You note that "there is no clearly logical next target control" in that tab order. Exactly. I like to on the spur of the moment decide that I'm going to do whatever as fast as I possibly can, with Tab, Ctrl-Tab, Alt-Tab, Shift-Tab altogether if possible. And sometimes if I don't have much open simultaneously the Ctrl-Tab and Alt-Tab work out just nice. But trying to get my keyboard time down to the equivalent of 80 words per minute (i.e., calling a change of focus a 'word') or more, there are times when I am popping up a dialog in an app run by the JVM, or one in which the tab order isn't the bare-bones Explorer dialog (Open/Save) and I often pass up a cyclic edit keystroke sequence that would work fast and haul the mouse over just because I can't anticipate precisely where the focus is going to shift to. I find I run so much different software, and there's always something new to try out; that I thought it might be handy to have a transparent overlay pop up on some designated keypress to show what the focus could shift to on these keys. Not that a novice would need it. Nobody's too dumb to figure stuff like that out on the spur of the moment by trial-and-error, or a good hunch. But when the UI is going the way of the tile-gaucho I tend to wonder, if you don't use the keyboard shortcuts a heck of a lot more, what DO you do? I phrased my query in the tentative way I did because, like you say, it's obvious that there's no standard way of doing things like that, and the mother of invention can bring us what she wants; and outside of that-- Who's going to care? Just me, I guess. I was using VS Express, so I don't have a real big option for personalization or enhancements in the IDE; but what I was thinking of was in general, over the whole gamut of dialogs and controls that have some spatially distinct distribution and ordering that leads to an as-logical-as-we-can-manage tab order; and when it goes beyond the obvious, that's where I'd think having an overlay pop up temporarily might come in handy.

Who's missing buttons and controls under big 7 and big 8 ? Not you ? Well, what if somebody decides to fix our problem by designing us a popup that will diagram, on a mouseover, the tab order of controls on any form or dialog, showing icons with interconnecting arrows unambiguously laying out for us what will happen when you press the Tab key (once, twice, ... ). Too much work for too little pay ? I thought so.

The values appear small or zero; and this has been the case for a good while. It used to be that I'd see they were in the thousands; but not anymore. Isn't that life? The good old days: "Not anymore".

http://www.wonderhowto.com/how-to-make-fake-boogers-with-rubber-cement-329686/[^]

That's exactly right; and the reason I object is because those formulas do not and cannot change. It is at the point where substitutions are made that other dependencies arise; and because of this, I think it only leads to confusion to argue about a metric for the angular quantity. I don't see how complex arithmetic involving expressions of exponentials in terms of the circular functions is going to improve under his (Bob Palais') suggested improvements, either. If you can see this all as some isomorphism in which nobody is going to be able to, or care to, distinguish between present and future methods, show me some proof. I'm listening. But I don't think it's there to prove. How important is it, merely as a matter of heuristics?

Apparently, Bob Palais of the University of Utah wants to prevent the use of the constant Pi in circular measure. He wants to substitute a value, symbolized by the Greek letter Tau, which has a value twice that of Pi; and to base all mathematical expressions on this Tau value rather than on Pi. The trigonometric double-angle formulas, sin(2A) = 2sin(A)cos(A) cos(2A) = cos2(A) - sin2(A) tan(2A) = 2tan(A)/(1 - tan2(A)) would then be replaced by sin(t) = 2sin(t/2)cos(t/2) cos(t) = cos2(t/2) - sin2(t/2) tan(t) = 2tan(t/2)/(1 - tan2(t/2)) and the trigonometric half-angle formulas, sin(A/2) = [+|-]SQRT((1-cos(A))/2) { + if A/2 is in quadrant I or II, - otherwise } cos(A/2) = [+|-]SQRT((1+cos(A))/2) { + if A/2 is in quadrant I or IV, - otherwise } tan(A/2) = [+|-]SQRT((1-cos(A))/(1-cos(A))) { + if A/2 is in quadrant I or III, - otherwise } would then be replaced by sin(t/4) = [+|-]SQRT((1-cos(t/2))/2) { + if t/4 is in quadrant I or II, - otherwise } cos(t/4) = [+|-]SQRT((1+cos(t/2))/2) { + if t/4 is in quadrant I or IV, - otherwise } tan(t/4) = [+|-]SQRT((1-cos(t/2))/(1-cos(t/2))) { + if t/4 is in quadrant I or III, - otherwise } The consequences of this alone don't look too intimidating. So what happens with the sum and product functions under this change? We have, currently, sin(A) + sin(B) = 2sin((1/2)(A+B))cos((1/2)(A-B)) sin(A) - sin(B) = 2cos((1/2)(A+B))sin((1/2)(A-B)) cos(A) + cos(B) = 2cos((1/2)(A+B))cos((1/2)(A-B)) cos(A) - cos(B) = 2sin((1/2)(A+B))sin((1/2)(B-A)) which would be replaced by sin(t/2) + sin(u/2) = 2sin((1/4)(t+u))cos((1/4)(t-u)) sin(t/2) - sin(u/2) = 2cos((1/4)(t+u))sin((1/4)(t-u)) cos(t/2) + cos(u/2) = 2cos((1/4)(t+u))cos((1/4)(t-u)) cos(t/2) - cos(u/2) = 2sin((1/4)(t+u))sin((1/4)(t-u)) and we have the current product formulas, sin(A)sin(B) = (1/2)(cos(A-B) - cos(A+B)) sin(A)cos(B) = (1/2)(sin(A-B) + sin(A+B)) cos(A)sin(B) = (1/2)(sin(A+B) - sin(A-B)) cos(A)cos(B) = (1/2)(cos(A-B) + cos(A+B)) which would be replaced by sin(t/2)sin(u/2) = (1/2)(cos((1/2)(t-u)) - cos((1/2)(t+u))) sin(t/2)cos(u/2) = (1/2)(sin((1/2)(t-u)) + sin((1/2)(t+u))) cos(t/2)sin(u/2) = (1/2)(sin((1/2)(t+u)) - sin((1/2)(t-u))) cos(t/2)cos(u/2) = (1/2)(cos((1/2)(t-u)) + cos((1/2)(t+u))) In these last four formulas, we have in particular to note that the quantities (1/2)(t[+|-]u) would be substituted for by the half-angle formulas noted above which involve the square roots. So

Derek M. Jones, "The New C Standard" Introduction, p.53 "Every language has a culture associated with its use. A culture entails thinking about and doing certain things in a certain way. How and why these choices originally came about may provide some interesting historical context and might be discussed in other sections of this book, but they are generally not relevant to Coding guideline sections." "Culture is perhaps too grand a word for the common existing practices of C developers. Developers are overconfident and insular enough already without providing additional blankets to wrap themselves in. The term existing practice is both functional and reduces the possibility of aggrandizement." "Existing practices could be thought of as a set of assumptions and expectations about how things are done (in C). The term C style is sometimes used to describe these assumptions and expectations. However, this term has so many different meanings, for different developers, in different contexts, that its use is very prone to misunderstanding and argument. Therefore every effort will be made to stay away from the concept of style in this book." "In many ways existing practice is a meme machine. Developers read existing code, learn about the ideas it contains, and potentially use those ideas to write new code. Particular ways of writing code need not be useful to the program that contains them. They only need to appear to be useful to the developer who writes the code, or fit in with a developers preferred way of doing things. In some cases developers do not thoroughly analyze what code to write, they follow the lead of others. Software development has its fads and fashions, just like any other information-driven endeavor."

mohammedghaderian.bp [^] said, quote: Hi everybody, I want to develop an application with c# without .net framework and I want to use win32 dlls? Can I do it? endquote. His Question was down-voted, considered solved, with a final "No." So if I want to "make trouble" by suggesting that CodeDom methods could be used to generate runtime code which references DLLs as imports in C# and that IL might be used directly to circumvent the problem, then what? Although C# would be almost impotent without the .NET runtime, it still might be possible to write a rudimentary and barely functional program that does almost nothing. So, would anyone try to create one? Obviously not. The clincher is that everyone knows that Microsoft has 'sewn up' the .NET framework into a patchwork that nobody can use to do things Microsoft hasn't foreseen and explicitly allowed for. Every security token and access right, and the entire 'managed' code structure, is designed deliberately in order to prevent any strategy which might be 'competitive' to Microsoft from achieving anything with their runtime. So if what you want to do hasn't already been condoned in advance, you can't do it. And everyone knows it. That's why you all gave such a quick answer to mohammedghaderian.bp's question and downvoted it.

I had not, nor have I, said anything at all about countries outside the U.S. in connection with the subject I had written about. If you would read it rather than blindly parading the fact that you haven't read it, you might understand that I WAS NOT REFERRING TO COUNTRIES OUTSIDE OF THE U.S. "Even a simple thought gets tangled when told to someone who has a knot in his head."

Trying to achieve high data density, the molecules would have to be brought close together, and not knowing the radiation spectra or amplitudes, you can't compare the distance with the useful minima we've seen for ferromagnetic storage. And one mol may bit-store reliably in isolation, but this says nothing about what happens in a close-packed matrix of such mols. I haven't read the text of the article this report refers to. If anybody knows where a copy can be found, gratis, please leave a link so I can look at it. Another thing is the decay chains and the idea of what happens to a matrix of these mols when decay occurs. Imagining a one-bit-per-mol data storage scheme leads to thinking of losing one bit of data with every decay. And one-bit-per-X-mols packed into an artificially created domain where all X mols store the same magnetization brings up the question of intra-domain thermal transfers, which although not scary for single-bit storage, might be a threat over a few billion such domains because they would have to produce radiation extending beyond the domain boundaries. I can imagine a problem there with isolating them from your microprocessors and other chips, which will quit working right if they're subjected to ion bombardment. I don't think the U of N would have suggested a data storage use for these mols, though, if something as obvious as that would preclude it; and I don't know enough about it to guess whether the emissive effects of U in this context would be different than in the pure state as we see it in fissionable material. Anybody know about this aspect of it?

Thanks for finding that dead link for me. I probably wouldn't have noticed that for another month, at least.

http://www.codeproject.com/Messages/3882879/When-mathematicians-die.aspx[^] http://www.codeproject.com/Messages/3882768/When-Programmers-Die.aspx[^] http://www.codeproject.com/Messages/3882745/When-physicists-die.aspx[^] http://www.codeproject.com/Messages/3882628/When-chemists-die.aspx[^] When I die, I don't want to go to Hollywood.

Potential new disk storage technology ? A report from the University of Nottingham for you: http://www.nottingham.ac.uk/news/pressreleases/2011/april/breakthroughdatastorage.aspx[^] Comments, anyone?

Re: The Lounge, and some of you. Here's what I have to say about it, in detail: http://cqvis.zxq.net/CMT/POH1.html[^]

It seems to me that they should be required to use SQL queries to collect the data they want from your computer system rather than being allowed to install tasks which run autonomously and without your express consent. If they were required to use SQL, connecting only to specialized provider modules which you install and configure in your system and which only provide data after you have inspected as much of the report they are asking for as you like, much of the privacy problem could be avoided because the individual could assess each data item in accordance with their own standards for personal privacy. To my knowledge, no one has offered to design such a provider, nor have I ever heard of anyone proposing an adequate interface which could support this. If a suitable solution in this respect could be found, it could defer the really tough privacy issues without taking the risk of doing anyone personal damage in the meantime; and there is no sense in hurrying the law when it comes to adapting generalized rulings which would affect constitutional rights and privileges without providing precedents which could be applied fairly in all special cases. Not everyone has an SQL server installed on their local computer, of course; and the operating system administrative management tools are not designed to filter, collate, and publish the underlying details of your operating system's functioning as well as your habitual usage of it, even though system logs and counters are available to the individual possessing administrative privileges (for instance, in the Windows Management Instrumentation objects, if you are running a modern Windows operating system) and can be used to collect a great deal of extremely detailed information even from remote systems using the Remote Procedure Call mechanisms. But the average person is not going to be paid or adequately compensated for their time spent in collecting the information these agents want; and it is beyond the knowledge and expertise of the average computer user to install, configure, maintain, and operate the kind of software instrumentation which this sort of data collection requires. So you're not going to be paid to go to school to learn how to do it, you're not going to be reimbursed for the cost of the software you would need to do it if you did have the time and expertise to deliver what these agents want, and in any event they aren't going to pay you even a nickel here or a dime there for whatever data you collect for them in a situation where yo

This links to my suggested improvements in the design of data structures for fuzzy logic: http://cqvis.zxq.net/MISC/Fuzzy.html[^]

You'll never see this one in DocExplorer. There are some things they can do, and we can't: https://sites.google.com/site/vieldekatt/projects/undocumenteddotnet[^]

The University of Manchester has released a paper about their new microscope, which uses a transparent microsphere as a far-field superlens in an optical microscope. The paper is at this download link: http://www.mace.manchester.ac.uk/aboutus/staff/academic/profile/publications/journal_pdfdownload.php?id=2669[^]

Someone was asking about that a while back: Why is there no scalar equation for a line in three dimensions ? http://www.physicsforums.com/showthread.php?t=165472[^] I think the R-difference (from R3 to R1) in dimensionality accounts for it. That's why there's no form analogous to Ax + By + C that includes a z-coefficient. Any ideas ? The only equation that might come close to fitting the bill is this one: Ax2 + Bxy + Cy2 + Dx + Ey + F = 0 If B2 - 4AC is zero this equation might represent a line, or two lines if less than zero. But this requires a section by a plane of a conic surface; and although it's a scalar equation, it still isn't linear; so I don't think it's suitable.

modified on Monday, March 14, 2011 5:09 PM

The Brits have an idea for "plain-box" cigarettes. They just want to do that because they heard somebody was trying to get programmers hooked on "C++" brand cigarettes.