They are almost by design so generic [...] that they are appropriate for nothing
I find them useful for communication. If my design is similar to a pattern (most patterns are only part of any solution) then I can use the name of the pattern to describe my design more quickly. Often, because a pattern will include several distinct elements, I am able to convey several nuances of a particular design with only one or two words.
Explaining a design in terms of patterns is also a cool way of mentoring junior developers. You can back them out of the details of "this problem" and take a look at the more general situation, what the pattern suggests, and then go back to "this problem" and see how it fits in. Often you'll be able to elicit multiple "a-ha!'s" from the junior.
Back to your point, I find that the best pattern authors understand that they must strike a useful balance between generality and specificity. You still have to design the solution, but the top-level "how the heck do I get started here" is what's described by the pattern. Fowler tends to be better than the GoF at this particular skill.
Of course, getting the most utility out of patterns absolutely requires that you are an experienced designer (or are working with one). This means applying all of the classic design skills, including patterns, as appropriate to solve the problem (and that may mean no use of patterns for a particular problem).
Just start a new public repository with the latest good version. Keep the history, but don't worry too much about making sure that everyone has access. Long term storage is fine.
CM systems improve communication between developers by allowing them to synchronize their work as well as preventing simple developer mistakes from turning into massive code rewrites (but you don't need more than two weeks of history to accomplish these goals). The reasons you usually carry around all of the extra baggage of the old versions is for (1) establishment of legal ownership (copyright information) (2) simultaneous maintenance of multiple versions in the field and (3) to show some history of how you got to where you are.
Legal ownership is important, but you get that by keeping a few backups in your long-term storage. You don't have versions in the field (not of the open-sourced version anyway) so that's a moot point. The "how we got here" argument is also of minimal value as long as someone who knows the code is still around. The knowledge of how things were developed in a decent developer's head will be much easier to use than attempting software archeology on a stale file repository.
First of all, i'd like to make it clear that believing in Intelligent Design or in Creationism does NOT in any way prohibit ideas such as evolution
True, but the proponents of teaching ID in schools pitch it as an alternative to natural selection (one of the fundamental theories for observed evolution), specifically for the explanation as to how and why man appeared on this planet.
And as long as that continues to be true, your clearly correct statement will also be unfortunately irrelevant to the argument at hand.
You need to be careful with your definition of "entertainment".
If "entertainment" includes all of the things we choose to do for the enjoyment of it, then I've got to disagree. Want to teach kids about ecosystems, animal habitats, plant biology, simple thermodynamics, simple geology, and a whole lot more? Go on a camping trip in the mountains (or backwoods... whatever's local) and insist that the GameBoy be left in the car. There are uncountable things to be learned around a campsite, and if they haven't had the thrill of learning burned out of them yet, you'll have a lot of questions to answer...
There exists a vision of education which states that "learning is work" or "learning should be work". In my experience, this is almost entirely bullshit. Most of the things I've really absorbed and retained for the long term were learned when I was interested in a subject and enjoying the process of learning. This is not to say that learning must be made entertaining, but that you will do better building on a natural interest than to try to force the memorization of facts that have no relevance to the learner.
Not that you were advocating that (I choose to believe that you were using "entertainment" as shorthand for "mindless entertainment" -- 90% of television, etc.).
it makes the positions of widgets predictable, so you can interact with your software much faster using a mouse once you learn where things are
Apple's argument for the menu bar at the top of the screen. I find that my mouse skills seem to have developed to compensate for the variable location of widgets. I may be slightly faster on a Mac than on Win XP or Linux, but it's not enough to notice.
using the physical limits of screen edges has advantages for usability, e.g., I know that I can throw my mouse point to top-right and click and I'll get the close button for the current window
Interesting. I'd never noticed that before.
using non-maximised windows adds even more window dressing
Yeah, but I've got a 1600 x 1200 display on this laptop. The window dressing isn't getting in the way of anything I'm trying to see or do so... On lower resolution screens, I can see your point. I've got an older laptop with an 800x600 display that use as a Subversion server. I feel very cramped when using it, and the Gnome/Enlightenment window dressing does feel like a waste of valuable screen area.
it's annoying if you keep accidentally moving or resizing a window, e.g., when trying to select text near the end of a line in an editor
I find that my mouse skills have developed in such a way that this almost never happens. If you're really used to the scrollbar being on the right screen border (so that your movement to grab the scrollbar doesn't need left-right precision), I can understand your frustration.
personally, I find overlapping windows to be confusing and unhelpful
This is the point that has real merit. But it's a highly subjective criteria and all it really tells me is that it's silly to pretend that there's "One Right Way" for everyone to use their computer screens.
Personally, I find overlapping windows to be highly informative and not at all confusing. I can start a build (and keep an eye on it), monitor some other process, and read slashdot all at the same time.
switching the active window based on where the mouse pointer is on most UNIX systems rates among my top-ten most annoying UI mistakes ever
Change the setting from "focus follows mouse" to "click to focus". I haven't used a *nix window manager in ten years that didn't offer that option. Actually, do any of the Gnome integrated window managers even offer "focus follows mouse" any more? I can't remember seeing it the last few times I was screwing around with the settings...
His foundation is the charity he's referring to, and it's already got a huge hunk of his MS money. Well, it's really a meta-charity, but that's how he sees it and his position is defensible. Money that's currently in the foundation is gone. He can't spend it on himself or his kids any more.
Charitable foundations are usually operated so that they can continue making grants and supporting their "worthy causes" indefinitely. I have a friend who is on the board of the Hewlett foundation and another friend who is a fund manager for the Getty foundation. They're in close agreement on how big foundations like to do things.
This means investing carefully and spending a little bit less than the interest brings in. I haven't seen the numbers (and I'm too lazy to look them up right now), but I would guess that the Gates foundation is spending more than 2% of it's endowment each year. Probably more like 6-8%.
Casing is a major problem for people that don't have access to the equipment required for PVC modelling.
How about access to a supermarket? Get a Rubbermaid or Tupperware sandwich container and some standoffs and you're off to the races with a fairly slick case for under $5 ($10-$15 for the Tupperware, but Tupperware has a stronger edge to the bowl part, which can be worth the extra cost).
If you're worried about the "ghetto" look, cut properly sized round holes and use rubber grommets or small metal bulkhead connectors to clean up the edges. If you do a decent job locating the holes, it will look very nice, and with the removeable top, it's quite functional too!
One of the skills I find most useful in amateur robotics is the ability to repurpose everyday items. When I go into Home Depot, I'll start playing with something on just about any aisle and my fiance will ask me, "What kind of a robot part are you going to make from that?" The worst aisles by far are in plumbing. Worst as in hardest for me to get through without stopping multiple times.
Actually, the real problem with the Hindenberg was the decision to stiffen the outer skin with a paint that was essentially hardened kerosene. As in what we use for modern jet fuel.
Hydrogen burns with a pale blue flame. All of the exciting footage showed lots of bright yellow flame from... the burning of the envelope.
It does give a little more lift, but as we all know, it burns.
Two points. First, hydrogen gives twice the lift of helium. A 100% bonus for the same sized envelope. Second, it only burns in the presence of oxygen (or another gaseous oxidizer). If the envelope is made from a nonflammable membrane impermeable to oxygen (any membrane decent at retaining hydrogen is completely impermeable to oxygen), explosions and dramatic flames become vanishingly unlikely.
The Hindenberg had problems, to be sure. IMHO, however, the use of hydrogen wasn't one of them.
Can't join a table to itself in SQL? You have to name the table each time, but it's not very difficult and certainly not impossible. Try something like this:
select parent.name from person as parent, person as child where child.parent_fk = parent.person_pk;
But a company can be a community. In my informed opinion, a company that is also a community has a much better shot of long-term success than a company made up of unmotivated loner mercenaries.
Skills being equal of course. At the end, you can only be a great employee if you both have the skills and can be a member of the community (a good culture fit).
Regards, Ross
Re:The real 90s versus outdated 00s software
on
Java Is So 90s
·
· Score: 2, Interesting
Java's "Write once run everywhere" motto is a joke, and everyone knows it.
"Write once, run anywhere" certainly was oversold. For server-side applications, however, it's awfully close to reality.
We currently test our server-side Java application on Solaris, RHEL, and Win32 (2K, XP, 2003) with JBoss, WebLogic, and WebSphere as the app server with zero changes outside of two configuration fields. All of the internal code that pays attention to the platform differences (and there's more for app server than for hardware/os differences) is in three classes that haven't been touched in almost two years (I just checked).
This isn't truly "run anywhere" as we only use commodity OS's on "PC or better" hardware, but it sure beats porting a C++ program across those same platforms.
[...] Ruby on Rails and the language itself deserves fame. It's well built, flexible, stable, and clearly the best competitor of Python.
I agree. Ruby and Python are rightly taking over light-weight web-application mindshare, but Java and.NET still seem to be more appropriate for more complex systems with big transactional database requirements (and there are a lot of business applications that have these requirements).
You're willing to bet the money of other people on the possibility of an abrupt oil production shortfall, but not your own (or very little of your own)?
You and I have very different understandings as to the purpose of government. I believe that government should use common resources to provide a stable context for commerce and to strike a balance between over and under-regulation towards a goal of maximizing long-term public value. Companies are not in the business of providing a stable context for economic growth because the profit motive dramatically weakens any motivation to maximize long-term public value. As such, a different kind of organization must provide this context, most urgently when destabilizing scenarios have been clearly identified.
Speculatively, you probably don't value government provided stability because you (1) don't believe that government is necessary to have stability (the evidence is strongly against you here) and (2) you haven't experienced a context without a strong stabilizing agent. The fact that government uses force to obtain resources (taxation) and to enforce stability also frightens you because it erodes your self-image of being a completely independent agent in the world. I don't want you to discard your independent self-image, but if you do choose to develop a more informed understanding of yourself and the world around you, I suspect you'll find that collective action through government isn't necessarily an evil thing. It just so happens that governments are made up of humans who covet power and who are corrupted by that power so that governments gradually become "less good".
As an aside, I consider myself a libertarian (which I define as: someone who has done more thinking about the nature of government and ethical uses of force than most Libertarians)
Actually, I have considered building a biodeisel plant when I have money. Two problems:
1) I'm not particularly wealthy and so, at the moment, I can't invest squat in an actual industrial plant unless an unrelated venture that I am invested in happens to strike it big.
2) California doesn't allow for the sale of new deisel cars, which seriously curtails the current market for biodeisel in my vicinity. I suspect that this would change in an oil crisis, but it makes it much harder to defray the costs of the investment today.
I don't see the problem. Oil will become gradually more expensive, cheaper replacements will be substituted, and the economy will change.
A seismic change would not be a gradual change. You assume that the eventual reductions in oil production will be gradual, leading to gradual increases in oil production. Oil demand is not very elastic, so dramatic changes to supply would cause dramatic changes in price (prices could double or triple very quickly if there was a production shortfall of 20%).
Can you come up with any scenarios that result in abrupt reductions in oil production? What if the OPEC countries are radically overestimating the remaining reserves and a production crisis across the middle east happens that reduces OPEC production by 75%? How about multiple simultaneous pipeline disasters?
I'm not reaching any conclusions either way, but I do think it's naive to plan based on an assumption that oil production will follow a gradual decline. Energy is ultimately a foundation of our economy and if that market is disrupted, our economy won't be able to work around that problem without a lot of pain and suffering. The smartest approach to planning is usually to plan for the worst and hope for the best. However, our leaders are planning for the best. If they're wrong, it's going to be a very scary time indeed.
It sounds like you really agree with him but you want to argue anyway.
No, I really do disagree with the original post, but what I disagree with is the "only goal" rhetoric and the implication of valuing profit at any cost.
Conclusive counter-example? Ha! You believe that crap?
Not so much, no. But for them to publically claim for sixty years that shareholders come fourth on the heirarchy of priorities means that it's not going to get you into trouble to have goals other than profit. Since that directly contradicts the original post I was replying to...
Unfortunately, you're showing your ignorance. Let me walk you through it:
Careful there. You might find yourself confused with the kettle, pot.
A Business' sole goal is to maximize profit for its shareholders, and nothing else.
False.
The goal that a business must keep as a top-level goal is to maximize shareholder value. This is not the same as "this quarter's profit" or even "profit" over any time frame (though they eventually become related).
Further, many companies interpret "shareholder value" as stock value over the long-term, which is often at odds with actions that would increase stock value in the short-term.
As a conclusive counter-example, check out Johnson & Johnson's credo. Shareholder value is fourth on that list and it's been below other goals for the past 60 years.
I've driven pretty fast. I once drove a Dodge Viper around a race track and got some pretty wicked speed, hitting about 150mph on the back straight. What didn't I see? Motion blur.
Actually, you don't remember seeing motion blur. But your eyes really did actually see a lot of motion blur and your visual cortex did it's best to figure things out for your concious mind to use at the time (and it figured things out slightly differently for your memory).
Your eyes don't capture a continuous image, each rod and cone accumulates luminance (and for cones, chrominance) information over time and sends a signal that approximates a rolling average value of that intensity.
Your visual cortex has a bunch of different components that try to make sense of this data stream coming from your retina. One component controls what goes into long-term memory and blurry data doesn't make the cut. Another component actually substitutes clear images for blurry images when figuring out what's going on around you... but where does it get clear images?
One of the things that you do unconciously when you're moving is to focus on many different things in your field of view. When you focus on something, your eye is tracking the item and blur is eliminated for that particular item at that moment. When you look somewhere else, the new item is sharp, but the previous item may be outside your fovea (the high-resolution area of your retina) and may be in motion relative to the new item (which your eye is now tracking). The part of your brain responsible for your "model of immediate surroundings" will put the sharp image of the previous item into your visual space where it expects the item to be, transformed for observed motion.
So your memory of your recent race will not recall any motion blurred objects for two reasons. Your brain did it's best to fill in blurry data with approximate sharp data and your memory didn't mark blurry data as "good enough to remember". As a result, you probably can't visualize all of the items along the sides of the track on each lap.
Next time, focus on the car ahead and then try to pay attention to some items on the side of the track for a moment or two. You'll see the motion blur for a second or so and it may be continuous if you're moving fast enough. The rods and cones in your peripheral vision are more susceptible to blur and the parts of your visual cortex that manage your focus take some time to start using information from your peripheral retina.
What the game is doing is pretending that you've got a laser-focus directly ahead, and simulating the blur that your brain would be dealing with from your eyes if the areas outside your focus were further into your peripheral vision. I suspect that this model "draws you in" and results in very quick and very deep immersion. My suspicions are confirmed by the terminology used in the review. When you focus on the blurry items, the blur is obvious (and you know that it's a deliberate artifact), but when you're focusing on driving the car, he says it feels like you're actually there.
SBC is not saying that they'll block websites or parts of the internet. At least, not in this interview they're not.
SBC is saying that they won't provide access to their "last leg" networks for free. As in, Google can't sell broadband service in West L.A. without paying some fee to SBC for the privilege of using the wires that SBC installed and is responsible for maintaining.
The real underlying issue is who decides how much SBC and other broadband hardware owners can charge broadband resellers for the use of their network. SBC would like to charge broadband resellers (including Google) at least as much as they directly charge to consumers so that the resellers can't compete with SBC in the broadband market. The resellers obviously want a price large enough to make a buck of their own (free would be great).
This is where the FCC has been tasked to find a balance. Unfortunately for consumers, their balance has shifted towards SBC and Verizon away from AOL, Earthlink, and Google broadband.
I told you to go back and reread my original statement until it made sense. Since you still don't understand it, you didn't follow my directions. Keep at it. One of these days, you'll be able to follow the discussion with the best of us.
Until then, I'll have a little mercy and explain that Occam's Razor is useful for reducing a list of unfalsified statements, but it is not "how science works." Not even a little bit.
Now go back to my original post as it's still right and you still haven't understood it.
what sprang from horses that arre a different species?
The simplest answer to your question: nothing yet. Hang around for a while and see.
Slightly longer answer: because horses are with us today, they're at the end of the chain of species, not the middle. I'm trying to come up with a non-insulting way to talk about the past vs. the present vs. the future and how since we're at the present, we shouldn't talk about the future as if it's the past... but I'm having a really hard time.
If you really understand even basic issues so poorly and you keep speaking up when your betters are having a discussion, you're going to sound like you learned logic and science (among other things) in Kansas. Best to keep your mouth shut.
"What was our Universe like at a fraction of a second before it started"
What you've got there is a nonsense question, and there are exactly zero scientists trying to figure it out. There are similar questions that a very few cosmologists are thinking about like, "If there is a meta-universe in which our universe appeared, what could be known about it?" but there's not a huge amount of interest in events outside our universe right now.
The question that a great deal more scientists are trying to answer is, "What was our universe like a very tiny fraction of a second after it started?" A great number of theorems to answer that question have been proposed and most have been disproven, or more precisely, are unlikely enough that nobody is seriously considering them any more. What's left are various models about our universe that extend back to 10e-35* seconds after the start of the universe.
And each and every one of those theorems is stated in terms of what they predict and how future observations could invalidate them. I.e. they are falsifiable, not provable.
Regards, Ross
* we may be farther back by now, but that's where we were when I was into cosmology about six years ago. Also, there were theoretical problems when you got to 10e-50 seconds since that's when all of the forces are acting as one force and you need to have the "unified theory" to describe conditions at that point. Since we're only beginning to gain some momentum behind M-theory as a likely candidate, it seems probable that 10e-35 is still where our ability to describe the early universe stops.
Think about how much work has been done in physics that is based on theorms, yet I don't see a huge list of conditions before each paper.
The caveats will frequently be phrased as the conditions or tests that would cause the conclusions to be invalid. And the only caveats you're likely to find will be around the new statements made in that paper. When you build on someone else's work, you leave their caveats to them. They will be present at least once, and then after that, it's assumed that you understood them.
"Scientists, therefore, are a great deal more interested in statements that can be tested (found to be false) than statements that can be proven."
No they are not. They are interested in finding how things are and generally why. They don't say "Theroem A cannot be proven, I won't even bother with it."
You've really missed my point. Go back and reread my statement until it makes sense.
No theorem can be proven. Theorems are not written as provable statements. However, theorems are falsifiable which is a hell of a lot more useful. Science works by observing something suprising, coming up with several explanations for what was observed, then coming up with tests to figure out which of those explanations can't possibly be right. I'll say that again because so few people appear to understand it:
Science works by ruling out mistaken ideas, not by proving some "right" idea.
How about "extremely likely to be correct."? Religions are the only groups that claim to have a line on "The Truth".
Science is the domain of conditional statements, usually starting with a caveat like, "To the best of our knowledge...". Further, science is not concerned with provability, but falsifiability. Scientists know that the only real "proofs" are found in mathematics and are rarely directly applicable in the real world. Scientists, therefore, are a great deal more interested in statements that can be tested (found to be false) than statements that can be proven.
Just looking at the language used (certainty vs. probability) will clearly distingush a religious statement from a scientific statement. The proponents of ID start from a position of certainty and therefore what they say is extremely likely to be religious and not scientific.
Exactly. Software subscriptions can provide more control to the customer, and I've considered them for my own software for exactly that reason.
Simple thought experiment: I assert that my software will provide enormous value for a customer for six years, but my customer only believes that he'll get three years of utility from it. If I'm willing to offer a subscription where the customer pays 10% of the negotiated total price every six months, the customer will pay substantially less if their analysis is better than mine. And if their analysis is wrong, they are getting more utility than they thought they would, which makes the continuing subscription fee easier to justify on an ROI basis.
The difficulty comes with how the price is set and explained. For personal use, the price will need to appear substantially below the best retail price (spread over at least 3-4 years) before it will stop feeling like I'm getting torn a new one. Would I pay $10/month for a personal subscription to MSOffice? Probabaly. $20/month? Probably not. MS site licenses pretty much are subscriptions already, so they've already got a lot of data on what companies can tolerate. Now they need to see if they can figure out what consumers will tolerate in the way of rental costs.
Slashdot Mirror
They are almost by design so generic [...] that they are appropriate for nothing
I find them useful for communication. If my design is similar to a pattern (most patterns are only part of any solution) then I can use the name of the pattern to describe my design more quickly. Often, because a pattern will include several distinct elements, I am able to convey several nuances of a particular design with only one or two words.
Explaining a design in terms of patterns is also a cool way of mentoring junior developers. You can back them out of the details of "this problem" and take a look at the more general situation, what the pattern suggests, and then go back to "this problem" and see how it fits in. Often you'll be able to elicit multiple "a-ha!'s" from the junior.
Back to your point, I find that the best pattern authors understand that they must strike a useful balance between generality and specificity. You still have to design the solution, but the top-level "how the heck do I get started here" is what's described by the pattern. Fowler tends to be better than the GoF at this particular skill.
Of course, getting the most utility out of patterns absolutely requires that you are an experienced designer (or are working with one). This means applying all of the classic design skills, including patterns, as appropriate to solve the problem (and that may mean no use of patterns for a particular problem).
Regards,
Ross
Just start a new public repository with the latest good version. Keep the history, but don't worry too much about making sure that everyone has access. Long term storage is fine.
CM systems improve communication between developers by allowing them to synchronize their work as well as preventing simple developer mistakes from turning into massive code rewrites (but you don't need more than two weeks of history to accomplish these goals). The reasons you usually carry around all of the extra baggage of the old versions is for (1) establishment of legal ownership (copyright information) (2) simultaneous maintenance of multiple versions in the field and (3) to show some history of how you got to where you are.
Legal ownership is important, but you get that by keeping a few backups in your long-term storage. You don't have versions in the field (not of the open-sourced version anyway) so that's a moot point. The "how we got here" argument is also of minimal value as long as someone who knows the code is still around. The knowledge of how things were developed in a decent developer's head will be much easier to use than attempting software archeology on a stale file repository.
Regards,
Ross
First of all, i'd like to make it clear that believing in Intelligent Design or in Creationism does NOT in any way prohibit ideas such as evolution
True, but the proponents of teaching ID in schools pitch it as an alternative to natural selection (one of the fundamental theories for observed evolution), specifically for the explanation as to how and why man appeared on this planet.
And as long as that continues to be true, your clearly correct statement will also be unfortunately irrelevant to the argument at hand.
Regards,
Ross
Education is not and cannot be entertainment.
You need to be careful with your definition of "entertainment".
If "entertainment" includes all of the things we choose to do for the enjoyment of it, then I've got to disagree. Want to teach kids about ecosystems, animal habitats, plant biology, simple thermodynamics, simple geology, and a whole lot more? Go on a camping trip in the mountains (or backwoods... whatever's local) and insist that the GameBoy be left in the car. There are uncountable things to be learned around a campsite, and if they haven't had the thrill of learning burned out of them yet, you'll have a lot of questions to answer...
There exists a vision of education which states that "learning is work" or "learning should be work". In my experience, this is almost entirely bullshit. Most of the things I've really absorbed and retained for the long term were learned when I was interested in a subject and enjoying the process of learning. This is not to say that learning must be made entertaining, but that you will do better building on a natural interest than to try to force the memorization of facts that have no relevance to the learner.
Not that you were advocating that (I choose to believe that you were using "entertainment" as shorthand for "mindless entertainment" -- 90% of television, etc.).
Regards,
Ross
it makes the positions of widgets predictable, so you can interact with your software much faster using a mouse once you learn where things are
Apple's argument for the menu bar at the top of the screen. I find that my mouse skills seem to have developed to compensate for the variable location of widgets. I may be slightly faster on a Mac than on Win XP or Linux, but it's not enough to notice.
using the physical limits of screen edges has advantages for usability, e.g., I know that I can throw my mouse point to top-right and click and I'll get the close button for the current window
Interesting. I'd never noticed that before.
using non-maximised windows adds even more window dressing
Yeah, but I've got a 1600 x 1200 display on this laptop. The window dressing isn't getting in the way of anything I'm trying to see or do so... On lower resolution screens, I can see your point. I've got an older laptop with an 800x600 display that use as a Subversion server. I feel very cramped when using it, and the Gnome/Enlightenment window dressing does feel like a waste of valuable screen area.
it's annoying if you keep accidentally moving or resizing a window, e.g., when trying to select text near the end of a line in an editor
I find that my mouse skills have developed in such a way that this almost never happens. If you're really used to the scrollbar being on the right screen border (so that your movement to grab the scrollbar doesn't need left-right precision), I can understand your frustration.
personally, I find overlapping windows to be confusing and unhelpful
This is the point that has real merit. But it's a highly subjective criteria and all it really tells me is that it's silly to pretend that there's "One Right Way" for everyone to use their computer screens.
Personally, I find overlapping windows to be highly informative and not at all confusing. I can start a build (and keep an eye on it), monitor some other process, and read slashdot all at the same time.
switching the active window based on where the mouse pointer is on most UNIX systems rates among my top-ten most annoying UI mistakes ever
Change the setting from "focus follows mouse" to "click to focus". I haven't used a *nix window manager in ten years that didn't offer that option. Actually, do any of the Gnome integrated window managers even offer "focus follows mouse" any more? I can't remember seeing it the last few times I was screwing around with the settings...
Regards,
Ross
His foundation is the charity he's referring to, and it's already got a huge hunk of his MS money. Well, it's really a meta-charity, but that's how he sees it and his position is defensible. Money that's currently in the foundation is gone. He can't spend it on himself or his kids any more.
Charitable foundations are usually operated so that they can continue making grants and supporting their "worthy causes" indefinitely. I have a friend who is on the board of the Hewlett foundation and another friend who is a fund manager for the Getty foundation. They're in close agreement on how big foundations like to do things.
This means investing carefully and spending a little bit less than the interest brings in. I haven't seen the numbers (and I'm too lazy to look them up right now), but I would guess that the Gates foundation is spending more than 2% of it's endowment each year. Probably more like 6-8%.
Regards,
Ross
Good link. I learned a lot. Looks like I will have to stop repeating that theory as fact.
Thanks,
Ross
Casing is a major problem for people that don't have access to the equipment required for PVC modelling.
How about access to a supermarket? Get a Rubbermaid or Tupperware sandwich container and some standoffs and you're off to the races with a fairly slick case for under $5 ($10-$15 for the Tupperware, but Tupperware has a stronger edge to the bowl part, which can be worth the extra cost).
If you're worried about the "ghetto" look, cut properly sized round holes and use rubber grommets or small metal bulkhead connectors to clean up the edges. If you do a decent job locating the holes, it will look very nice, and with the removeable top, it's quite functional too!
One of the skills I find most useful in amateur robotics is the ability to repurpose everyday items. When I go into Home Depot, I'll start playing with something on just about any aisle and my fiance will ask me, "What kind of a robot part are you going to make from that?" The worst aisles by far are in plumbing. Worst as in hardest for me to get through without stopping multiple times.
Regards,
Ross
Actually, the real problem with the Hindenberg was the decision to stiffen the outer skin with a paint that was essentially hardened kerosene. As in what we use for modern jet fuel.
Hydrogen burns with a pale blue flame. All of the exciting footage showed lots of bright yellow flame from... the burning of the envelope.
It does give a little more lift, but as we all know, it burns.
Two points. First, hydrogen gives twice the lift of helium. A 100% bonus for the same sized envelope. Second, it only burns in the presence of oxygen (or another gaseous oxidizer). If the envelope is made from a nonflammable membrane impermeable to oxygen (any membrane decent at retaining hydrogen is completely impermeable to oxygen), explosions and dramatic flames become vanishingly unlikely.
The Hindenberg had problems, to be sure. IMHO, however, the use of hydrogen wasn't one of them.
Regards,
Ross
Can't join a table to itself in SQL? You have to name the table each time, but it's not very difficult and certainly not impossible. Try something like this:
select parent.name from person as parent, person as child
where child.parent_fk = parent.person_pk;
Regards,
Ross
Dude,
A company is not a country.
But a company can be a community. In my informed opinion, a company that is also a community has a much better shot of long-term success than a company made up of unmotivated loner mercenaries.
Skills being equal of course. At the end, you can only be a great employee if you both have the skills and can be a member of the community (a good culture fit).
Regards,
Ross
Java's "Write once run everywhere" motto is a joke, and everyone knows it.
.NET still seem to be more appropriate for more complex systems with big transactional database requirements (and there are a lot of business applications that have these requirements).
"Write once, run anywhere" certainly was oversold. For server-side applications, however, it's awfully close to reality.
We currently test our server-side Java application on Solaris, RHEL, and Win32 (2K, XP, 2003) with JBoss, WebLogic, and WebSphere as the app server with zero changes outside of two configuration fields. All of the internal code that pays attention to the platform differences (and there's more for app server than for hardware/os differences) is in three classes that haven't been touched in almost two years (I just checked).
This isn't truly "run anywhere" as we only use commodity OS's on "PC or better" hardware, but it sure beats porting a C++ program across those same platforms.
[...] Ruby on Rails and the language itself deserves fame. It's well built, flexible, stable, and clearly the best competitor of Python.
I agree. Ruby and Python are rightly taking over light-weight web-application mindshare, but Java and
Regards,
Ross
You're willing to bet the money of other people on the possibility of an abrupt oil production shortfall, but not your own (or very little of your own)?
You and I have very different understandings as to the purpose of government. I believe that government should use common resources to provide a stable context for commerce and to strike a balance between over and under-regulation towards a goal of maximizing long-term public value. Companies are not in the business of providing a stable context for economic growth because the profit motive dramatically weakens any motivation to maximize long-term public value. As such, a different kind of organization must provide this context, most urgently when destabilizing scenarios have been clearly identified.
Speculatively, you probably don't value government provided stability because you (1) don't believe that government is necessary to have stability (the evidence is strongly against you here) and (2) you haven't experienced a context without a strong stabilizing agent. The fact that government uses force to obtain resources (taxation) and to enforce stability also frightens you because it erodes your self-image of being a completely independent agent in the world. I don't want you to discard your independent self-image, but if you do choose to develop a more informed understanding of yourself and the world around you, I suspect you'll find that collective action through government isn't necessarily an evil thing. It just so happens that governments are made up of humans who covet power and who are corrupted by that power so that governments gradually become "less good".
As an aside, I consider myself a libertarian (which I define as: someone who has done more thinking about the nature of government and ethical uses of force than most Libertarians)
Regards,
Ross
Actually, I have considered building a biodeisel plant when I have money. Two problems:
1) I'm not particularly wealthy and so, at the moment, I can't invest squat in an actual industrial plant unless an unrelated venture that I am invested in happens to strike it big.
2) California doesn't allow for the sale of new deisel cars, which seriously curtails the current market for biodeisel in my vicinity. I suspect that this would change in an oil crisis, but it makes it much harder to defray the costs of the investment today.
Regards,
Ross
I don't see the problem. Oil will become gradually more expensive, cheaper replacements will be substituted, and the economy will change.
A seismic change would not be a gradual change. You assume that the eventual reductions in oil production will be gradual, leading to gradual increases in oil production. Oil demand is not very elastic, so dramatic changes to supply would cause dramatic changes in price (prices could double or triple very quickly if there was a production shortfall of 20%).
Can you come up with any scenarios that result in abrupt reductions in oil production? What if the OPEC countries are radically overestimating the remaining reserves and a production crisis across the middle east happens that reduces OPEC production by 75%? How about multiple simultaneous pipeline disasters?
I'm not reaching any conclusions either way, but I do think it's naive to plan based on an assumption that oil production will follow a gradual decline. Energy is ultimately a foundation of our economy and if that market is disrupted, our economy won't be able to work around that problem without a lot of pain and suffering. The smartest approach to planning is usually to plan for the worst and hope for the best. However, our leaders are planning for the best. If they're wrong, it's going to be a very scary time indeed.
Regards,
Ross
It sounds like you really agree with him but you want to argue anyway.
No, I really do disagree with the original post, but what I disagree with is the "only goal" rhetoric and the implication of valuing profit at any cost.
Conclusive counter-example? Ha! You believe that crap?
Not so much, no. But for them to publically claim for sixty years that shareholders come fourth on the heirarchy of priorities means that it's not going to get you into trouble to have goals other than profit. Since that directly contradicts the original post I was replying to...
Regards,
Ross
Unfortunately, you're showing your ignorance. Let me walk you through it:
Careful there. You might find yourself confused with the kettle, pot.
A Business' sole goal is to maximize profit for its shareholders, and nothing else.
False.
The goal that a business must keep as a top-level goal is to maximize shareholder value. This is not the same as "this quarter's profit" or even "profit" over any time frame (though they eventually become related).
Further, many companies interpret "shareholder value" as stock value over the long-term, which is often at odds with actions that would increase stock value in the short-term.
As a conclusive counter-example, check out Johnson & Johnson's credo. Shareholder value is fourth on that list and it's been below other goals for the past 60 years.
Regards,
Ross
I've driven pretty fast. I once drove a Dodge Viper around a race track and got some pretty wicked speed, hitting about 150mph on the back straight. What didn't I see? Motion blur.
Actually, you don't remember seeing motion blur. But your eyes really did actually see a lot of motion blur and your visual cortex did it's best to figure things out for your concious mind to use at the time (and it figured things out slightly differently for your memory).
Your eyes don't capture a continuous image, each rod and cone accumulates luminance (and for cones, chrominance) information over time and sends a signal that approximates a rolling average value of that intensity.
Your visual cortex has a bunch of different components that try to make sense of this data stream coming from your retina. One component controls what goes into long-term memory and blurry data doesn't make the cut. Another component actually substitutes clear images for blurry images when figuring out what's going on around you... but where does it get clear images?
One of the things that you do unconciously when you're moving is to focus on many different things in your field of view. When you focus on something, your eye is tracking the item and blur is eliminated for that particular item at that moment. When you look somewhere else, the new item is sharp, but the previous item may be outside your fovea (the high-resolution area of your retina) and may be in motion relative to the new item (which your eye is now tracking). The part of your brain responsible for your "model of immediate surroundings" will put the sharp image of the previous item into your visual space where it expects the item to be, transformed for observed motion.
So your memory of your recent race will not recall any motion blurred objects for two reasons. Your brain did it's best to fill in blurry data with approximate sharp data and your memory didn't mark blurry data as "good enough to remember". As a result, you probably can't visualize all of the items along the sides of the track on each lap.
Next time, focus on the car ahead and then try to pay attention to some items on the side of the track for a moment or two. You'll see the motion blur for a second or so and it may be continuous if you're moving fast enough. The rods and cones in your peripheral vision are more susceptible to blur and the parts of your visual cortex that manage your focus take some time to start using information from your peripheral retina.
What the game is doing is pretending that you've got a laser-focus directly ahead, and simulating the blur that your brain would be dealing with from your eyes if the areas outside your focus were further into your peripheral vision. I suspect that this model "draws you in" and results in very quick and very deep immersion. My suspicions are confirmed by the terminology used in the review. When you focus on the blurry items, the blur is obvious (and you know that it's a deliberate artifact), but when you're focusing on driving the car, he says it feels like you're actually there.
Regards,
Ross
SBC is not saying that they'll block websites or parts of the internet. At least, not in this interview they're not.
SBC is saying that they won't provide access to their "last leg" networks for free. As in, Google can't sell broadband service in West L.A. without paying some fee to SBC for the privilege of using the wires that SBC installed and is responsible for maintaining.
The real underlying issue is who decides how much SBC and other broadband hardware owners can charge broadband resellers for the use of their network. SBC would like to charge broadband resellers (including Google) at least as much as they directly charge to consumers so that the resellers can't compete with SBC in the broadband market. The resellers obviously want a price large enough to make a buck of their own (free would be great).
This is where the FCC has been tasked to find a balance. Unfortunately for consumers, their balance has shifted towards SBC and Verizon away from AOL, Earthlink, and Google broadband.
Regards,
Ross
I told you to go back and reread my original statement until it made sense. Since you still don't understand it, you didn't follow my directions. Keep at it. One of these days, you'll be able to follow the discussion with the best of us.
Until then, I'll have a little mercy and explain that Occam's Razor is useful for reducing a list of unfalsified statements, but it is not "how science works." Not even a little bit.
Now go back to my original post as it's still right and you still haven't understood it.
Ross
what sprang from horses that arre a different species?
The simplest answer to your question: nothing yet. Hang around for a while and see.
Slightly longer answer: because horses are with us today, they're at the end of the chain of species, not the middle. I'm trying to come up with a non-insulting way to talk about the past vs. the present vs. the future and how since we're at the present, we shouldn't talk about the future as if it's the past... but I'm having a really hard time.
If you really understand even basic issues so poorly and you keep speaking up when your betters are having a discussion, you're going to sound like you learned logic and science (among other things) in Kansas. Best to keep your mouth shut.
Regards,
Ross
"What was our Universe like at a fraction of a second before it started"
What you've got there is a nonsense question, and there are exactly zero scientists trying to figure it out. There are similar questions that a very few cosmologists are thinking about like, "If there is a meta-universe in which our universe appeared, what could be known about it?" but there's not a huge amount of interest in events outside our universe right now.
The question that a great deal more scientists are trying to answer is, "What was our universe like a very tiny fraction of a second after it started?" A great number of theorems to answer that question have been proposed and most have been disproven, or more precisely, are unlikely enough that nobody is seriously considering them any more. What's left are various models about our universe that extend back to 10e-35* seconds after the start of the universe.
And each and every one of those theorems is stated in terms of what they predict and how future observations could invalidate them. I.e. they are falsifiable, not provable.
Regards,
Ross
* we may be farther back by now, but that's where we were when I was into cosmology about six years ago. Also, there were theoretical problems when you got to 10e-50 seconds since that's when all of the forces are acting as one force and you need to have the "unified theory" to describe conditions at that point. Since we're only beginning to gain some momentum behind M-theory as a likely candidate, it seems probable that 10e-35 is still where our ability to describe the early universe stops.
Think about how much work has been done in physics that is based on theorms, yet I don't see a huge list of conditions before each paper.
The caveats will frequently be phrased as the conditions or tests that would cause the conclusions to be invalid. And the only caveats you're likely to find will be around the new statements made in that paper. When you build on someone else's work, you leave their caveats to them. They will be present at least once, and then after that, it's assumed that you understood them.
"Scientists, therefore, are a great deal more interested in statements that can be tested (found to be false) than statements that can be proven."
No they are not. They are interested in finding how things are and generally why. They don't say "Theroem A cannot be proven, I won't even bother with it."
You've really missed my point. Go back and reread my statement until it makes sense.
No theorem can be proven. Theorems are not written as provable statements. However, theorems are falsifiable which is a hell of a lot more useful. Science works by observing something suprising, coming up with several explanations for what was observed, then coming up with tests to figure out which of those explanations can't possibly be right. I'll say that again because so few people appear to understand it:
Science works by ruling out mistaken ideas, not by proving some "right" idea.
Regards,
Ross
[...] does not make evolution the "truth".
How about "extremely likely to be correct."? Religions are the only groups that claim to have a line on "The Truth".
Science is the domain of conditional statements, usually starting with a caveat like, "To the best of our knowledge...". Further, science is not concerned with provability, but falsifiability. Scientists know that the only real "proofs" are found in mathematics and are rarely directly applicable in the real world. Scientists, therefore, are a great deal more interested in statements that can be tested (found to be false) than statements that can be proven.
Just looking at the language used (certainty vs. probability) will clearly distingush a religious statement from a scientific statement. The proponents of ID start from a position of certainty and therefore what they say is extremely likely to be religious and not scientific.
Regards,
Ross
Exactly. Software subscriptions can provide more control to the customer, and I've considered them for my own software for exactly that reason.
Simple thought experiment: I assert that my software will provide enormous value for a customer for six years, but my customer only believes that he'll get three years of utility from it. If I'm willing to offer a subscription where the customer pays 10% of the negotiated total price every six months, the customer will pay substantially less if their analysis is better than mine. And if their analysis is wrong, they are getting more utility than they thought they would, which makes the continuing subscription fee easier to justify on an ROI basis.
The difficulty comes with how the price is set and explained. For personal use, the price will need to appear substantially below the best retail price (spread over at least 3-4 years) before it will stop feeling like I'm getting torn a new one. Would I pay $10/month for a personal subscription to MSOffice? Probabaly. $20/month? Probably not. MS site licenses pretty much are subscriptions already, so they've already got a lot of data on what companies can tolerate. Now they need to see if they can figure out what consumers will tolerate in the way of rental costs.
Regards,
Ross