And in fact, the Sahara was inhabited following the last ice age (see this), and the final drying started around 5000 BP. Some believe the root cause was variations in the earth's tilt (this) -- which doesn't mean that a change in circulation cells wasn't the direct cause of the drying.
Actually, the business model has been a wide success (i.e. cellphones for one). The internet appliance people got caught by the fact that their hardware was pretty easily used for other purposes, and those purposes were attractive enough that lots of people (read geeks/slashdotters) went out and bought every unit they could get their hands on. (I know, I saw a friend do this.)
How does that "empower" the people who can't afford the up-front cost?
Because it gives them a piece of hardware that they wouldn't have had otherwise. Because the act of hacking the box is a LEARNING experience, and paves the way for more learning experiences, which might just open up a way to a better life. Knowledge is power.
You missed my point: it doesn't empower them because either (a) the company that made them went out of business shortly thereafter, dis-empowering lots of people like them, or (b) the company never produced them in the first place because they were scared away. The general idea here was to provide access to data and communication and tools, not as a toy or learning experience for computer hackers.
A non-profit company could certainly license, build and sell these - so where does that leave your theory?
Why would a legitimate charity have any incentive or motivation to lock their beneficiaries in to a propriatary operating system? Why would a non-profit spend money on software licensing fees when it could help more people by using Free software? I can't think of any reason to do so unless they have ulterior motives -- like shilling for Microsoft.
Ummmm - because the operating system wasn't the point? This thing isn't running WinXP (or Word). It's running CE, and the cost for that in an embedded use is around $5 as stated here. Some charities might be focused on improving the lives of people by giving them access to things like weather reports, crop prices, etc, etc, not on what software it runs.
And more importantly, you missed my point again. My point was that a non-profit _could_ sell an unlocked version of this - nothing requires that it be locked. Also, as others here have said, Linux is an option for PIC; the current shipping ones happen not to use it because the companies making them decided not to (or the ancillary applications such as those on the WinCE PIC weren't available or up to the level of the WinCE ones).
No one is stopping these manufacturers from selling PICs running Linux or unlocked CE. But for most (I didn't say all) users, this is a good device as it is. And as has been stated many times here, you are not the target for this. Nor are proto-geeks in these countries really the target market either (though they could be targetted by say a Linux PIC or an unlocked PIC).
Getting Linux to run on this isn't just a Geek Toy. Doing so will empower ALL people to use a cheap (possibly subsidized) general-purpose computing device in ways other than their corporate masters want them to use it. The telcos and ISPs who will be licensing this want nice obediant consumers who'll buy the device, subscribe to their service, and use it to buy stuff from their advertisers & partners.
Or, like all the other "Internet appliances" that came out and were hacked, allowing people to bypass the service that provides the subsidy will drive the company producing them out of (this) business.
How does that "empower" the people who can't afford the up-front cost?
Corporations aren't looking to enslave people, really. Make a profit, yes (though a non-profit company could certainly license, build and sell these - so where does that leave your theory?) Someone licensing these could leave it unlocked I'd bet if they want to. If they're selling them at breakeven or a loss (aka subsidized), they're not likely to leave them unlocked, since they wouldn't last long doing so.
As for AMD: they may well be selling these processors at effectively cost, in the hopes of creating markets for their higher-end stuff in the longer run. One way to make a profit in the long run is to do things to help create demand by improving the economic/educational situation of people/countries that are not viable current customers.
I can envision a spreadsheet at AMD showing how a small investment (developing PIC, selling the geodes at cost, etc) could result in considerable long-term profits 10 or 15 years down the road (and some profits much sooner). Not to mention getting a marketing leg up in 3rd-world/developing countries on Intel.
and use RAS (Remote Access Server) to connect to the voting machines through telephone lines.
One detail left out: did it answer calls from every phone number in existence, or just the ones on an approved list?
Caller-id _can_ be spoofed, of course, even without actually tapping the phone lines, which could be done.
Note: I'm just saying it can be done, not that it was. Plus I'll bet it does accept calls from anywhere anyways, and has some sort of login protocol/password that's supposed to protect it. If they were smart, it's some variant of certificate-based authentication. If they weren't, it's plaintext or the like.
Videophones don't have to use a ton of BW over a G.711 call. Worldgate [disclaimer: I work for them] and Motorola will be selling a videophone (Ojo) that works well at a steady 30fps over cable or DSL connections, at total bandwidth (including overhead) of 100Kbps (dsl) to 150Kbps (cable), audio and video. A G.711 (20ms) audio call is 64Kbps data, plus around 16Kbps overhead for around 80Kbps total. Don't knock visual and non-verbal communication until you've tried it, even ignoring people who are hearing-impaired. Random example: The actress we used to talk to people at shows (from our offices in Philly) wants a pair so she can keep in touch with her mother when she moves to LA, for example.
Also, encrypting the packets would really mess with things if some packets were missing. (very common, called jitter)
Umm, you're a network admin who deals with VoIP? Missing packets are packet loss, not jitter (which is variance in transit/arrival times (delay) of packets, though extreme jitter can cause the (de)jitter buffers to underflow and drop packets that arrive especially late). And packet loss doesn't mess up SRTP (encrypted) streams; they're typically block-coded with an AES key that's exchanged at the start of a session; each block of data can be decrypted without reference to the others.
I was part of an NIST program in '92/93 or so where they paid people to make calls to each other and talk about a particular subject. I believe they took the data, made transcripts, and it was fed to companies as a database of US english speakers (of varying accents) to use to tune speech recognition. (I think I got involved with it via Bell Labs, who we were working with on using their 3210 DSP for speech recognition in Amigas.
So there may be multi-accent databases out there to develop base profiles for speech recognition.
I'm not saying that the republicans stole the election in florida. But could they steal it with enough installed Diebold machines? Looks pretty easy to me. Thats not a conspiracy theory - thats Diebold looking for americans to not care how their votes are counted.
Agreed. Diebold as a company per se didn't (IMO) go out to rig elections - but a series of decisions made there, both by execs and so-so programmers/designers has lead to that being a real possibility (plus the possibility that this person under fraud indictment did want to make it possible - it's far from certain that he was involved, but it's plausible).
What this shows (especially the bit with this guy) is that voting systems need to be open to inspection, at least by officials and their representatives. It also points out how important monitoring, control and recording of physical and network access is. You'll note that almost all protocols for paper ballots involve controlling who has access to ballots and machinery and under what supervision.
Most of what you describe to me above is a long explanation of why we need to be more distrustful of ties between government and business. How can so many people be trustful of a VP that is pouring money into his former company? Perhaps at this point he's not directly benefiting, but i bet he left behind a few friends that are rather well taken care of.
Absolutely agreed. "You scratch my back, I'll scratch yours" is firmly entrenched in almost all levels of politics, not just national. That's one reason why involvement in local politics is important, as are open government laws, etc. And neither party has an exclusive on this, though the republicans are definitely more tied into the powerful/rich corporations. To an extent, all politics is involved in this: they tell groups "get me elected, and I'll do thing X that will make you happy", where X could be removing pollution controls or adding them, or raising taxes on the rich in favor of the middle class, or the reverse. That said, these different X's aren't all morally or ethically equivalent in my point of view, and some are about fundamentals of who we are as a people.
Electing these people simply continues an aggregation of power into the hands of the wealthy.
(I guess I feel your post makes these people look more innocent when from many perspectives they're causing some serious damage in the world.)
From my observations, most of them (like most people in any group) are not evil per se - from their point of view they're mostly looking out for themselves, their friends, and their points of view. Many/most of them are doing so innocently (as in not intending harm, thinking they're doing the "right" thing), some are cynical and/or not caring much about the harm they do if they reach their goals. True evil (as in doing harm to see others suffer, or total lack of caring over how much damage is done to others) is rare - in any group. Of course, situations like politics and the associated temptations and requirements to constantly raise money can both select for the less-ethical to a degree, and at the same time tempt people who otherwise would not do questionable things into doing so. Power is also tempting, and using power to further ones own personal agenda (like Bush and religion).
I just wanted people to have an insider's view of what these schools and social circles are like. Interestingly, you'll note that Connecticut went firmly for Gore (all congressional districts) in 2000, and is consistently one of the last bastions of moderatism in the Republican party. Some moderate/liberal (at least socially and environmentally) Republicans there have been cut adrift by the increasingly right-wing and religious mainstream of the party over the last 20 years, as the party lines have "rotated". Lowell Weicker for example (who by weird coincidence recently bought my grandmother's old house I recently found out - we sold it 20 years ago).
Having gone to these "elite private schools" in NYC and Connecticut, having an uncle who went to 1-12th grade and Yale with the elder George Bush and who was his roommate, I have to say that the idea of a Connecticut/Yale/Tory/whatever conspiracy is simply amazingly unlikely.
A lot of the people in these schools aren't that smart (though there is a pecking order academically, all of them have their share of the less-smart (or don't-care) legacy-types). Pretty much the primary determinator of who goes to these schools is who a) can afford it, and b) wants to. After those are passed, then legacies, academics, and other factors (attempts to provide a somewhat diverse enrollment, etc) are considered. Most have (through various scholarships, foundations, etc) a moderate percentage of "disadvantaged" students.
A classmate of mine was another of the Bush crony's kids: Doug Baker, James Baker's (former chief of staff to G.H.W. Bush) son. (This was 1977-1980.) Not shall was say one of the sharp ones in the class (hardly), but a good football/lacrosse player and partier. At my 15th reunion (1995) he had become a lobbyist (what a shock). Others I went to school with include JFK. Jr, David Duchovny, and various sons of very well-off businesspeople. There was a sizable contingent at boarding school from Midland TX around 1980; sons of oil men and the like (many of them like Bush, transplants following the money).
My uncle went to day school with G.H.W. Bush, then to boarding prep school, then to Yale with him. In prep school they were roommates at one point. Both flew in WWII, but my uncle was in P-51's over Germany, and unlike Bush didn't go back to Yale. He continued to live in CT (New Canaan), and was a stock broker and staunch Republican for many many years, was Chief of Police in New Canaan after got tired of hunting and fishing in retirement, etc. When G.H.W. Bush was running for re-election, Frontline interviewed my uncle about Bush's school days. One of my uncle's comments: Bush was an idiot. Almost all of it (including the idiot comment) was edited out. Today he's an independant who REALLY wants to see W go down in flames. He supported Dean in fact.
Which brings me to the comments I'm replying to. While in theory there could be a conspiracy by some nebulous east-coast preppy elite, the reality as I see it from having grown up and gone to school with many in Bush's circle is far more simple and easy to swallow - the Bushes (and most presidents, with the odd exception like Clinton) are from rich families, and those families have connections to other rich families, and draw on them for their closest advisors and supporters. A lot of these people get into prestigious schools, colleges, and jobs via family connections and history (legacies). Not everyone in these schools does, in fact it's probably a minority nowadays, but it was and still is common in many of them if not most.
These people are rich, they go to school mostly with other upper-middle-class or rich people, and they form friendships for life with the people they went to school with (and often with others of similar backgrounds, which is hardly unique). This applies to the majority of politicians, especially at the upper levels. It takes money and even more so connections to get to elected office, especially high office (and promises for a lot of back-scratching).
This isn't to say that none of them do bad/questionable things - hardly. Many do. But as others I'm sure have said here, never attribute to malice (or conspiracy) what is adequately explained by stupidity (or just plain normal social class cliquiness (sp)). Honestly, these people _aren't_ smart enough to pull such a huge conspiracy (let alone for so long) off.
p.s. While I attended these schools and have a long family history associated with them, I was not one of the "rich" kids or legacies - my mother was director of development at one (which got us in, free I think), and the other I went to not as a legacy, though my father and grandmother did pay for it. I'd be considered probably one of the middle/upper-middle class students with a family tradition of prep schools.
Dummynet can absolutely do it. Put a PC with BSD & Dummynet and two ethernet interfaces in to simulate delay/loss/BW restrictions/etc. Very configurable. You can chose which packets are affected.
Worst problem: fixed delay, not bell-curve/whatever. You can roughly approximate delay variance by several rules of varying probabilities. Also, loss is random not bursty. For most testing, this is fine.
It can take a little while to get used to configuring it. Don't forget to make it act like a network in both directions!
This "disk drives will be obsolete" assumes that disk drive prices are flat. Drive prices are one of the few things that has (if anything) beaten Moore's Law. Eventually they'll probably flatten out - but not yet.
The "death knell of rotating media" has been sounded more times than I can remember. Anyone remember the front-page stories that by late 80's bubble memory would have replaced hard disks?:-)
Those are standards for digitizing NTSC/PAL video. Note however that NTSC (broadcast/cable) has a maximum (best-case) luma bandwidth of 4.2 MHz, and max chroma bandwidth of circa 1 MHz (I forget the exact value). So while 600+ pixels are needed to represent 4.2 Mhz, chroma information is FAR more limited - best case around 150. (All numbers from memory.) Also, realize that 4.2 MHz is an upper limit that is rarely reached by an actual TV/signal combination.
Killing power "at night and on holidays" to monitors, area lights, etc might be ok in some cases/businesses (such as a call center, perhaps). Rather a bad idea if someone is working late on an important project...
There are better ways to save power on monitors (as mentioned in my original post and others). Not to mention the power button... I never leave my office without hitting the monitor power button.
As mentioned elsewhere in this thread, the DOE has free software (for Windows, sigh) that helps organizations control/set monitor power management.
Our small company (when money was tight) shut down lots of extraneous unused servers and got serious about turning monitors off at night, and saved a fair bunch. (We had tons of servers lying around.)
Re:Swap caps lock and control
on
Is Caps Lock Dead?
·
· Score: 3, Insightful
Ah. A vi user. If you're an Emacs user, having the capslock key mapped to control is the ONLY way to fly. As others have said, that's the One True Position for the control key. (Check Sun keyboards, for example.)
My "CapsLock" keys have the legend worn off of them, and one even has a groove from where my pinkie's fingernail hits it.
Minimo (as the memory graphs show) does not suck memory as time goes on. Firefox and Mozilla don't either, so long as you don't set it for a huge memory cache or keep opening more tabs or some other such thing. The perf and footprint teams worked quite a bit to improve both from the days of 0.X.
While mozilla and it's friends may use more memory than you could make them use (with a ground-level redesign/recode), they're also very flexible.
Opera's code is small, and it's fast. Minimo and Firefox are a bit larger codewise, and fast (though perhaps not as fast as opera) but the biggest hit is runtime memory usage, and a lot of that is images and datastructures, etc, that aren't vastly different in size.
Worldgate and Mozilla/minimo
on
Mozilla's Mini-Me
·
· Score: 3, Informative
While minimo targets Linux; it inherently is largely applicable to another environment - especially since they expect the front-end to be rewritten by someone using it in a real application.
Worldgate was going to use Mozilla for it's next-generation browse-the-web-on-your-cable-box application, where the browsers all run in servers at the headend and send screen images down to the settops as MPEG stills. We ran over 20 copies of Mozilla (tuned in ways similar to minimo) on 500Mhz P3's with 512MB of memory, and performance was reasonable. We lived with scroll bars where we had to (we subverted a few things to let pages fit tighter, but we also had to use larger-than-normal fonts). For added fun we had no mouse, but we had keyboards.
The toughest part was "geometric navigation" of links/etc with arrow keys; before development on that ended when we sold off our patents/business we'd mostly gotten that working, but there are more edge cases than you can count (nested and inline frames, imagemaps, etc).
I was a nielsen "family" a couple of years post-college (~1988) for sweeps. Probably skewed the ratings by working at home and coding all night long while watching Discovery and CNN.
A friend was a Nielsen family (wired) for a year, also a SW engineer. Realize that Nielsen probably also misses out on a fair number of the working poor - but I agree their samples can be pretty issue, especially for "niche" shows. They probably try to match nationwide demographics with their sample, but in things they don't ask about or level on it's probably significantly off.
The whole diary thing was a problem from the start, but with the first remote controls, then the explosion of channels (and other video sources) it got far worse. And, as you say, people have trouble remembering what they were watching something on, or write in things they wanted to watch or shows they want to "reward", or leave off things they "shouldn't" have watched. For example, I imagine far fewer admit to watch Friday-night Skinemax movies than actually watch them - especially the teenagers.
I wonder when Nielsen will start feeding data to the Matrix....:-):-( "You watched the same news shows as a known terrorist! Come out for preventive deprogramming before we have to drag you out!" 1/2:-)
Ah, for the days when a good fraction of Usenet postings had NSA Food signatures... Though I don't think they're looking for references to Kremvax anymore.
We here (techies) understand how easy it is in hindsight to set up a scoring system so that the already-known bad guys end up on top. How well such an after-the-fact-designed (after 9/11) system is at picking out _other_ terrorists is unknown, and iffy at best (unless one of the inputs is "known to have trained in Afganistan with Al Queda"). They say that investigations were launched and "several arrests were made". It doesn't say that those arrests had anything to do with terrorism, or that they weren't later released or just deported for immigration violations. If they _were_ terrorist-related arrests, you can bet Siesint would be crowing about it and it would be part of Matrix now (even if it were pure chance or blind luck).
They even say it was dropped because it was 9/11-specific.
Your point is well taken - he may not be a C/C++/java/whatever programmer, and certainly he's most comfortable in Erlang - and as I mentioned, that's important to his point.
However, my point was that if a certain performance level was important, doing the initial implementation in a language (any of several) that was appropriate might have obviated the need to do multiple optimization passes, analysis, etc. I.e. a simpler algorithm implemented in a "faster" language might have run as fast as his 5-phase-optimized Erlang algorithm.
And yes, the performance of the development team matters a lot. There are all sorts of parameters that go into deciding how to solve a problem, obviously: team skills, tools, how often will this be run, what the requirements are, expected hardware, lifetime, maintenance (a simple algorithm in a "faster" language will have an advantage here over a more complex/heavily-optimized one in a slower language, for example), etc, etc.
In his "test" case, apparently on the order of 60+ fps was "good enough", though he was trying to illustrate what seemed general optimization methods - looking for work that can be factored out; work that can be collapsed; work that wasn't really needed; and ways to do the work faster (better O() algorithms) before micro-optimizing. And if you do optimize, make sure it's actually a part that's contributing to it being slow (profile, or analyze).
66 fps on a 3 GHz machine, doing a 600x600 simple RLE decode...
Ok, it's not bad for a a language like Erlang, but it's not exactly fast.
The big point here for the author is "it's fast enough". Lots of micro- (and macro-) optimizations are done when it turns out they aren't needed. And writing in a high level language you're comfortable in is important, if it'll do the job. This is a good point.
On the other hand, even a fairly naive implementation in something like C or C++ (and perhaps Java) would probably have acheived the goal without having to make 5 optimization passes (and noticable time examining behavior).
And even today, optimizations often do matter. I'm working on code that does pretty hard-real-time processing on multiple threads and keeps them synchronized while communicating with the outside world. A mis-chosen image filter or copy algorithm can seriously trash the rest of the system (not overlapping DMA's, inconvenient ordering of operations, etc). The biggest trick is knowing _where_ they will matter, and generally writing not-horrible-performance (but very readable) code as a matter of course as a starting point.
Disclaimer: I was a hard-core ASM & C programmer who for years beta-tested 680x0 compilers by critiquing their optimizers.
Re:SIlence is a pipe dream for me
on
A Silent PC Solution?
·
· Score: 2, Interesting
rdesktop on Linux/BSD works _really_ well. I pretty much no longer turn on the monitor on my XP machine. I can also bounce the X window over a tunnel to my machine at home, and control the XP machine at work from home.
I replaced both my and my wife's cases with Antec Sonatas a few weeks ago (we have both of them in the "office" section of our master bedroom at the moment). Quick summary: Nice Case! and quiet too!
In a bedroom in the woods, my machine is almost silent; the remaining noise it almost totally due to a slightly whiney WD 120GB 8-mb-cache drive. My video card (an NVidia GeForce 3 Ti-200) fan is either dead (possible) or is temp-aware. In any case I don't have a problem with it. My chipset fan is dead. After switching to the Sonata case, I found my old CPU fan was (a) hitting it's cage, and (b) noisy as a vacuum. I switched to a ThermalTake Volcano II+ (rebadged as Mad something, bought at CompUsa for ~$35) and that did the trick; I could barely tell the system was on other than the drive whine.
My wife's had a quieter CPU fan to start, but it's now the loudest thing in her system. Her motherboard/CPU are due for replacement, so we'll be rebuilding her machine with an ear to noise.
Even without that, we find the ReplayTV in the bedroom louder than both our PC's - and it's meant for use in A/V racks. Of course, on AVSForum there are a number of threads on how to quiet a ReplayTV even more....
Slashdot Mirror
And in fact, the Sahara was inhabited following the last ice age (see this), and the final drying started around 5000 BP. Some believe the root cause was variations in the earth's tilt (this) -- which doesn't mean that a change in circulation cells wasn't the direct cause of the drying.
You missed my point: it doesn't empower them because either (a) the company that made them went out of business shortly thereafter, dis-empowering lots of people like them, or (b) the company never produced them in the first place because they were scared away. The general idea here was to provide access to data and communication and tools, not as a toy or learning experience for computer hackers.
Ummmm - because the operating system wasn't the point? This thing isn't running WinXP (or Word). It's running CE, and the cost for that in an embedded use is around $5 as stated here. Some charities might be focused on improving the lives of people by giving them access to things like weather reports, crop prices, etc, etc, not on what software it runs.
And more importantly, you missed my point again. My point was that a non-profit _could_ sell an unlocked version of this - nothing requires that it be locked. Also, as others here have said, Linux is an option for PIC; the current shipping ones happen not to use it because the companies making them decided not to (or the ancillary applications such as those on the WinCE PIC weren't available or up to the level of the WinCE ones).
No one is stopping these manufacturers from selling PICs running Linux or unlocked CE. But for most (I didn't say all) users, this is a good device as it is. And as has been stated many times here, you are not the target for this. Nor are proto-geeks in these countries really the target market either (though they could be targetted by say a Linux PIC or an unlocked PIC).
Corporations aren't looking to enslave people, really. Make a profit, yes (though a non-profit company could certainly license, build and sell these - so where does that leave your theory?) Someone licensing these could leave it unlocked I'd bet if they want to. If they're selling them at breakeven or a loss (aka subsidized), they're not likely to leave them unlocked, since they wouldn't last long doing so.
As for AMD: they may well be selling these processors at effectively cost, in the hopes of creating markets for their higher-end stuff in the longer run. One way to make a profit in the long run is to do things to help create demand by improving the economic/educational situation of people/countries that are not viable current customers.
I can envision a spreadsheet at AMD showing how a small investment (developing PIC, selling the geodes at cost, etc) could result in considerable long-term profits 10 or 15 years down the road (and some profits much sooner). Not to mention getting a marketing leg up in 3rd-world/developing countries on Intel.
One detail left out: did it answer calls from every phone number in existence, or just the ones on an approved list?
Caller-id _can_ be spoofed, of course, even without actually tapping the phone lines, which could be done.
Note: I'm just saying it can be done, not that it was. Plus I'll bet it does accept calls from anywhere anyways, and has some sort of login protocol/password that's supposed to protect it. If they were smart, it's some variant of certificate-based authentication. If they weren't, it's plaintext or the like.
I was part of an NIST program in '92/93 or so where they paid people to make calls to each other and talk about a particular subject. I believe they took the data, made transcripts, and it was fed to companies as a database of US english speakers (of varying accents) to use to tune speech recognition. (I think I got involved with it via Bell Labs, who we were working with on using their 3210 DSP for speech recognition in Amigas.
So there may be multi-accent databases out there to develop base profiles for speech recognition.
Agreed. Diebold as a company per se didn't (IMO) go out to rig elections - but a series of decisions made there, both by execs and so-so programmers/designers has lead to that being a real possibility (plus the possibility that this person under fraud indictment did want to make it possible - it's far from certain that he was involved, but it's plausible).
What this shows (especially the bit with this guy) is that voting systems need to be open to inspection, at least by officials and their representatives. It also points out how important monitoring, control and recording of physical and network access is. You'll note that almost all protocols for paper ballots involve controlling who has access to ballots and machinery and under what supervision.
Most of what you describe to me above is a long explanation of why we need to be more distrustful of ties between government and business. How can so many people be trustful of a VP that is pouring money into his former company? Perhaps at this point he's not directly benefiting, but i bet he left behind a few friends that are rather well taken care of.
Absolutely agreed. "You scratch my back, I'll scratch yours" is firmly entrenched in almost all levels of politics, not just national. That's one reason why involvement in local politics is important, as are open government laws, etc. And neither party has an exclusive on this, though the republicans are definitely more tied into the powerful/rich corporations. To an extent, all politics is involved in this: they tell groups "get me elected, and I'll do thing X that will make you happy", where X could be removing pollution controls or adding them, or raising taxes on the rich in favor of the middle class, or the reverse. That said, these different X's aren't all morally or ethically equivalent in my point of view, and some are about fundamentals of who we are as a people.
Electing these people simply continues an aggregation of power into the hands of the wealthy.
(I guess I feel your post makes these people look more innocent when from many perspectives they're causing some serious damage in the world.)
From my observations, most of them (like most people in any group) are not evil per se - from their point of view they're mostly looking out for themselves, their friends, and their points of view. Many/most of them are doing so innocently (as in not intending harm, thinking they're doing the "right" thing), some are cynical and/or not caring much about the harm they do if they reach their goals. True evil (as in doing harm to see others suffer, or total lack of caring over how much damage is done to others) is rare - in any group. Of course, situations like politics and the associated temptations and requirements to constantly raise money can both select for the less-ethical to a degree, and at the same time tempt people who otherwise would not do questionable things into doing so. Power is also tempting, and using power to further ones own personal agenda (like Bush and religion).
I just wanted people to have an insider's view of what these schools and social circles are like. Interestingly, you'll note that Connecticut went firmly for Gore (all congressional districts) in 2000, and is consistently one of the last bastions of moderatism in the Republican party. Some moderate/liberal (at least socially and environmentally) Republicans there have been cut adrift by the increasingly right-wing and religious mainstream of the party over the last 20 years, as the party lines have "rotated". Lowell Weicker for example (who by weird coincidence recently bought my grandmother's old house I recently found out - we sold it 20 years ago).
p.s. I don't live in CT anymore.
Having gone to these "elite private schools" in NYC and Connecticut, having an uncle who went to 1-12th grade and Yale with the elder George Bush and who was his roommate, I have to say that the idea of a Connecticut/Yale/Tory/whatever conspiracy is simply amazingly unlikely.
A lot of the people in these schools aren't that smart (though there is a pecking order academically, all of them have their share of the less-smart (or don't-care) legacy-types). Pretty much the primary determinator of who goes to these schools is who a) can afford it, and b) wants to. After those are passed, then legacies, academics, and other factors (attempts to provide a somewhat diverse enrollment, etc) are considered. Most have (through various scholarships, foundations, etc) a moderate percentage of "disadvantaged" students.
A classmate of mine was another of the Bush crony's kids: Doug Baker, James Baker's (former chief of staff to G.H.W. Bush) son. (This was 1977-1980.) Not shall was say one of the sharp ones in the class (hardly), but a good football/lacrosse player and partier. At my 15th reunion (1995) he had become a lobbyist (what a shock). Others I went to school with include JFK. Jr, David Duchovny, and various sons of very well-off businesspeople. There was a sizable contingent at boarding school from Midland TX around 1980; sons of oil men and the like (many of them like Bush, transplants following the money).
My uncle went to day school with G.H.W. Bush, then to boarding prep school, then to Yale with him. In prep school they were roommates at one point. Both flew in WWII, but my uncle was in P-51's over Germany, and unlike Bush didn't go back to Yale. He continued to live in CT (New Canaan), and was a stock broker and staunch Republican for many many years, was Chief of Police in New Canaan after got tired of hunting and fishing in retirement, etc. When G.H.W. Bush was running for re-election, Frontline interviewed my uncle about Bush's school days. One of my uncle's comments: Bush was an idiot. Almost all of it (including the idiot comment) was edited out. Today he's an independant who REALLY wants to see W go down in flames. He supported Dean in fact.
Which brings me to the comments I'm replying to. While in theory there could be a conspiracy by some nebulous east-coast preppy elite, the reality as I see it from having grown up and gone to school with many in Bush's circle is far more simple and easy to swallow - the Bushes (and most presidents, with the odd exception like Clinton) are from rich families, and those families have connections to other rich families, and draw on them for their closest advisors and supporters. A lot of these people get into prestigious schools, colleges, and jobs via family connections and history (legacies). Not everyone in these schools does, in fact it's probably a minority nowadays, but it was and still is common in many of them if not most.
These people are rich, they go to school mostly with other upper-middle-class or rich people, and they form friendships for life with the people they went to school with (and often with others of similar backgrounds, which is hardly unique). This applies to the majority of politicians, especially at the upper levels. It takes money and even more so connections to get to elected office, especially high office (and promises for a lot of back-scratching).
This isn't to say that none of them do bad/questionable things - hardly. Many do. But as others I'm sure have said here, never attribute to malice (or conspiracy) what is adequately explained by stupidity (or just plain normal social class cliquiness (sp)). Honestly, these people _aren't_ smart enough to pull such a huge conspiracy (let alone for so long) off.
p.s. While I attended these schools and have a long family history associated with them, I was not one of the "rich" kids or legacies - my mother was director of development at one (which got us in, free I think), and the other I went to not as a legacy, though my father and grandmother did pay for it. I'd be considered probably one of the middle/upper-middle class students with a family tradition of prep schools.
Dummynet can absolutely do it. Put a PC with BSD & Dummynet and two ethernet interfaces in to simulate delay/loss/BW restrictions/etc. Very configurable. You can chose which packets are affected.
Worst problem: fixed delay, not bell-curve/whatever. You can roughly approximate delay variance by several rules of varying probabilities. Also, loss is random not bursty. For most testing, this is fine.
It can take a little while to get used to configuring it. Don't forget to make it act like a network in both directions!
Sheesh, that'll teach me to post without preview before 10am.
This "disk drives will be obsolete" assumes that disk drive prices are flat. Drive prices are one of the few things that has (if anything) beaten Moore's Law. Eventually they'll probably flatten out - but not yet. The "death knell of rotating media" has been sounded more times than I can remember. Anyone remember the front-page stories that by late 80's bubble memory would have replaced hard disks? :-)
Those are standards for digitizing NTSC/PAL video. Note however that NTSC (broadcast/cable) has a maximum (best-case) luma bandwidth of 4.2 MHz, and max chroma bandwidth of circa 1 MHz (I forget the exact value). So while 600+ pixels are needed to represent 4.2 Mhz, chroma information is FAR more limited - best case around 150. (All numbers from memory.) Also, realize that 4.2 MHz is an upper limit that is rarely reached by an actual TV/signal combination.
Killing power "at night and on holidays" to monitors, area lights, etc might be ok in some cases/businesses (such as a call center, perhaps). Rather a bad idea if someone is working late on an important project...
There are better ways to save power on monitors (as mentioned in my original post and others). Not to mention the power button... I never leave my office without hitting the monitor power button.
For reference, Philadelphia area (PECO) is circa 13-14 cents/kWh. LCD saving would be $40ish by your calculation.
My Viewsonic G810 21" (20.0" vis) CRT has a typical power rating of 142 watts, not 200, so my savings would be less - around $30.
Note that if they can't reliably get people to shut the monitor off or use the power-saving modes, the savings would be appreciably higher.
As mentioned elsewhere in this thread, the DOE has free software (for Windows, sigh) that helps organizations control/set monitor power management.
Our small company (when money was tight) shut down lots of extraneous unused servers and got serious about turning monitors off at night, and saved a fair bunch. (We had tons of servers lying around.)
Ah. A vi user. If you're an Emacs user, having the capslock key mapped to control is the ONLY way to fly. As others have said, that's the One True Position for the control key. (Check Sun keyboards, for example.)
My "CapsLock" keys have the legend worn off of them, and one even has a groove from where my pinkie's fingernail hits it.
Minimo (as the memory graphs show) does not suck memory as time goes on. Firefox and Mozilla don't either, so long as you don't set it for a huge memory cache or keep opening more tabs or some other such thing. The perf and footprint teams worked quite a bit to improve both from the days of 0.X.
While mozilla and it's friends may use more memory than you could make them use (with a ground-level redesign/recode), they're also very flexible.
Opera's code is small, and it's fast. Minimo and Firefox are a bit larger codewise, and fast (though perhaps not as fast as opera) but the biggest hit is runtime memory usage, and a lot of that is images and datastructures, etc, that aren't vastly different in size.
While minimo targets Linux; it inherently is largely applicable to another environment - especially since they expect the front-end to be rewritten by someone using it in a real application.
Worldgate was going to use Mozilla for it's next-generation browse-the-web-on-your-cable-box application, where the browsers all run in servers at the headend and send screen images down to the settops as MPEG stills. We ran over 20 copies of Mozilla (tuned in ways similar to minimo) on 500Mhz P3's with 512MB of memory, and performance was reasonable. We lived with scroll bars where we had to (we subverted a few things to let pages fit tighter, but we also had to use larger-than-normal fonts). For added fun we had no mouse, but we had keyboards.
The toughest part was "geometric navigation" of links/etc with arrow keys; before development on that ended when we sold off our patents/business we'd mostly gotten that working, but there are more edge cases than you can count (nested and inline frames, imagemaps, etc).
I was a nielsen "family" a couple of years post-college (~1988) for sweeps. Probably skewed the ratings by working at home and coding all night long while watching Discovery and CNN.
:-) :-( "You watched the same news shows as a known terrorist! Come out for preventive deprogramming before we have to drag you out!" 1/2 :-)
A friend was a Nielsen family (wired) for a year, also a SW engineer. Realize that Nielsen probably also misses out on a fair number of the working poor - but I agree their samples can be pretty issue, especially for "niche" shows. They probably try to match nationwide demographics with their sample, but in things they don't ask about or level on it's probably significantly off.
The whole diary thing was a problem from the start, but with the first remote controls, then the explosion of channels (and other video sources) it got far worse. And, as you say, people have trouble remembering what they were watching something on, or write in things they wanted to watch or shows they want to "reward", or leave off things they "shouldn't" have watched. For example, I imagine far fewer admit to watch Friday-night Skinemax movies than actually watch them - especially the teenagers.
I wonder when Nielsen will start feeding data to the Matrix....
Ah, for the days when a good fraction of Usenet postings had NSA Food signatures... Though I don't think they're looking for references to Kremvax anymore.
We here (techies) understand how easy it is in hindsight to set up a scoring system so that the already-known bad guys end up on top. How well such an after-the-fact-designed (after 9/11) system is at picking out _other_ terrorists is unknown, and iffy at best (unless one of the inputs is "known to have trained in Afganistan with Al Queda"). They say that investigations were launched and "several arrests were made". It doesn't say that those arrests had anything to do with terrorism, or that they weren't later released or just deported for immigration violations. If they _were_ terrorist-related arrests, you can bet Siesint would be crowing about it and it would be part of Matrix now (even if it were pure chance or blind luck).
They even say it was dropped because it was 9/11-specific.
Your point is well taken - he may not be a C/C++/java/whatever programmer, and certainly he's most comfortable in Erlang - and as I mentioned, that's important to his point.
However, my point was that if a certain performance level was important, doing the initial implementation in a language (any of several) that was appropriate might have obviated the need to do multiple optimization passes, analysis, etc. I.e. a simpler algorithm implemented in a "faster" language might have run as fast as his 5-phase-optimized Erlang algorithm.
And yes, the performance of the development team matters a lot. There are all sorts of parameters that go into deciding how to solve a problem, obviously: team skills, tools, how often will this be run, what the requirements are, expected hardware, lifetime, maintenance (a simple algorithm in a "faster" language will have an advantage here over a more complex/heavily-optimized one in a slower language, for example), etc, etc.
In his "test" case, apparently on the order of 60+ fps was "good enough", though he was trying to illustrate what seemed general optimization methods - looking for work that can be factored out; work that can be collapsed; work that wasn't really needed; and ways to do the work faster (better O() algorithms) before micro-optimizing. And if you do optimize, make sure it's actually a part that's contributing to it being slow (profile, or analyze).
66 fps on a 3 GHz machine, doing a 600x600 simple RLE decode...
Ok, it's not bad for a a language like Erlang, but it's not exactly fast.
The big point here for the author is "it's fast enough". Lots of micro- (and macro-) optimizations are done when it turns out they aren't needed. And writing in a high level language you're comfortable in is important, if it'll do the job. This is a good point.
On the other hand, even a fairly naive implementation in something like C or C++ (and perhaps Java) would probably have acheived the goal without having to make 5 optimization passes (and noticable time examining behavior).
And even today, optimizations often do matter. I'm working on code that does pretty hard-real-time processing on multiple threads and keeps them synchronized while communicating with the outside world. A mis-chosen image filter or copy algorithm can seriously trash the rest of the system (not overlapping DMA's, inconvenient ordering of operations, etc). The biggest trick is knowing _where_ they will matter, and generally writing not-horrible-performance (but very readable) code as a matter of course as a starting point.
Disclaimer: I was a hard-core ASM & C programmer who for years beta-tested 680x0 compilers by critiquing their optimizers.
rdesktop on Linux/BSD works _really_ well. I pretty much no longer turn on the monitor on my XP machine. I can also bounce the X window over a tunnel to my machine at home, and control the XP machine at work from home.
tightVNC, while useful, has a tendency to crash.
I replaced both my and my wife's cases with Antec Sonatas a few weeks ago (we have both of them in the "office" section of our master bedroom at the moment). Quick summary: Nice Case! and quiet too!
In a bedroom in the woods, my machine is almost silent; the remaining noise it almost totally due to a slightly whiney WD 120GB 8-mb-cache drive. My video card (an NVidia GeForce 3 Ti-200) fan is either dead (possible) or is temp-aware. In any case I don't have a problem with it. My chipset fan is dead. After switching to the Sonata case, I found my old CPU fan was (a) hitting it's cage, and (b) noisy as a vacuum. I switched to a ThermalTake Volcano II+ (rebadged as Mad something, bought at CompUsa for ~$35) and that did the trick; I could barely tell the system was on other than the drive whine.
My wife's had a quieter CPU fan to start, but it's now the loudest thing in her system. Her motherboard/CPU are due for replacement, so we'll be rebuilding her machine with an ear to noise.
Even without that, we find the ReplayTV in the bedroom louder than both our PC's - and it's meant for use in A/V racks. Of course, on AVSForum there are a number of threads on how to quiet a ReplayTV even more....