Problem here is that, to be effective, you'd want a reasonable difference in 'altitude' between the 'corners' to make this effective, but to maintain a similar orbit, objects with lower altitude must have higher speeds. That is, your low-altitude 'corners' will tend to speed away from the high altitude corners. The end-result of that happening is left as an exercise to the reader.
I have to agree with this. Just about every time I have some problem that I cannot immediately understand and fix, I'll try a few different directions (usually at the same time): find a HOWTO page, do a google search, and jump on some relevant IRC channel to talk to a real person with real experience. Sadly, this last option just hasn't worked for me in over a year. I had a minor disk failure other day that caused, among other things I finally had working again, a problem in evolution. The only responses I got (at least on the servers I was trying) was in #gnome, and the response was along the lines of "that suck-@ss ximian sh|t ain't got no place on a REAL gnome system". Yeah, it was real helpful, eh? I understand the parent poster's problem and plea. If you don't have anything helpful to say, then STFU. *shrug*
Though I haven't read through the article as yet, I do find myself occasionally (and more frequently, of late) almost consider "going back" myself. I've been on Linux for, hmm.., four years now. I have stuck with my distribution throughout, and have really only bought the CD's to upgrade twice. (That since people are saying "well, if he didn't chase every upgrade, he wouldn't feel that way.) I try to leave my system alone; I'm quite adept at causing myself plenty of grief and stress w/o changing kernels, distro's, or the like. That stress I do have in "dealing" with Linux, granted, stem from my messing with other stuff: NFS exporting a network directory, poking at NIS, poking at NTP, getting wine to compile/work, and other Unix/Linux things. Don't get me wrong: I love the command line; I grew up on (MS)DOS before I knew, much less had, windows 3.x, so the command line is comforting to have around. On the other hand, when my NAT'ed boxen stopped resolving names via DNS other day, it took over an hour of pouring back through man pages, HOWTO's, and/etc/foo.conf files to figure out that/etc/sysconfig/??/forward_ip4 was now "0" instead of "1". (And I'm not sure just what happened to cause that.) Dumb little thing, sure, and granted when I understood what the problem was, I knew what else to go looking for. Thing is, I just can't find enough time in the day to fully comprehend every nuance of the underlying structure that makes Linux (and perhaps UNIX) what it is. Who's failing is that? Mine, perhaps, if anyones.
Anyway, those times I get the tempting voice to "go back", I remind myself that, more than the above issues I inflict upon myself, I stay with Linux since the philosophy suits me. I don't pour through source code myself, but I trust there are plenty of you out there who can and do. The whole open-ness of approach, for me, offsets the problems I have in the preceding paragraph as I'm not that worried about having "modify-system-at-will" stuff secretly lurking under the surface of the OS. If I was to switch back over, I'd just have stress associated with untrustworthy computing practices. I think for the near term I'll stick with that stress which is mainly self-imposed.
Re:the only real windowmanager
on
Gnome 2.0 RC1
·
· Score: 1
"AquaGraphite" themes (now that themes.org site has returned) for gtk and sawfish get a similar overall look. I have slightly hacked versions of both running on my desktop. Very slick appearance. Can probably find the WM theme for your WM if you look. [shrug]
I do. I even have a fairly nice one and could probably turn up a photocopy of an original (50's-era) guide book for one.
For the record, I'm 28; getting, learning, and understanding a slide rule was a personal curiosity. I did all of high school, college, and graduate school with a one-line, no-graphing calculator and did just fine that way, so I'm a bit shocked (though not necessarily surprised) about the comments of dependence on graphing calculators. I mean... "circle"... its round... "ellipse"... its round and oblong... sheesh. I'll go kick back in my crotchety chair now.:-/
... but lets be honest here; people are stupid, so there will be millions left vulnerable no matter what MS does because those millions are too ignorant to protect themselves.
Interesting (and depressing) thing occured last week here at work. Couple of us "linux" nuts were talking to a "windows" nut about the need to at least keep up on system patches, etc. Now, he's a very brilliant engineer and can get around in a computer system more so than you'd otherwise think when you heard his reply: "I don't care. I really don't." This even after we explained it wasn't about someone taking stuff from his system as much as it was about someone using his system to attack others. He is smart enough to do it, understands the repurcussions of not doing it, and still doesn't care. It was at this point that the couple pro-linux nuts in the discussion realized that there was honestly nothing we could say to move his opinion.
In other words, you have to figure that, as many clueless people are not patching their systems, our co-worker represents a large number of quite saavy people that are completely apathetic to wanting to be bothered. They don't have the interest to want to take the time; we can't reach these people using fear or logic. How, then, do we protect ourselves?
News of any Big Name (tm) 3D CAD software for Linux is huge in the engineering field. Where I work (as engineer in a large aerospace-industry company with infrequent-but-present CAD needs), we have been using Unigraphics on Solaris boxes. For whatever bone-head reason, couple years ago the entire company decided to begin replacement of several thousand Sun boxes with Dell/Intel solutions running W2K. The real geeks here amongst us made a good cry for putting Linux on the machines instead (save money, yes you can still get support, we still have Unix, no productivity drops, etc.) but, among many other "reasons" was one of the strongest motivators that Unigraphics (UG) didn't work under Linux but did under windows. Now, while the company deployment of and reliance on "UG" is large (so much so that there's no hope of switching to anything else; there is far too much inertia at this point), us pro-Linux guys didn't have too much come back for the pro-MS upper management point that there wasn't any top-of-the-line CAD packages for Linux. (As an aside for those interested, we are expected to be a fully-MS-campus by end of the year; already the amount of things that just don't work, courtesy of unknown coded limits in W2K and/or its basic non-Unix-ness, is enough to make you either laugh out loud or go cry in a corner.)
I haven't seen the comment you are referring to (nor am I that interested, really), but I did just want to raise a very generic and non-personal point that whether something is funny or not doesn't have much to do with whether something is on-topic or not; I've modded an otherwise decent humor attempt as off-topic before... yes, I got the [attempted?] joke, but there wasn't any context that related it to the posting or parent comment(s). (Yeah, the fact that this is off-topic with respect to the submission isn't lost on me either. Guess those are the breaks.) For my two cents, if someone is going to be funny, they at least should be passably on-topic about it.
I am not really into the "get / listen to music online" scene (so keep this in mind as you read the next paragraph). Steadily over the past few years, I have tapered off buying CD's to almost nil, and I've sworn off the local radio stations (not so much their CC tie-ins, but that their playlists are insanely short; I heard the same song going to work three days in the same week... and it wasn't a great song to begin with. ARGH).
So now I'm wondering... would it be feasible to setup a slashdot-style online music forum for independent artists to submit music tracks to, such that members of forum could categorize, comment on, rate/rank those submissions to get streamed online? Wouldn't need much of a DJ (per se) as much as a few people to keep the forum running. Heck, if it works, you could probably let members design their on CD for burning for, say, $5 to get sent to them, the proceeds of which go to the forum upkeep as well as the artists themselves. Granted it would take a little while to work up a good user base, but would something like this work? Pros and/or cons?
I used to use a spreadsheet years ago before juggling the cells/formulas across many accounts (1 checking, 2 savings), multiple school and car loans, credit accounts, and now-a-days, 401k tracking, became a pain and I switched over to *cough*Money*cough*. MS sent me a 90-day freebie trial of the '98 version, and that was pretty much that; I've been using it since. (It is the *one* piece of software that I can actually give MS full props for, even if it is (or seems) just a glorified MSHTML app with a database backend. )
So, anyway, while I stipulate that a spreadsheet might work, the main thread was with respect to financial packages.
I have actually poked at each release of GnuCash, but consistently it lacks one feature that I am rather addicted to: scheduling income and expenses, and combining this into a budget forecast for the next N months. With something like this, correctly setup, I not only know how much money I have an any account *right now*, but I will have a reasonable ball-park figure for how much money I will have in three months, six months, etc. A nifty line-plot is handy to see where, when, and how bad the next "low point" is going to be, and as necessary I can adjust funds to deal with it gracefully before it has a chance to bite my sorry a$$. Very useful planning tool that, now, I cannot do without. This is the one single feature keeping a '98 partition hanging around my house.
Now, I'm not too bad writing bits of code and what-not (it's a tangential part of my day job), and I appreciate that, to some extent, linux money applications can be scripted and stuff; maybe I could roll my own forecaster this way, but I really don't want to feel like I need to kludge together such a relatively 'big' feature when I don't have the time and interest after getting home. (Maybe it's just me, and, yes, I'm a bit lazy once I'm off the clock.;)
I probably haven't looked into all possible alternatives for a linux-based financial program, but so far I haven't noticed one that really handles this.
Since the/.'ed page appears to be posted in multiple copies already, how about a couple viewpoints from someone whose mom actually tried this?
This definitely has to be involved in the whole Herbalife Independent Distributor scheme. Just to satisfy your curiosity, yes it is a MLM. You sign up under another distributor (the person that lets you in on what the product actually is, etc.) and they get a small cut off of your sales. Quite pyramid-like, since the best profits come not by selling the product itself, but by recruiting other independent distributors underneath you. (Otherwise, you probably *could* make money, but not the amounts you hear glamorized).
Now, the rest of my information is, to be fair, several years out of date. I can only relay how it was all seemingly setup in the early-mid '90s. Basically, the top-level Herbalife company (corporation? I wasn't too sure exactly) works very hard to keep a legally viable separation distance from all the indenpendent distributors. (This is perhaps why the company doesn't get in trouble for the street spam, since, as another poster pointed out, the ID's are the ones putting them up.) This went so far as prohibiting ID's from using any sort of Herbalife company logo on business cards to maintain a clear separation. (IN fact, the letter of law at the time was that you were only allowed to say "Independent Herbalife Distributor" in a plain font face. Nothing else.)
The signs you see all over creation that make no reference to the product or company arise from this same seeming need to maintain separation. You, as an ID, are not allowed to advertise what the product is or might be; you have to entice interest in the customer, and after *they* make the initial contact can you explain what it all is.
They even went so far as laying down policy on accepted forms of advertisement with respect to various media. Mom, trying to play fair and honest with them, inquired about constructing a store-front web site. They gave consistent, repeated, flat "No"s when she described it. They only saw a web site as a means to advertise for potential business, not as a possible means to *DO* business. (Remember the time frame involved here: early-mid '90s.) The company's own site was the only place permitted to have logo's, product information, etc., and even *they* themselves didn't have a mechanism to take money. (Since, I guess, only the ID's were allowed to move merchandise and collect cash directly.) It was weird.
Now, it got interesting to see that the only people making cash were the select few people at the top of the pyramid. The dominant exceptions are people that are providing these signs for ID's to put up. (Wonder why they all look the same in every city? There's a place that an ID order's them from.) In store product demo's and the like were also something for and ID to spend money on. (Physical stores. You could have the product displayed all over the counter, but don't you dare have any mention of Herbalife in your store window for passer-by's to see.)
Needless to say, it is in fact quite hard to actually set this up as a lucrative business. You really need a big downstream pyramid of ID's underneath you, but then they are competing against each other (and you, perhaps) to get the same thing, so any normal size town (or sub-section thereof) can get saturated by various ID's competing for either customers or other ID's). Mom worked at this for over a year, and I don't think she ever really turned a dime over in her favor.
(Oh, and if any company Nazi's start looking for her as a result of this post, she has a different last name than me.:P )
I was taking a few numerical methods classes in my engineering classwork not *too* long ago. At the same time, I was trying to learn how to apply OO concepts to new problems as well as learn the finer points of Java. So I started doing my various homework assignments and projects in an OO fashion with Java. Maybe a few ideas can be a starting point. (Bear in mind what I did was somewhat crude, but it was personal homework assignments, not full-on production code releases. I'm also not advocating use Java for high-performance-demanding code per se; I was only killing several birds with a single stone;)
For a finite element methods class, I created classes to represent various element types. (Bars, thin triangles, thin quads, etc.) There was also a "matrix" object. Now, each element type object would have a method to populate its coefficients into the matrix, and the matrix would have a method with which to solve itself. Now, the implementation of this particular method is quite procedural if you only look at that part; the OO comes about providing a more abstracted method of assembling the matrix, dealing with neighbor-to-neighbor effects, and the like. Ultimately, I had different classes to solve matrix objects by different techniques, so I could easily change how the problem was being solved by just tossing the matrix object to a different object/method (which was already fairly polished in the more general numerical methods class).
For things similar to CFD solutions (a la your finite volume question), you can likely use OO to organize your flow domain into different region objects, each of which has some notional grid object (or perhaps collection of volume objects). The methods in these classes can give you the ability, again, to assemble matrix objects which can be passed off to a particular solver. (Again, this implementation of the solver will likely appear quite procedural; I didn't carry the idea too much farther before gradution to know if the various matrix inversion techniques could be cleverly OO-ized.)
So in other words, use OO where the principles can help you. It does take some amount of thought to redesign your approach to the problem. Hope this gives a few things to think about.
Color me confused, but the whole behavior of a sonic boom is that "information" (such as the length of the craft) cannot travel upstream across shock waves, so the nose "boom" won't know how long the aircraft is. Besides, it is the Mach number of the craft in flight that determines the angle of the primary nose shock, not the geometry of the nose (or remainder of the craft) itself. So I figure the story has to be horribly mangled from the original source at this point to imply that stretching the craft can even "flatten" the generated shock waves. Very little to see here folks.
First off, let me say I am a supporter of Java as a language and platform. I started with a working knowledge of C before C++ or Java entered the mainstream consciousness, but I found picking up C++ to be troublesome whereas picking up Java was a snap. Then, having a good knowledge of C (for the syntax) and Java (for the OO concepts) making the final transition to C++ was a snap.
Personally, I find these accusations of Java being clunky to be interesting, as I see Java as being more clean and consistent with its class API than what I've seen in C++, and ever time I try to accomplish something that was trivial in Java into C++, I come up against syntactical brick walls. (Okay, I will grant that I don't *know* C++ nearly as deeply as I do Java).
In fact, I have a RedHat box at home that is serving various utility services to my two other boxes (one's for the wife running '98), and it is all being done with Java since client/server code is hardly any more difficult that printing "hello world". Speed is not a problem once the virtual machine loads up the classes, and this is what I think most people perceive as slowness. (And it also bears noting that load-up is a bit faster with whatever JRE I have under linux (IBM, IIRC) than with Sun's JRE under windows.
Anyway, I digress. If the course material is mainly about the OO approach to code design, then I would favor Java over C++ just since Java's syntax seems to get less in the way, and thus seems to distract less from the matter at hand. In fact, in less than a week, I am presenting OOP to my fellow engineers (many of them "greybeards" who grew up on FORTRAN), but my presentation is neutral of any language mainly because I don't want implementation details to interfere with the larger picture.
Sure, C++ has more in terms of raw power, but is that the focus of the class? Cool tricks we can do with pointer arithmetic in a double-ended queue? Yes, CS majors should delve all the way into C++ (and or assembler, as some posts are mentioning, if hardware becomes a focus), but for the conceptual differences that OO presents compared to procedural code, Java gets less in the way and leaves more mental resources available to the student.
... what happened to all the finery Sally was wearing on one of her last campaigns (save the children). No doubt she donated them to some needy, down-and-out dot-commer that could make better use of such nice boots and trenchcoats than she could in those poor third-world countries. She should serve as a model to us all. *grin*
That occured to me also. Either you use this kind of thing on a one-way trip (good for expendable probes) or you make a sling-shot around the planet of interest and conduct your business *real fast* on the fly-by.;) I haven't heard anything written on the concept of "tacking" upwind like you would do in a sailboat (since I believe that makes use of lateral pressure exerted on the boat from the water). The only thing left would be some other (more conventional?) thrust device, but now we're stacking up lots of extra mass to the point that the magnetic sail wouldn't work nearly as well to begin with, rending the discussion somewhat moot.
Unrelated to this, though, I have to wonder how massive the magnetic generator itself is. Sounds like they can meet the 1kW power requirement well enough, but it needs [nearly] 1kg of "fuel" (for plasma) also per day. The numbers quoted at the top for a 200kg vehicle isn't all the heavy. By the time you add yourself, a friend, and the obligatory Beowolf/Linux cluster, well, I just start wondering if it would truly attain their listed speed *or* endurance rating. Just my thoughts.
I purchased two of these [new-line?] linksys cards several months ago to start my home network. I had heard from someone that Linksys cards had a habit of being NE2000 compatible, but this isn't the case with these cards. Turns out, the floppy disk that came along with mine (at least the one in the network starter kit which has cat-5 and a hub) has a "linux" directory on it, inside of which is a modified "tulip.c" driver. There were even instructions on how to compile the driver. When it was all said-and-done, the card worked perfectly for me (with RH6.0). (NOTE: I did have to rebuild the driver after I rebuilt the kernel though; the stock tulip.c in my distro isn't totally compatible.)
What to me is funny, though, is that win'98 had a real issue of NOT working on my box also. Turned out this was due to win'98's "PCI-steering feature". When that was turned off (and the system was forced to look for devices where the BIOS had put them on bootup) it worked there also.
It did take a few days though. The Linksys website didn't have any clear mention of Linux-specific help, and their "support" personnel never did return two calls I placed, so the quality of their alleged support is very much suspect IMO.
Slashdot Mirror
Problem here is that, to be effective, you'd want a reasonable difference in 'altitude' between the 'corners' to make this effective, but to maintain a similar orbit, objects with lower altitude must have higher speeds. That is, your low-altitude 'corners' will tend to speed away from the high altitude corners. The end-result of that happening is left as an exercise to the reader.
I have to agree with this. Just about every time I have some problem that I cannot immediately understand and fix, I'll try a few different directions (usually at the same time): find a HOWTO page, do a google search, and jump on some relevant IRC channel to talk to a real person with real experience. Sadly, this last option just hasn't worked for me in over a year. I had a minor disk failure other day that caused, among other things I finally had working again, a problem in evolution. The only responses I got (at least on the servers I was trying) was in #gnome, and the response was along the lines of "that suck-@ss ximian sh|t ain't got no place on a REAL gnome system". Yeah, it was real helpful, eh? I understand the parent poster's problem and plea. If you don't have anything helpful to say, then STFU. *shrug*
Though I haven't read through the article as yet, I do find myself occasionally (and more frequently, of late) almost consider "going back" myself. I've been on Linux for, hmm.., four years now. I have stuck with my distribution throughout, and have really only bought the CD's to upgrade twice. (That since people are saying "well, if he didn't chase every upgrade, he wouldn't feel that way.) I try to leave my system alone; I'm quite adept at causing myself plenty of grief and stress w/o changing kernels, distro's, or the like. That stress I do have in "dealing" with Linux, granted, stem from my messing with other stuff: NFS exporting a network directory, poking at NIS, poking at NTP, getting wine to compile/work, and other Unix/Linux things. Don't get me wrong: I love the command line; I grew up on (MS)DOS before I knew, much less had, windows 3.x, so the command line is comforting to have around. On the other hand, when my NAT'ed boxen stopped resolving names via DNS other day, it took over an hour of pouring back through man pages, HOWTO's, and /etc/foo.conf files to figure out that /etc/sysconfig/??/forward_ip4 was now "0" instead of "1". (And I'm not sure just what happened to cause that.) Dumb little thing, sure, and granted when I understood what the problem was, I knew what else to go looking for. Thing is, I just can't find enough time in the day to fully comprehend every nuance of the underlying structure that makes Linux (and perhaps UNIX) what it is. Who's failing is that? Mine, perhaps, if anyones.
Anyway, those times I get the tempting voice to "go back", I remind myself that, more than the above issues I inflict upon myself, I stay with Linux since the philosophy suits me. I don't pour through source code myself, but I trust there are plenty of you out there who can and do. The whole open-ness of approach, for me, offsets the problems I have in the preceding paragraph as I'm not that worried about having "modify-system-at-will" stuff secretly lurking under the surface of the OS. If I was to switch back over, I'd just have stress associated with untrustworthy computing practices. I think for the near term I'll stick with that stress which is mainly self-imposed.
"AquaGraphite" themes (now that themes.org site has returned) for gtk and sawfish get a similar overall look. I have slightly hacked versions of both running on my desktop. Very slick appearance. Can probably find the WM theme for your WM if you look. [shrug]
For the record, I'm 28; getting, learning, and understanding a slide rule was a personal curiosity. I did all of high school, college, and graduate school with a one-line, no-graphing calculator and did just fine that way, so I'm a bit shocked (though not necessarily surprised) about the comments of dependence on graphing calculators. I mean... "circle"... its round... "ellipse"... its round and oblong... sheesh. I'll go kick back in my crotchety chair now.
In other words, you have to figure that, as many clueless people are not patching their systems, our co-worker represents a large number of quite saavy people that are completely apathetic to wanting to be bothered. They don't have the interest to want to take the time; we can't reach these people using fear or logic. How, then, do we protect ourselves?
News of any Big Name (tm) 3D CAD software for Linux is huge in the engineering field. Where I work (as engineer in a large aerospace-industry company with infrequent-but-present CAD needs), we have been using Unigraphics on Solaris boxes. For whatever bone-head reason, couple years ago the entire company decided to begin replacement of several thousand Sun boxes with Dell/Intel solutions running W2K. The real geeks here amongst us made a good cry for putting Linux on the machines instead (save money, yes you can still get support, we still have Unix, no productivity drops, etc.) but, among many other "reasons" was one of the strongest motivators that Unigraphics (UG) didn't work under Linux but did under windows. Now, while the company deployment of and reliance on "UG" is large (so much so that there's no hope of switching to anything else; there is far too much inertia at this point), us pro-Linux guys didn't have too much come back for the pro-MS upper management point that there wasn't any top-of-the-line CAD packages for Linux. (As an aside for those interested, we are expected to be a fully-MS-campus by end of the year; already the amount of things that just don't work, courtesy of unknown coded limits in W2K and/or its basic non-Unix-ness, is enough to make you either laugh out loud or go cry in a corner.)
I haven't seen the comment you are referring to (nor am I that interested, really), but I did just want to raise a very generic and non-personal point that whether something is funny or not doesn't have much to do with whether something is on-topic or not; I've modded an otherwise decent humor attempt as off-topic before... yes, I got the [attempted?] joke, but there wasn't any context that related it to the posting or parent comment(s). (Yeah, the fact that this is off-topic with respect to the submission isn't lost on me either. Guess those are the breaks.) For my two cents, if someone is going to be funny, they at least should be passably on-topic about it.
Paranoid and/or a slashdot regular.
You installed the fan backwards.
I am not really into the "get / listen to music online" scene (so keep this in mind as you read the next paragraph). Steadily over the past few years, I have tapered off buying CD's to almost nil, and I've sworn off the local radio stations (not so much their CC tie-ins, but that their playlists are insanely short; I heard the same song going to work three days in the same week... and it wasn't a great song to begin with. ARGH).
So now I'm wondering... would it be feasible to setup a slashdot-style online music forum for independent artists to submit music tracks to, such that members of forum could categorize, comment on, rate/rank those submissions to get streamed online? Wouldn't need much of a DJ (per se) as much as a few people to keep the forum running. Heck, if it works, you could probably let members design their on CD for burning for, say, $5 to get sent to them, the proceeds of which go to the forum upkeep as well as the artists themselves. Granted it would take a little while to work up a good user base, but would something like this work? Pros and/or cons?
Excellent...
</montgomery burns>
I used to use a spreadsheet years ago before juggling the cells/formulas across many accounts (1 checking, 2 savings), multiple school and car loans, credit accounts, and now-a-days, 401k tracking, became a pain and I switched over to *cough*Money*cough*. MS sent me a 90-day freebie trial of the '98 version, and that was pretty much that; I've been using it since. (It is the *one* piece of software that I can actually give MS full props for, even if it is (or seems) just a glorified MSHTML app with a database backend. )
So, anyway, while I stipulate that a spreadsheet might work, the main thread was with respect to financial packages.
I have actually poked at each release of GnuCash, but consistently it lacks one feature that I am rather addicted to: scheduling income and expenses, and combining this into a budget forecast for the next N months. With something like this, correctly setup, I not only know how much money I have an any account *right now*, but I will have a reasonable ball-park figure for how much money I will have in three months, six months, etc. A nifty line-plot is handy to see where, when, and how bad the next "low point" is going to be, and as necessary I can adjust funds to deal with it gracefully before it has a chance to bite my sorry a$$. Very useful planning tool that, now, I cannot do without. This is the one single feature keeping a '98 partition hanging around my house.
;)
Now, I'm not too bad writing bits of code and what-not (it's a tangential part of my day job), and I appreciate that, to some extent, linux money applications can be scripted and stuff; maybe I could roll my own forecaster this way, but I really don't want to feel like I need to kludge together such a relatively 'big' feature when I don't have the time and interest after getting home. (Maybe it's just me, and, yes, I'm a bit lazy once I'm off the clock.
I probably haven't looked into all possible alternatives for a linux-based financial program, but so far I haven't noticed one that really handles this.
Personally, I was hoping the Congressman's letter would end with:
:-P~~"
"p.s. Our Bill is better than your Bill.
We know grammar too. You need either a comma or a hyphen in your first sentence. Lemme suggest either:
or
depending on what you meant. (Sorry, man, but opportunity presented itself.
Since the /.'ed page appears to be posted in multiple copies already, how about a couple viewpoints from someone whose mom actually tried this?
:P )
This definitely has to be involved in the whole Herbalife Independent Distributor scheme. Just to satisfy your curiosity, yes it is a MLM. You sign up under another distributor (the person that lets you in on what the product actually is, etc.) and they get a small cut off of your sales. Quite pyramid-like, since the best profits come not by selling the product itself, but by recruiting other independent distributors underneath you. (Otherwise, you probably *could* make money, but not the amounts you hear glamorized).
Now, the rest of my information is, to be fair, several years out of date. I can only relay how it was all seemingly setup in the early-mid '90s. Basically, the top-level Herbalife company (corporation? I wasn't too sure exactly) works very hard to keep a legally viable separation distance from all the indenpendent distributors. (This is perhaps why the company doesn't get in trouble for the street spam, since, as another poster pointed out, the ID's are the ones putting them up.) This went so far as prohibiting ID's from using any sort of Herbalife company logo on business cards to maintain a clear separation. (IN fact, the letter of law at the time was that you were only allowed to say "Independent Herbalife Distributor" in a plain font face. Nothing else.)
The signs you see all over creation that make no reference to the product or company arise from this same seeming need to maintain separation. You, as an ID, are not allowed to advertise what the product is or might be; you have to entice interest in the customer, and after *they* make the initial contact can you explain what it all is.
They even went so far as laying down policy on accepted forms of advertisement with respect to various media. Mom, trying to play fair and honest with them, inquired about constructing a store-front web site. They gave consistent, repeated, flat "No"s when she described it. They only saw a web site as a means to advertise for potential business, not as a possible means to *DO* business. (Remember the time frame involved here: early-mid '90s.) The company's own site was the only place permitted to have logo's, product information, etc., and even *they* themselves didn't have a mechanism to take money. (Since, I guess, only the ID's were allowed to move merchandise and collect cash directly.) It was weird.
Now, it got interesting to see that the only people making cash were the select few people at the top of the pyramid. The dominant exceptions are people that are providing these signs for ID's to put up. (Wonder why they all look the same in every city? There's a place that an ID order's them from.) In store product demo's and the like were also something for and ID to spend money on. (Physical stores. You could have the product displayed all over the counter, but don't you dare have any mention of Herbalife in your store window for passer-by's to see.)
Needless to say, it is in fact quite hard to actually set this up as a lucrative business. You really need a big downstream pyramid of ID's underneath you, but then they are competing against each other (and you, perhaps) to get the same thing, so any normal size town (or sub-section thereof) can get saturated by various ID's competing for either customers or other ID's). Mom worked at this for over a year, and I don't think she ever really turned a dime over in her favor.
(Oh, and if any company Nazi's start looking for her as a result of this post, she has a different last name than me.
I was taking a few numerical methods classes in my engineering classwork not *too* long ago. At the same time, I was trying to learn how to apply OO concepts to new problems as well as learn the finer points of Java. So I started doing my various homework assignments and projects in an OO fashion with Java. Maybe a few ideas can be a starting point. (Bear in mind what I did was somewhat crude, but it was personal homework assignments, not full-on production code releases. I'm also not advocating use Java for high-performance-demanding code per se; I was only killing several birds with a single stone
For a finite element methods class, I created classes to represent various element types. (Bars, thin triangles, thin quads, etc.) There was also a "matrix" object. Now, each element type object would have a method to populate its coefficients into the matrix, and the matrix would have a method with which to solve itself. Now, the implementation of this particular method is quite procedural if you only look at that part; the OO comes about providing a more abstracted method of assembling the matrix, dealing with neighbor-to-neighbor effects, and the like. Ultimately, I had different classes to solve matrix objects by different techniques, so I could easily change how the problem was being solved by just tossing the matrix object to a different object/method (which was already fairly polished in the more general numerical methods class).
For things similar to CFD solutions (a la your finite volume question), you can likely use OO to organize your flow domain into different region objects, each of which has some notional grid object (or perhaps collection of volume objects). The methods in these classes can give you the ability, again, to assemble matrix objects which can be passed off to a particular solver. (Again, this implementation of the solver will likely appear quite procedural; I didn't carry the idea too much farther before gradution to know if the various matrix inversion techniques could be cleverly OO-ized.)
So in other words, use OO where the principles can help you. It does take some amount of thought to redesign your approach to the problem. Hope this gives a few things to think about.
Had a college roommate who followed some advice from the school newspaper regarding jerking telemarketers around. One such call went as follows...
Telemarketer: "Hello, sir. We're running a special on $FOO and for a limited time, we'd like to make it available..."
Roommate: "I worship Satan!"
Short pause...
Telemarketer, less confidently: "Umm, ok, well, see the special we have is that --"
Roommate: "No! You don't understand... I worship the dark prince Lucifer who rules from his dark throne!"
Telemarketer:"Umm... ok. Have a nice day then."
... problem solved. Whoever they were, they did stop calling. It still brings a tear to my eye.
Color me confused, but the whole behavior of a sonic boom is that "information" (such as the length of the craft) cannot travel upstream across shock waves, so the nose "boom" won't know how long the aircraft is. Besides, it is the Mach number of the craft in flight that determines the angle of the primary nose shock, not the geometry of the nose (or remainder of the craft) itself. So I figure the story has to be horribly mangled from the original source at this point to imply that stretching the craft can even "flatten" the generated shock waves. Very little to see here folks.
First off, let me say I am a supporter of Java as a language and platform. I started with a working knowledge of C before C++ or Java entered the mainstream consciousness, but I found picking up C++ to be troublesome whereas picking up Java was a snap. Then, having a good knowledge of C (for the syntax) and Java (for the OO concepts) making the final transition to C++ was a snap.
Personally, I find these accusations of Java being clunky to be interesting, as I see Java as being more clean and consistent with its class API than what I've seen in C++, and ever time I try to accomplish something that was trivial in Java into C++, I come up against syntactical brick walls. (Okay, I will grant that I don't *know* C++ nearly as deeply as I do Java). In fact, I have a RedHat box at home that is serving various utility services to my two other boxes (one's for the wife running '98), and it is all being done with Java since client/server code is hardly any more difficult that printing "hello world". Speed is not a problem once the virtual machine loads up the classes, and this is what I think most people perceive as slowness. (And it also bears noting that load-up is a bit faster with whatever JRE I have under linux (IBM, IIRC) than with Sun's JRE under windows.
Anyway, I digress. If the course material is mainly about the OO approach to code design, then I would favor Java over C++ just since Java's syntax seems to get less in the way, and thus seems to distract less from the matter at hand. In fact, in less than a week, I am presenting OOP to my fellow engineers (many of them "greybeards" who grew up on FORTRAN), but my presentation is neutral of any language mainly because I don't want implementation details to interfere with the larger picture.
Sure, C++ has more in terms of raw power, but is that the focus of the class? Cool tricks we can do with pointer arithmetic in a double-ended queue? Yes, CS majors should delve all the way into C++ (and or assembler, as some posts are mentioning, if hardware becomes a focus), but for the conceptual differences that OO presents compared to procedural code, Java gets less in the way and leaves more mental resources available to the student.
... what happened to all the finery Sally was wearing on one of her last campaigns (save the children). No doubt she donated them to some needy, down-and-out dot-commer that could make better use of such nice boots and trenchcoats than she could in those poor third-world countries. She should serve as a model to us all. *grin*
That occured to me also. Either you use this kind of thing on a one-way trip (good for expendable probes) or you make a sling-shot around the planet of interest and conduct your business *real fast* on the fly-by. ;) I haven't heard anything written on the concept of "tacking" upwind like you would do in a sailboat (since I believe that makes use of lateral pressure exerted on the boat from the water). The only thing left would be some other (more conventional?) thrust device, but now we're stacking up lots of extra mass to the point that the magnetic sail wouldn't work nearly as well to begin with, rending the discussion somewhat moot.
Unrelated to this, though, I have to wonder how massive the magnetic generator itself is. Sounds like they can meet the 1kW power requirement well enough, but it needs [nearly] 1kg of "fuel" (for plasma) also per day. The numbers quoted at the top for a 200kg vehicle isn't all the heavy. By the time you add yourself, a friend, and the obligatory Beowolf/Linux cluster, well, I just start wondering if it would truly attain their listed speed *or* endurance rating. Just my thoughts.
I purchased two of these [new-line?] linksys cards several months ago to start my home network. I had heard from someone that Linksys cards had a habit of being NE2000 compatible, but this isn't the case with these cards. Turns out, the floppy disk that came along with mine (at least the one in the network starter kit which has cat-5 and a hub) has a "linux" directory on it, inside of which is a modified "tulip.c" driver. There were even instructions on how to compile the driver. When it was all said-and-done, the card worked perfectly for me (with RH6.0). (NOTE: I did have to rebuild the driver after I rebuilt the kernel though; the stock tulip.c in my distro isn't totally compatible.)
What to me is funny, though, is that win'98 had a real issue of NOT working on my box also. Turned out this was due to win'98's "PCI-steering feature". When that was turned off (and the system was forced to look for devices where the BIOS had put them on bootup) it worked there also.
It did take a few days though. The Linksys website didn't have any clear mention of Linux-specific help, and their "support" personnel never did return two calls I placed, so the quality of their alleged support is very much suspect IMO.