
You see, Photoshop hasn't added anything I really care about since they added layers and layer effects. There are a few minor enhancements that are nice to have, but I was happy on Photoshop 7. I buy upgrades to Photoshop whenever an OS X upgrade breaks it badly enough. It annoyed me badly enough paying for bug fixes when I was doing it on my own terms. When Adobe tried to push me to a monthly fee schedule for Photoshop, I walked away and haven't looked back.
I still buy Lightroom about every second release (about $40 annually), because unlike Photoshop, its upgrades actually provide tangible benefits—new camera support, face recognition, etc. (Yeah, theoretically Photoshop upgrades technically add new camera support, too, but I've imported RAW files into Photoshop a total of three or four times ever; typically, I start out with the processed output from Lightroom.) The problem with Photoshop is that it is a mature product, and there aren't any cool new features left to add that provide enough benefit to pony up an extra $80 a year to get them.
That's a serious problem for companies like Adobe. You see, they're in a position where they command the market. The only place to go from there is down, and the writing is on the wall. Flash is a failure. Photoshop is rapidly seeing the bottom portion of its market worn away by competition. This leaves Adobe's last remaining market—big graphics shops. Those folks will keep buying Photoshop until they no longer find themselves exchanging Photoshop files with other companies. Adobe is thus bleeding those companies for every possible penny they can before Photoshop eventually fades into obsolescence. Or at least that's what it looks like from where I'm standing.
No, it really doesn't. There aren't too many new shows. There are simply too few shows getting renewed. Every episode that has aired before is "new programming". There's certainly not too much of that. If anything, there's too little. If you watch whatever is on Netflix, you'll run out of new content before you watch TV every evening for two years. After that, you'll be sitting there looking for new content and not finding any that looks interesting to you. Yet the shows that you aren't interested in are probably interesting to other people, so those shows should still continue to exist.
The real problem, unfortunately, is what it has always been: ratings-driven executives who are not content to pay for shows unless they find their audience instantly and are broadly popular. This leads over time to progressively lower-quality programming that dumbs down everything to be "good enough" for the largest possible audience, while being seen as "good" by none of them.
What we need is what we have always needed: executives who have the courage to allow a show the time to find its audience, even if that takes three or four seasons, who are willing to guarantee more than a half season at a time, giving people a reason to think long-term about a show's future. Without that, most TV will continue to be what it has become: crap.
Pretty much. There are a lot of other things that also kind of fall into that category for me—salary, for example. I'm not likely to jump ship for a higher salary (in the absence of other reasons to leave a company), but if you stop paying me a salary, I'm not likely to stick around.:-D
Maybe pay is a problem for IT, but it usually isn't an issue at all for programmers. For example, here in the Bay Area, at roughly $90-$150k per year, the hourly rate would still be better than most jobs even if it required working 80-hour weeks. The problem is that most people can't survive an 80-hour work week for more than a couple of weeks, and even a 40-hour week is horribly inefficient and, frankly, exhausting at times.
The 40-hour work week is optimal for menial tasks that require very little thinking. For a technical workforce that spends most of their day thinking and trying to solve complex problems, workers are most efficient when working six or seven hours per day, not eight, and certainly not 12. Long before they reach the 50-hour mark, they're actually getting less work done per week than somebody working a 35-hour week, because they have less energy and are less focused. In fact, I rather suspect that the optimal work week in tech is somewhere closer to 20 hours, and that even a 35-hour week involves significant loss of efficiency.
What we need is for employers to hire twice as many people, pay them half as much, and work them half as long. Doing so on a broad scale, however, would require some serious changes, particularly in the way we try to attract people to the field. But it should be done, not just because employees would be happier with a better work-life balance, but also because employers would be getting what they paid for instead of only about two-thirds of it.
Things that don't matter:
Lunch or snacks (free or otherwise)
I'm currently working for a company that feeds us three days a week. I'm very grateful for that, because we moved from an area that had a few restaurants within easy walking distance to a new location that has none. At our exit off the 101, there's one restaurant that's kind of like a Denny's (not fast at all), one Mexican restaurant, and that's it. None of them are within walking distance (okay, so there's one Mexican food truck, but...), and the shared parking lot for those two restaurants is always full. Other than that, there's no food until you get all the way to the next exit off the 101 in either direction.
Worse, when you do go one exit away in either direction, you similarly find only two or three restaurants, also with inadequate parking. Most of the time, I end up driving three exits north on the 101 to where there are enough restaurants to actually be practical (including a few with drive-thru windows), but that means burning about half of my lunch hour just driving to and from the restaurant.
So yes, it doesn't matter even slightly whether an employer provides food for free, or even provides it on-site, but it matters a great deal whether there is ready access to food within a reasonable walking distance. If there isn't, and if the employer doesn't provide food, it can get rather annoying.:-)
We've seen what happens when you treat unrepentant criminals as outcasts. They remain unrepentant criminals. The recidivism rate in American prisons should be proof enough that such tactics don't work. Or take a look at countries that try to imprison the opposing forces after a civil war. Things never stabilize. Sometimes, forgiveness by others is the first step towards repentance—turn the other cheek, and all that.
Slashdot Mirror
I'm not sure I believe any of those numbers. A decade ago, I think I got $3k–$4k (I forget the exact number) for teaching a 10-week CS class as an adjunct. I think they're paying closer to $6k per quarter per class now. The only problem is, most folks only get to teach one class per quarter (at most), and you can't live on $18k per year in the Bay Area, so college adjunct faculty end up taking two or three jobs just to make ends meet. The problem isn't the pay per hour so much as the lack of sufficient hours. That and the fact that most classes involve spending lots of time outside of class grading papers, preparing lectures, etc. Three classes could easily be close to a full 40-hour work week, depending on the class. So even if you could manage three classes at $6k per class, you'd be working full-time for $54k per year (assuming you can't teach summer school). Again, you can't live on that out here.
As for your current salary, that isn't much higher than what I'm making at a startup, but I'm working roughly a 40-hour week, give or take. Where I work, we manage our project goals sensibly, we don't over-commit, and we deliver a quality product that is used by over a million people (and by about a quarter million in any given day). If you're really working 100 hours per week, your management is incompetent. No competent manager would work people 100 hours in a week, or even 60, because the productivity of programmers starts to diminish after barely twenty. By 60 hours, you're getting less done in a week than if you only worked 40 hours. By 100 hours, you'd get more work done by coming in for a single eight-hour day and spending the other four sitting around doing nothing.
You see, 100 hours per week, if it is a 5-day week, means that even if you have a cot to sleep on in the office, you're still getting less than 4 hours sleep. If it is a 7-day week, you would still have to live within a few blocks of the office to get eight hours of sleep, without doing anything else but working. Besides being very psychologically damaging, that is physically damaging, putting workers at dramatically elevated risk of premature death due to heart attacks, strokes, cancer, and pretty much every other disease you can think of. And on top of that, working so close to when you go to sleep would cause severe sleep disturbance, so even if everyone somehow managed to do that for more than two or three weeks without completely burning out and saying "screw this" and spending most of the day goofing off, they would still not be able to function at more than a fraction of their normal mental capacity.
Even menial jobs like factory work require better concentration than can reasonably be achieved with a 60-hour work week, much less a 100-hour work week. There's simply no way anyone remotely competent as a manager could ask for such things of employees, so if your management is doing so, you should start looking for another job right now. Because that's not the way the Silicon Valley normally works—not even at startups... except maybe game companies.
That and there are fewer of them trying to hump her leg at staff parties....
What is "full price"? We're not talking about a product sold directly by the manufacturer here; it's a product sold by a store at an arbitrary price point. The discounted ink cartridge at Office Depot might cost the same as the non-discounted version from another local retailer, at which point even that theoretical consideration goes out the window unless the cheapest local vendor sells both products.
And unless the customer was made clearly aware of the obligation to return the product prior to purchase (rather than tossing it out or returning it via the reseller's normal recycling channel)—and no, mere text on the box probably isn't sufficient, for the same reason that shrinkwrap licenses are dubious—I would argue that the discount likely wouldn't be perceived by the purchaser as consideration in exchange for agreeing to the license in the first place, in which case it isn't a valid contract, and because sale of a product is generally considered to automatically confer the use of all patents as embodied in that product by default (or at least that's my understanding), a preexisting contract between Lexmark and the customer is the only way in which such a license should be binding in the first place.
Besides, given Lexmark's sleazy DMCA abuse just a few years ago, I'd hope that any judge who knows their history would toss about the word "barratry" by the end of Lexmark's opening arguments, but maybe that's just me. :-)
That's a little different, as it isn't really a consumer product that you would purchase off the shelf. Presumably that's software, which typically does require someone to agree to licensing terms. Whether those terms are valid or not is... complex, but given that you're talking about specialized software that is presumably sold directly to specific customers, there's nothing preventing you from requiring them to sign a contract that limits their use with a noncompete clause prior to delivering the goods. Either way, it would be contract law and/or copyright law that would be used for enforcing such terms, rather than patent law.
More importantly, there's no consideration. Once you have purchased the product, you own it, and presumptively have a legal right to use it. Lexmark might claim that they're giving consideration with their discount, but that discount was granted at the time of purchase, prior to when you were asked to agree to this so-called license. They have no legitimate legal right to demand further money after purchase in exchange for you not having to return something that you have purchased. Purchases don't work that way. So in the absence of evidence to the contrary, such as a signed agreement made at the time of purchase, that so-called license grants you nothing that you didn't already have. Because there is no consideration, there is no contract. Because there is no contract, you are not bound by the terms of the license. Or at least that's the way I would interpret the law as it pertains to the latest in Lexmark's long string of idiotic schemes to rip off their customers.
I don't know what is more ridiculous: the fact that they keep pulling this crap or the fact that people keep buying their crap knowing what an abusive company Lexmark is.
Was every store required to carry both SKUs? If the answer is "no" (and given how valuable shelf space is, I'd bet that it is), they're effectively forcing customers to buy what is readily available. And even if the answer is "yes", they still probably don't have a case.
Normally, a physical product (as opposed to something that has actual copyright protection) cannot be licensed. It can be sold, or it can be rented. Either the customer owns it or they don't. If they do, then they can do anything they want to with it. Once a product is sold, the original manufacturer has no legal right to limit its use. The SCOTUS has consistently tossed out attempts at post-sale restraint.
If a product is rented, there must be a legal contract in place, which means, among other things, that the customer must clearly understand that he or she is just renting the product, rather than buying it. If it looks too much like a sale, once again, post-sale restraint gets tossed out. And that's true even for cases involving patents.
I'd be utterly shocked if Lexmark won this, assuming it bubbles up to a high enough court. There's way too much case law precedent saying that they should get their a**es handed to them. It would require the courts doing a complete 180 from their consistent position on this issue over the last century.
First, I should point out that when I said that 5 MPH was unlikely to increase the risk of death in an accident, I was talking about for the driver of the car, not for a cyclist hit by the car. I was comparing the increased risk to a car's driver from driving in an unsafe manner against the increased risk to a cyclist from riding in an unsafe manner. I guess that wasn't quite as clear as I'd intended it to be. :-)
Second, you're presumably basing your assumptions on speeds of roads in Europe, where minor streets often have much lower limits than here in the 'States. With the exception of school zones, which are typically 15 MPH (~24 km/h), very few American roads are posted lower than 25 MPH (~40 km/h), even in residential areas. So even on the safest American roads, you're halfway from your never fatal speed to your usually fatal speed.
Outside of residential neighborhood streets (and sometimes even on residential streets), the vast majority of roads are posted at either 30 or 35 MPH (48 or 56 km/h), which means they're solidly at the "usually fatal" end of your scale.
So basically, speeding by 5 MPH on most streets will make little difference to cyclist fatality rates because almost all wrecks at full speed will be fatal either way. Either the driver sees the bicycle and successfully stops (or nearly stops) or doesn't. In the first case, the cyclist lives. In the second case, the cyclist doesn't. Short of getting the meatbags out from behind the wheel, the only thing that will significantly improve cyclist safety is cyclists avoiding roads with high speed limits and being extremely cautious in their riding behavior.
That's in large part because speed limits are so frequently set 10 MPH or more below the prevailing speed of traffic on the roads. The correct question is what percentage of people travel faster than the maximum safe speed for the roads. That number is much, much less than a third.
The purpose of speed limits is not to minimize the risk to cyclists. If that were the case, all road speed limits would be set at 15 MPH and nobody would get anywhere. The purpose of speed limits is to mitigate the risk of cars pulling out in front of you, based on the number of driveway entrances, the number of them that are blind, etc. And occasionally, speed limits are designed to reduce the risk of death if a kid runs out in front of you near a school, where there is likely to be a high concentration of young children.
Note that most deaths by bicycles are on large urban arteries, which you would typically expect to have speed limits of at least 40 MPH. At that rate, a collision has a 50% fatality rate, give or take. There's only one good way to prevent those accidents: get the bicycles off of those major arteries and onto secondary streets where they won't be consistently traveling at a fraction of the prevailing speed of the road. The problem is not the drivers. The problem is that those bicycles shouldn't have been on those roads. They simply aren't capable of achieving the speed necessary to not pose a serious safety risk when used on those larger streets.
Blame is too strong a word, as that implies primary fault. I'm saying that they bear some responsibility. If I break into your house and steal something, your insurance pays for it. If you leave your house unlocked and I walk in and steal something, your insurance probably doesn't. It is the burglar's fault primarily in either case, but in the second situation, you failed to take the expected basic precautions required to ensure the safety of your property, and as such, the insurance company is within its rights to deny the claim because of contributory negligence on your part.
From what I've seen, drivers' attitudes have absolutely nothing to do with the problem. Drivers are almost invariably courteous around bicycles when they see them in time. However, the reality is that driving can be enormously complex, and it is very easy to make mistakes, whether through fatigue, distraction, or simply having too many pieces of information flying at you all at once.
The only people whose attitudes can change in a way that would have a meaningful effect on bicycling safety are the cyclists themselves. Cyclists must consistently avoid getting too close to cars, avoid taking unnecessary risks, always wear attire that makes them easily visible, avoid making sudden stops that can get them rear-ended by cars that are trying to get through traffic lights, and generally assume that nobody else on the road can see them until proven otherwise. Ride defensively. If everyone did this, we would have a lot fewer cycling deaths.
Breaking the speed limit by 5 MPH results in a fairly small increase in the odds of an accident being fatal. Speeding across an intersection on a bicycle without stopping greatly increases the risk of a fatal accident, from zero to 100% if you get hit. And it is fatal accidents we're talking about here. The statistical difference in risk makes that comparison utter crap. Your argument is like calling someone a hypocrite for complaining about the guy doing 120 MPH in a 25 MPH zone simply because that person is known for going two or three MPH over the limit.
I'm guessing their cyclists are more careful, they have fewer drunk drivers and drunk cyclists, they have more bike lanes, and fewer people ride bicycles on major streets there.
American roads have a tendency to be a lot more uneven than roads in Europe, at least from what I've seen. In Europe, most of the roads I saw were almost either highways (no bicycles allowed) or they were two-lane city streets. Here in the U.S., we seem to have a much larger percentage of four-, six-, and eight-lane major city streets, often with no smaller parallel streets that are suitable for bicycles (because they all get cut by highways every two or three miles). The largest percentage of bicycle fatalities happens on those large urban streets.
I blame, at least in part, the "traffic calming" nonsense that has become the norm in lots of cities, where it is impossible to go anywhere on back streets without eventually getting to a point where you're forced out onto one of those major arteries by side streets that don't connect. I also blame the haphazard way in which American highways are allowed to carve a slice right through a city, rather than being elevated or built underground as is common in much of Europe, allowing bicycles to pass over or under it on every existing cross street.
But the biggest problem, IMO, is the steady increase in average commute times. This leads to drivers being more and more fatigued by the time they get off the highway at the end of the day, and surprise, surprise, that's when most of the fatalities happen. We have an epidemic of long commutes in California, brought on in no small part by prop 13 creating a tax disincentive for selling your home and moving closer to your job. This, coupled with poor regional land use planning, has created a situation where about 80% of the roughly 120,000 people living in the greater Morgan Hill/Gilroy/Hollister area commute an hour or more to work in the Silicon Valley because there are almost no tech companies with offices in that area. Same goes for a much smaller (but still sizable) percentage of the greater Santa Cruz area, plus everything from the East Bay all the way up to greater Sacramento.
That's technically correct—the best kind of correct. However, according to New York's statistics:
When you add
Well, it doesn't give you the right to hit the cyclist, if that's what you mean. It does give you the right to tap your horn, or maybe even call the police, if you're feeling particularly vindictive that day. :-)
It is legal in most natural languages, and is quite common in non-imperative sentences. However, for imperative sentences (commands), putting the action before the condition is not the normal way that people speak in most natural languages, precisely because it leads to mistakes. For example, in a high-stress situation, if one person shouts to another "Shoot her if she moves", there's a high probability that the other person will shoot reflexively before even hearing the last half of the sentence.
Similarly, shifting the "if" statement off to the right end of the line makes it much more likely to be missed by someone scanning down the column of text, because people naturally skim down the justified edge of a block of code, rather than the ragged edge. This language design anti-pattern makes Perl code considerably harder to skim than languages that do things in the more typical order of condition followed by action. (To be pedantic, C does provide one statement type—do-while—that puts the action before the condition, but the action is preceded by an additional token that makes it obvious that the action is conditional. This distinction matters.)
If someone uses brackets as delimiters, your code has to also parse character classes correctly.
Even if true, the risks to a cyclist doing something stupid are much greater than the risks to an automobile driver, statistically speaking, because they are much less visible, and they don't have a giant metal box wrapped around them to offer protection. There should be exactly zero cyclists doing these things. If there are enough bad cyclists for us to even notice their poor habits, that's already way too many bad cyclists. If the percentage of cyclists doing these things is comparable to the percentage of motorists doing so, then that is likely to be the cause of a sizable percentage of those cycling deaths, and a negligible percentage of deaths by motorists.
Yes, and bicycles traveling in a traffic lane at speeds significantly below the posted limit are required by law to periodically pull off the road to let cars around, just like any other slow-moving vehicle. Neither bicycles nor cars nor any other vehicle are allowed to block traffic for an extended period of time.
Not really. You are allowed to do so for a short time, but not indefinitely. In particular, most states have slow vehicle laws that require you to pull off the road if you are unable to drive at a speed that does not impede the flow of traffic. The exact laws vary widely from state to state.
California, for example, requires that you pull off the road if there are five or more cars behind you; the same law also applies to slow-moving cars, trucks, etc. (California Vehicle Code section 21656)
This. I would have put it more tactfully, but a thousand times, this.
Almost without fail, cyclists ignore stop signs, ignore traffic lights, turn without signaling, and generally ignore traffic laws willy nilly. It's one thing to make a right turn on a bicycle without stopping; it is quite another to speed past a stop sign, going straight through an intersection where the opposite direction doesn't have any obligation to stop.
The number of serious mistakes made by cyclists in my presence has been climbing steadily over the past decade or so, which suggests that the level of recklessness exhibited by cyclists is increasing at an alarming rate, or at least the number of idiot cyclists is increasing at an alarming rate. You can't have cyclists ignoring the laws of the road and somehow expect drivers of cars to magically take up the slack. Responsibility for cyclists' safety must be shared.
No chance of that, buuuuuuuuuddy.
From my perspective, regular expressions outside the context of a string leads to, among other things, all sorts of fun for IDEs, where they can't properly do brace matching because you have a brace in a regex and they don't know what to do with it. The string syntax is much less problematic because they invariably follow bog-standard language parsing rules, which are much easier to handle than parsing regular expressions (where the rules are inconsistent from one language to the next in subtle ways, with Perl being the biggest mess).
What makes Perl's regular expression engine particularly nasty is that there's more than one way to do it. Don't like slashes as regex relimiters? Why not use brackets? Or parentheses? Or curly braces? The complexity of the parsing rules goes up by a sizable margin every time you add such "handy" features. And the way special characters are handled inside a regular expression depends on context, which makes it a serious pile of hurt.
By contrast, I can parse a regex in a C string in about three lines of code (count the number of double quote marks not preceded by an odd number of backslashes or inside single quotes, and if you get an odd number of double quote marks, you're inside a string), and any brace matching engine that can't handle such simple rules is just sad.
By backwards instructions, I mean that in C, you write this:
if (foo) bar();
whereas in Perl, you write this:
bar() if foo();
That design violates the norms of natural languages, where conditional expressions almost invariably precede the actions that they modify, and as such, leads to many an incorrect assumption when skimming other people's code.
Yeah, I almost made a comment about the vast quantities of PHP code out there, but decided against it because so much of that code probably shouldn't be out there. These days, before I bring in any piece of PHP software, I tend to spend at least a day or two doing a cursory security audit in which I look for SQL injection and shell command argument passing vulnerabilities. I usually find some.
But yeah, at least from a jobs perspective, the community, codebase, support, and ease of hiring are what makes a language valuable. And the more bad code there is out there, the easier it is to get a job fixing all the holes. :-D
Sure. Lots of folks use PHP for writing quick server code, because it makes light work of prototyping things. If you know what you're doing, it is a solid programming language, offering a fairly clean, C-like syntax, without the horror of Perl's backwards instructions and bare regular expressions, the OO bloat of Java, the need to install additional interpreters (problematic on shared hosting services), etc.
You know that Lightroom is still available for sale, right?
When Adobe announced the "Creative Cloud" nonsense, I bought a copy of CS6, upgrading from CS3, knowing that if I didn't act, I would never be able to do so again. At that point, I immediately began phasing out my use of Photoshop. Unfortunately, I haven't created any big new projects since then, and I'm stuck using Photoshop for existing projects because Pixelmator doesn't handle manual text kerning changes correctly on import. But the cover art for my fourth novel will be done entirely in Pixelmator.
You see, Photoshop hasn't added anything I really care about since they added layers and layer effects. There are a few minor enhancements that are nice to have, but I was happy on Photoshop 7. I buy upgrades to Photoshop whenever an OS X upgrade breaks it badly enough. It annoyed me badly enough paying for bug fixes when I was doing it on my own terms. When Adobe tried to push me to a monthly fee schedule for Photoshop, I walked away and haven't looked back.
I still buy Lightroom about every second release (about $40 annually), because unlike Photoshop, its upgrades actually provide tangible benefits—new camera support, face recognition, etc. (Yeah, theoretically Photoshop upgrades technically add new camera support, too, but I've imported RAW files into Photoshop a total of three or four times ever; typically, I start out with the processed output from Lightroom.) The problem with Photoshop is that it is a mature product, and there aren't any cool new features left to add that provide enough benefit to pony up an extra $80 a year to get them.
That's a serious problem for companies like Adobe. You see, they're in a position where they command the market. The only place to go from there is down, and the writing is on the wall. Flash is a failure. Photoshop is rapidly seeing the bottom portion of its market worn away by competition. This leaves Adobe's last remaining market—big graphics shops. Those folks will keep buying Photoshop until they no longer find themselves exchanging Photoshop files with other companies. Adobe is thus bleeding those companies for every possible penny they can before Photoshop eventually fades into obsolescence. Or at least that's what it looks like from where I'm standing.
No, it really doesn't. There aren't too many new shows. There are simply too few shows getting renewed. Every episode that has aired before is "new programming". There's certainly not too much of that. If anything, there's too little. If you watch whatever is on Netflix, you'll run out of new content before you watch TV every evening for two years. After that, you'll be sitting there looking for new content and not finding any that looks interesting to you. Yet the shows that you aren't interested in are probably interesting to other people, so those shows should still continue to exist.
The real problem, unfortunately, is what it has always been: ratings-driven executives who are not content to pay for shows unless they find their audience instantly and are broadly popular. This leads over time to progressively lower-quality programming that dumbs down everything to be "good enough" for the largest possible audience, while being seen as "good" by none of them.
What we need is what we have always needed: executives who have the courage to allow a show the time to find its audience, even if that takes three or four seasons, who are willing to guarantee more than a half season at a time, giving people a reason to think long-term about a show's future. Without that, most TV will continue to be what it has become: crap.
Pretty much. There are a lot of other things that also kind of fall into that category for me—salary, for example. I'm not likely to jump ship for a higher salary (in the absence of other reasons to leave a company), but if you stop paying me a salary, I'm not likely to stick around. :-D
Maybe pay is a problem for IT, but it usually isn't an issue at all for programmers. For example, here in the Bay Area, at roughly $90-$150k per year, the hourly rate would still be better than most jobs even if it required working 80-hour weeks. The problem is that most people can't survive an 80-hour work week for more than a couple of weeks, and even a 40-hour week is horribly inefficient and, frankly, exhausting at times.
The 40-hour work week is optimal for menial tasks that require very little thinking. For a technical workforce that spends most of their day thinking and trying to solve complex problems, workers are most efficient when working six or seven hours per day, not eight, and certainly not 12. Long before they reach the 50-hour mark, they're actually getting less work done per week than somebody working a 35-hour week, because they have less energy and are less focused. In fact, I rather suspect that the optimal work week in tech is somewhere closer to 20 hours, and that even a 35-hour week involves significant loss of efficiency.
What we need is for employers to hire twice as many people, pay them half as much, and work them half as long. Doing so on a broad scale, however, would require some serious changes, particularly in the way we try to attract people to the field. But it should be done, not just because employees would be happier with a better work-life balance, but also because employers would be getting what they paid for instead of only about two-thirds of it.
I mostly agree, except for this one:
I'm currently working for a company that feeds us three days a week. I'm very grateful for that, because we moved from an area that had a few restaurants within easy walking distance to a new location that has none. At our exit off the 101, there's one restaurant that's kind of like a Denny's (not fast at all), one Mexican restaurant, and that's it. None of them are within walking distance (okay, so there's one Mexican food truck, but...), and the shared parking lot for those two restaurants is always full. Other than that, there's no food until you get all the way to the next exit off the 101 in either direction.
Worse, when you do go one exit away in either direction, you similarly find only two or three restaurants, also with inadequate parking. Most of the time, I end up driving three exits north on the 101 to where there are enough restaurants to actually be practical (including a few with drive-thru windows), but that means burning about half of my lunch hour just driving to and from the restaurant.
So yes, it doesn't matter even slightly whether an employer provides food for free, or even provides it on-site, but it matters a great deal whether there is ready access to food within a reasonable walking distance. If there isn't, and if the employer doesn't provide food, it can get rather annoying. :-)
We've seen what happens when you treat unrepentant criminals as outcasts. They remain unrepentant criminals. The recidivism rate in American prisons should be proof enough that such tactics don't work. Or take a look at countries that try to imprison the opposing forces after a civil war. Things never stabilize. Sometimes, forgiveness by others is the first step towards repentance—turn the other cheek, and all that.