Ugh. Regarding Gore/Bush, well, I loathe Gore and Bush is looking remarkably Clintonesque in his attempts to evade questions -- but that's another rant.
Another thing is that here, we don't use proportional representation. A single minority party could win 15% of the votes in all the House races, but not win a single seat; on the other hand, with massive numbers of parties and a relative rarity of runoffs (normally: plurality wins), a party could theoretically sweep both houses w/ 15% in all races if the rest of the votes were divided among many others.
My suspicion would be that a lot of GOPers would be absolutely thrilled if Nader ran for Pres. under the Green party, since that might be enough to deny the Democrats all the electoral votes of California. OTOH, you do get the occasional viable (at least at the state level) third-party candidacy, like Ventura and the Reform Party...
As for NOTA... I'm not sure that it would matter that much. Remember that already our President does not have much of a mandate, having won only something like 43% of the (popular, not electoral) votes cast, and a relatively low percentage of eligible voters actually voted. That's public information, and it doesn't seem to have hurt him in the slightest. Given that few would vote for NOTA (in all odds)...
Note to previous posters: crypt(1) != crypt(3), and is in fact a fully reversible, albeit rather weak, encryption (versus hashing/message-digest) implementation.
According to the man page for crypt.1 on this box (a sun4u running Solaris), it uses the same algorithm as ed, ex and vi in encryption mode. So, any of those should work...
When MS software works well, it can be very easy to use -- via installation wizards, deinstallation support (Add/Remove Program), and all sorts of configuration tools. I'll admit that it's pretty spiffy to be able to install a program, select checkboxes for the various components, and have it automatically added to menus.
The problems occur when it breaks -- and generally then the user has no recourse to fix it, precisely because so much is inherently automated for the user and has been made utterly inaccessible, except possibly through the Registry -- and that can be made difficult to reach if any problems occur during booting.
This can and does happen with even fairly vanilla hardware configurations, with results ranging from inability to install (as in the NT4 installer blue-screening deterministically while probing devices), to incredible inanity (like Win95 insisting that it knows (incorrectly, as it happens) what network card is installed and switching to a malfunctioning driver on every boot. When stuff like that happens, you tend to be SOL -- and get stuck with answers like "re-install your applications", "re-install Windows (and your applications)", "re-arrange your cards", and so forth.
A more charitable reading of that is, bug fixes do not constitute an entirely new version of software. That is, adding service packs / hot-fixes should normally not boost the version number.
That's not quite true, since certain SPs have mattered a lot in terms of functionality (e.g. IIRC, NT DX3 support came in a service pack...), but it's why you don't hear that suddenly MS released MS Windows NT 4.39.110+ or so. Release a patch, but it's not a full release; they're not going to ask stores to discard their stock and issue newly mastered versions; and it's not going to be billed as another "release".
That's the same way that, say, most Linux distributors (probably all), do not increment their version numbers for every single Errata patch and make sure to add more features before calling it a new release.
* I'll give you two guesses as to Holling's political party. Hint: Its symbol is an ass.
* How much mail-ordering and online-ordering do you think the poor do? Seriously, now..
Would you complain about a sales tax if it only applied to luxury goods? Which, incidentally, was repealed after revenues *dropped*; apparently the sponsors of that law never studied elementary economics or property rights & philosophy...
* Income taxes hurt the rich and employed. The employed have income, after all. In fact, given the AMT, you can make an extra $X but pay *more* than $X in additional taxes; or, you can marry somebody and instantly your total tax liability grows. Or, because you earn more, every $1 you make is worth far less than every $1 of somebody else, since the Gov't takes the rest. Sound fair to you? More like it promotes being poor, jobless, single and fertile.
Hm. It's not a GOP proposal, though, unless Sen. Hollings has suddenly chosen to defect.
The other thing is that this isn't your average sales tax. Odds are, most people buying online aren't your average pauper -- you're affecting a different market. How many people buy their basics (like groceries, etc) online? Not too many, so far. So it'll be hitting those who buy their books or cars online, in constrast to those who go down to their local Albertson's or Fred Meyer's.
This also hits the people who want the latest from fashion catalogs, or those who buy computers (caveat. Big computer companies probably won't be heavily affected, because they tend to have presences in many, many states).
That's in constrast to your average in-person-transaction sales tax, which hits those in everyday transactions and *is* blatantly regressive. Then again, so is the lottery, with the fundamental differences that a) the latter is completely voluntary, and b) it's mathematically a bad transaction...
* If you already pay state or local sales tax on the merchandise, that amount (up to 5%) is credited towards the tax. So it's not additional to state/local unless your s/l taxes are below that amount...
* Retailers that do business in your state, and are subject to taxing jurisdiction of the state, qualify as 'local merchants' and are excluded.
* The bill *does* specify a fund for education spending. Nominally, salaries, but states w/ above the average (mean, presumably) in teacher salaries (although it says nothing about adjustment for cost-of-living... !) can use the money for other educational purposes.
* It is an excise tax that only applies to products both bought and sold within this country. It's not attempting to tax international sales.
Reasons to go nuts:
* The funding can be withheld, basically at the Secretary's (read: President's) discretion. Read: blackmail opportunity.
* It includes a vague reference to excluding non-local transactions. Possibly, that'd make for an interesting political poker game as to what sort to exclude -- so more patronage.
How odd. Puerto Rico's explicitly included to benefit from the tax, but they won't pay it...
Oh, er, one more thing: Sec. 4059, Imposition of Tax, (c) specifically gives credit for sales tax paid regarding state/local sales tax. So, if that total already is at least 5%, it would appear that no additional tax is levied.
Neither sprintf nor sscanf has an ability to limit how many bytes it reads into, say, a (char*). This means that if you get input from a user (perhaps copying from argv, or the environment variables; or so forth...) and use these routines to copy 'em without checking: you lose.
--- char foo[10];
/* don't do this */ sprintf(foo, getenv("HOME")); --- for instance, or something to that effect, can copy more than 9 characters (plus NULL) to your fixed-size buffer. If you don't check and HOME is set to a very long string, you might lose *very* badly.
Hence, functions like snprintf, or using field length limits on scanf/sscanf -- but these aren't often used by people who aren't up on paranoid/defensive coding.
Re:The real loser isn't MS, its users who needed a
on
Hotmail Cracked Badly
·
· Score: 1
I believe the Finnish server you were referring to was an anonymous remailer service at anon.penet.fi; one that, if memory serves, anonymized both ways (one could anonymously send mail to a user of the service, as well).
Word is, that the service was shut down after the judicial system was used to disclose account information, after the Church of Scientology went after a disgruntled ex-member who was using anon.penet.fi. However, that might only have been possible since it was a remailer service, and thus had to know about the actual e-mail address if memory serves. Thus, the real (non-anonymous) account could be revealed.
Web-based system might change that, if the admins -- and users -- actually care about security and anonymity. Hotmail clearly does not, as it puts IP addresses in mail sent via itself -- addresses that could point to a whistleblower's work machine, for instance, and it also requires a bit of information for registration.
Just keep in mind that other programs don't have to come from MSFT to be coded badly. Remember the bad ol' days of Sendmail popping up on BUGTRAQ every so often, along w/ imapd and wuftpd? So switch if you like, but don't get too complacent and neglect to lock down a critical box.
You can have the safest OS in the world, and still have lousy security if a single privileged, network-accessible program is written with the slightest bit of carelessness...
At some schools, you apply as a graduate student and only after admissions do you attempt Ph.D. candidacy status; essentially, you have the option of going for either when you arrive.
At others, you're admitted directly to a specific M.S. or Ph.D. program, and switching may be problematic (although Ph.D. -> M.S. can perhaps be done if you meet requirements and specifically petition for it...).
M.S. programs are far less likely to guarantee funding for you, while some Ph.D. programs will (for everybody). OTOH, you might be there for 6-7 years in the latter case...
The job doors opened up by the degrees may vary. Keep that all in mind if you're not sure. You may not want to burn your bridges...
Seriously. While the oft-quoted top four (MIT, Stanford, CMU, Berkeley) are all good, it depends upon your intended focus. In certain areas, other schools are also superb choices. Graphics? From what I hear, GA Tech ain't bad at that... And so forth...
It also depends on peculiarities, like: Do you mind having to TA? Do you want to be right inside a major city? And so forth.
Think it's going to be expandable? Somehow, I've got my doubts. Call me a nay-sayer, but it strikes me that with a machine built mostly using *their* stuff, and apparently meant for consumers (not hard-core geeks), they don't intend for that box to come open very often.
Why make it easy to replace, oh, the audio with the latest from Voyetra Turtle Beach? Or the video card with something fancier from the high-end-CAD world? A *lot* of users have never installed software on their own (I'm not kidding!), let alone hardware; and they have an interest in people not buying competitor's products and flooding the online auction houses w/ used, Creative hardware.
But hopefully I'm just being waaaaaaaaaay too cynical here.:)
{shrug} Why would MS build MS-specific hardware, when they get the same thing done (with less DoJ scrutiny... doing something so blatant might bring down some heat, I'd think} by promoting software-driven modems, software-driven printers, and so forth?
There's *already* hardware that'll only run on, say Win9X (not even on NT...), and so forth -- and that trend does not seem to be ending.
* It might be considered interstate commerce. * It might be considered 'necessary and proper'. * The Feds pretty much always overstep their jurisdictional boundaries, even when it's blatantly absurd.
Er, if you think you might be, don't ask. It brings to mind the fool who called his local police, asking whether there were any warrants for his arrest (There were. They got used, pretty darn quick.).
I doubt that they'd let you know, since it'd imperil their investigation. Afterwards, they *might* have to let you know if you were tapped at some point (or at least, find out what they found out), but...
* On stocks, it wouldn't surprise me if a LOT of people held at least some -- especially given the availability of employee stock purchase plans.
Most people don't play a whole lot, nor do they have vast investments, but they may hold a couple. People who have cable TV might have a slightly higher average income than those that don't...
* The poor have less to spend, and most likely what they do spend on, will be necessities and products w/ low profit margins (high competition). While an executive might get an Audi or BMW and not really care that much about getting the best deal, you can't say as much for groceries. Generally, the former will have enough of a margin for a heavy shotgun-style ad budget...
...Hades, sometimes ads seem to be bought just for the *station* and not because of the program. I'm not sure how many people still watching old "ST:TNG" re-runs care that much about designer clothing or cosmetics... but when margins are high, a shotgun approach is affordable.
And that's why advertising aims towards the rich. You nailed it: money.
Given that the study claims to have tracked usage by time, this must have been somewhat controlled. Either people are reporting their own estimates of usage (which is naturally going to skew things), or monitoring software of some kind (browser plugin, otherwise) is being used -- and if the people know that they are being specifically monitored, this too might skew the results. In addition, such a study would be limited either by site, or ISP, or voluntary self-selection, or some other criterion that is not mentioned. I'm curious about the particulars.
As for Yahoo!, well, doing... what? Navigating through a hierarchy of links, trying to desperately escape Yahoo! and find specific content? Checking delayed stock quotes? What, in particular? If, say, they're snagging users via online games and e-mail accounts, that may matter more than those who simply grab the latest delayed (oxymoron? eh) stock quotes and don't care for any other services -- or advertising -- provided.
As for "contraction", sure. Seems plausible; some sites are going to be famous, and others not. More users are going to stay on the "famous" sites, now that they offer more services. Given that many new users mostly want basic information and services and aren't searching for, say, an online library of statistics papers to find the latest approaches to outlier detection on small datasets with no assumed prior distribution, or copies of the Federalist Papers, that's understandable.
Well, it's on their list (first, 'coz the different OSes are in alphabetical order, and no vendor name was given for Linux), presumably Intel itself is fairly confident that it'll not make them look like utter fools. It's going to be hard for them to justify Linux remaining on that list unless it demonstrably does better than, with, say, just 4 processors.
* use obscure software and setups that investigators would have a darn hard time figuring out?
* maintain backups?
* undo any eavesdropping devices they install?
* have, say, otherwise-legit anti-tampering countermeasures protecting the machine, room, and so forth?
For instance, suppose one used a fully cryptographic filesystem requiring a specific boot disk and some pretty darn secure authorization. Furthermore, use video systems to monitor the whole place. For the heck of it, rig the box to something spew dry ice or random debris all over the electronics if opened...
Would the owner be required to actually *be* there to disable 'em, leave handy instructions... or cheerfully watch, cackling gleefully as they try and fail?
If they botch the job (either tweaking my software configuration badly, or breaking something when they plant a bug), are they liable?
Tricky. Against the really clueful and paranoid, this might be fairly difficult; it'd be easier, perhaps, just to plant hidden cameras and such rather than learn all software configs.
More frequently, police might conduct what is technically ruled an illegal search; for instance, my memory is telling me that it is an open question whether police are allowed to search *passengers* when they stop a car. The driver, sure; the passengers, maybe not. If they find contraband on the passengers, then is this evidence valid? That's a hell of a lot more frequent and grey-area, and probably was what was meant.
In the cases you cite, the violations are pretty blatant. In my book (not a law book, just opinion...) in either case, the police officers, their department, and probably their city would all get nailed with a hefty lawsuit at the very minimum; the latter would probably result in jail time, and the evidence in that case would clearly be discarded as it was obtained under duress. The evidence in the break-in case is less disputable.
The article comes dangerously close to tarring all frequent users of computers as mentally-ill nerds.
Or, more cynically:
[rant] Psychologists and psychiatrists have a vested interest in labelling new syndromes, addictions and conditions -- the more 'conditions' a person can have, the more study is justified (and thus research grants), the more lectures that can be given and papers written, and the more exorbitantly-priced therapy to be dispensed. Unlike, say, oncology, standards of evidence need not be high. Witness the current dispensation of ADD diagnoses and associated drugs -- Ritalin, et al -- despite a continuing debate as to whether it is a *real* syndrome, or simply a convenient label for a common personality trait. This plays perfectly well with current trends towards holistic medicine, "victim" theory, and so forth, allowing people to bilk others of their time, money and dignity.
History lesson, folks: many 'syndromes' and conditions, even those with actual manifested syndromes, have been groundless. When was the last time you saw a dame suddenly faint? Yet, years ago, that was far more in vogue... Just a few years ago, it used to be 'the thing' for "therapists" to "uncover" allegedly repressed memories, usually of sexual abuse by parents. Well, that practice has been faded a tad after numerous court cases brought in which these "memories" were shown to be completely and utterly bogus, and often the results of suggestion from the therapists... and nowadays, there's a tendency to point at things and say, "That's evil."; "That's addictive"; "That's what's wrong with society today." Remember the L*******n shootings and the immediate quest for easy answers? Remember the press reporting 'bout how the suspects spent time online? And so forth. Odds are, the next 'massacre' or 'mass shooting' (in a country where shooting >2-3 people now becomes 'mass', whereas in others, terrorists beheading and otherwise mutilating scores of people with machetes is not that unusual...) that involves a 'nerd' will "explore" this even further.
Then there's the frequent argument that denial is confirmation: to paraphrase, "Thou doth protest too much". That barely even deserves a response, other than: "Have you stopped beating your wife?" Guilt does not manifest into denial automatically, and neither is evidence of the other. Heck, I'm being a rather arrogant bastard here, and I'm fully aware of that -- knowing that I can be like that when annoyed, which happens a tad too often... On the other hand, I fully deny secretly being the right-hand man for the CIA's Deputy Director of Operations -- ergo, I'm a spymaster, right? Nope.
As for their alleged evidence, a few examples do not a syndrome make. Are microwaves bad, because bozos have used them to -- messily -- dry their cats? That's not specifically 'microwave abuse' so much as 'cluelessness'. The fact that some people abuse foo does not mean that all those who use foo are "sick". Many people habitually speed: should we now list an addiction to speeding? Or jaywalking? In some of these cases, the underlying causes are simpler: pure ignorance, willful rejection of authority, and disregard of the consequences. If a marriage is happy and joyous, is somebody really going to turn to the 'net and reject his or her partner -- or is such behavior merely a manifestation of discontent? If students spend that much time online, does it not behoove us to ask -- "why"? And consider what else they might be doing, and whether it is specific to the 'net (i.e. take that away, and do you have a "normal" -- whatever that is -- individual, or does he shift his unusual focus onto something else?)
For just about any foo, there are people who will misuse it. Given that there are people so mind-bogglingly stupid enough to play a game in which they aim (pref. unloaded) automatic pistols at each other and see how fast they can pull the triggers (True. In the case I'm thinking of, the pair once used loaded weapons. One got Darwinned, and the other's paralyzed -- and facing criminal charges.) Is this the fault of the firearms, or simply a case of almost unbelievable idiocy? We've got, what, 6 billion people on the planet? Expect this.
But be sure before tarring millions with an alleged addiction. [/rant]
Fine. Y'all can moderate this down as 'Reactionary curmudgeon's flamebait from Hell' if y'like.
Slashdot Mirror
Ugh. Regarding Gore/Bush, well, I loathe Gore and Bush is looking remarkably Clintonesque in his attempts to evade questions -- but that's another rant.
Another thing is that here, we don't use proportional representation. A single minority party could win 15% of the votes in all the House races, but not win a single seat; on the other hand, with massive numbers of parties and a relative rarity of runoffs (normally: plurality wins), a party could theoretically sweep both houses w/ 15% in all races if the rest of the votes were divided among many others.
My suspicion would be that a lot of GOPers would be absolutely thrilled if Nader ran for Pres. under the Green party, since that might be enough to deny the Democrats all the electoral votes of California. OTOH, you do get the occasional viable (at least at the state level) third-party candidacy, like Ventura and the Reform Party...
As for NOTA... I'm not sure that it would matter that much. Remember that already our President does not have much of a mandate, having won only something like 43% of the (popular, not electoral) votes cast, and a relatively low percentage of eligible voters actually voted. That's public information, and it doesn't seem to have hurt him in the slightest. Given that few would vote for NOTA (in all odds)...
Note to previous posters: crypt(1) != crypt(3), and is in fact a fully reversible, albeit rather weak, encryption (versus hashing/message-digest) implementation.
According to the man page for crypt.1 on this box (a sun4u running Solaris), it uses the same algorithm as ed, ex and vi in encryption mode. So, any of those should work...
A note of caution:
When MS software works well, it can be very easy to use -- via installation wizards, deinstallation support (Add/Remove Program), and all sorts of configuration tools. I'll admit that it's pretty spiffy to be able to install a program, select checkboxes for the various components, and have it automatically added to menus.
The problems occur when it breaks -- and generally then the user has no recourse to fix it, precisely because so much is inherently automated for the user and has been made utterly inaccessible, except possibly through the Registry -- and that can be made difficult to reach if any problems occur during booting.
This can and does happen with even fairly vanilla hardware configurations, with results ranging from inability to install (as in the NT4 installer blue-screening deterministically while probing devices), to incredible inanity (like Win95 insisting that it knows (incorrectly, as it happens) what network card is installed and switching to a malfunctioning driver on every boot. When stuff like that happens, you tend to be SOL -- and get stuck with answers like "re-install your applications", "re-install Windows (and your applications)", "re-arrange your cards", and so forth.
A more charitable reading of that is, bug fixes do not constitute an entirely new version of software. That is, adding service packs / hot-fixes should normally not boost the version number.
That's not quite true, since certain SPs have mattered a lot in terms of functionality (e.g. IIRC, NT DX3 support came in a service pack...), but it's why you don't hear that suddenly MS released MS Windows NT 4.39.110+ or so. Release a patch, but it's not a full release; they're not going to ask stores to discard their stock and issue newly mastered versions; and it's not going to be billed as another "release".
That's the same way that, say, most Linux distributors (probably all), do not increment their version numbers for every single Errata patch and make sure to add more features before calling it a new release.
*Plonk*.
* I'll give you two guesses as to Holling's political party. Hint: Its symbol is an ass.
* How much mail-ordering and online-ordering do you think the poor do? Seriously, now..
Would you complain about a sales tax if it only applied to luxury goods? Which, incidentally, was repealed after revenues *dropped*; apparently the sponsors of that law never studied elementary economics or property rights & philosophy...
* Income taxes hurt the rich and employed. The employed have income, after all. In fact, given the AMT, you can make an extra $X but pay *more* than $X in additional taxes; or, you can marry somebody and instantly your total tax liability grows. Or, because you earn more, every $1 you make is worth far less than every $1 of somebody else, since the Gov't takes the rest. Sound fair to you? More like it promotes being poor, jobless, single and fertile.
Software prices actually do go down for old products -- or, perhaps, you're saying that your local stores sell old software at full price?
Also, occasionally something like Quattro Pro gets marked down... it does happen, one somebody wants more market/mind share.
Hm. It's not a GOP proposal, though, unless Sen. Hollings has suddenly chosen to defect.
The other thing is that this isn't your average sales tax. Odds are, most people buying online aren't your average pauper -- you're affecting a different market. How many people buy their basics (like groceries, etc) online? Not too many, so far. So it'll be hitting those who buy their books or cars online, in constrast to those who go down to their local Albertson's or Fred Meyer's.
This also hits the people who want the latest from fashion catalogs, or those who buy computers (caveat. Big computer companies probably won't be heavily affected, because they tend to have presences in many, many states).
That's in constrast to your average in-person-transaction sales tax, which hits those in everyday transactions and *is* blatantly regressive. Then again, so is the lottery, with the fundamental differences that a) the latter is completely voluntary, and b) it's mathematically a bad transaction...
Reasons not to panic:
* If you already pay state or local sales tax on the merchandise, that amount (up to 5%) is credited towards the tax. So it's not additional to state/local unless your s/l taxes are below that amount...
* Retailers that do business in your state, and are subject to taxing jurisdiction of the state, qualify as 'local merchants' and are excluded.
* The bill *does* specify a fund for education spending. Nominally, salaries, but states w/ above the average (mean, presumably) in teacher salaries (although it says nothing about adjustment for cost-of-living... !) can use the money for other educational purposes.
* It is an excise tax that only applies to products both bought and sold within this country. It's not attempting to tax international sales.
Reasons to go nuts:
* The funding can be withheld, basically at the Secretary's (read: President's) discretion. Read: blackmail opportunity.
* It includes a vague reference to excluding non-local transactions. Possibly, that'd make for an interesting political poker game as to what sort to exclude -- so more patronage.
How odd. Puerto Rico's explicitly included to benefit from the tax, but they won't pay it...
Oh, er, one more thing: Sec. 4059, Imposition of Tax, (c) specifically gives credit for sales tax paid regarding state/local sales tax. So, if that total already is at least 5%, it would appear that no additional tax is levied.
Sure. But there's nothing illegal with that; the one concern might be jurisdiction (since it's not interstate commerce in the traditional sense).
You're already most likely triple-taxed on your income, anyway (city/state/Federal).
Neither sprintf nor sscanf has an ability to limit how many bytes it reads into, say, a (char*). This means that if you get input from a user (perhaps copying from argv, or the environment variables; or so forth...) and use these routines to copy 'em without checking: you lose.
---
char foo[10];
/* don't do this */
sprintf(foo, getenv("HOME"));
---
for instance, or something to that effect, can copy more than 9 characters (plus NULL) to your fixed-size buffer. If you don't check and HOME is set to a very long string, you might lose *very* badly.
Hence, functions like snprintf, or using field length limits on scanf/sscanf -- but these aren't often used by people who aren't up on paranoid/defensive coding.
I believe the Finnish server you were referring to was an anonymous remailer service at anon.penet.fi; one that, if memory serves, anonymized both ways (one could anonymously send mail to a user of the service, as well).
Word is, that the service was shut down after the judicial system was used to disclose account information, after the Church of Scientology went after a disgruntled ex-member who was using anon.penet.fi. However, that might only have been possible since it was a remailer service, and thus had to know about the actual e-mail address if memory serves. Thus, the real (non-anonymous) account could be revealed.
Web-based system might change that, if the admins -- and users -- actually care about security and anonymity. Hotmail clearly does not, as it puts IP addresses in mail sent via itself -- addresses that could point to a whistleblower's work machine, for instance, and it also requires a bit of information for registration.
Just keep in mind that other programs don't have to come from MSFT to be coded badly. Remember the bad ol' days of Sendmail popping up on BUGTRAQ every so often, along w/ imapd and wuftpd? So switch if you like, but don't get too complacent and neglect to lock down a critical box.
You can have the safest OS in the world, and still have lousy security if a single privileged, network-accessible program is written with the slightest bit of carelessness...
Warning:
At some schools, you apply as a graduate student and only after admissions do you attempt Ph.D. candidacy status; essentially, you have the option of going for either when you arrive.
At others, you're admitted directly to a specific M.S. or Ph.D. program, and switching may be problematic (although Ph.D. -> M.S. can perhaps be done if you meet requirements and specifically petition for it...).
M.S. programs are far less likely to guarantee funding for you, while some Ph.D. programs will (for everybody). OTOH, you might be there for 6-7 years in the latter case...
The job doors opened up by the degrees may vary.
Keep that all in mind if you're not sure. You may not want to burn your bridges...
Disclaimer: I'm arguably biased, being at CMU...
Seriously. While the oft-quoted top four (MIT, Stanford, CMU, Berkeley) are all good, it depends upon your intended focus. In certain areas, other schools are also superb choices. Graphics? From what I hear, GA Tech ain't bad at that... And so forth...
It also depends on peculiarities, like: Do you mind having to TA? Do you want to be right inside a major city? And so forth.
Think it's going to be expandable? Somehow, I've got my doubts. Call me a nay-sayer, but it strikes me that with a machine built mostly using *their* stuff, and apparently meant for consumers (not hard-core geeks), they don't intend for that box to come open very often.
:)
Why make it easy to replace, oh, the audio with the latest from Voyetra Turtle Beach? Or the video card with something fancier from the high-end-CAD world? A *lot* of users have never installed software on their own (I'm not kidding!), let alone hardware; and they have an interest in people not buying competitor's products and flooding the online auction houses w/ used, Creative hardware.
But hopefully I'm just being waaaaaaaaaay too cynical here.
{shrug} Why would MS build MS-specific hardware, when they get the same thing done (with less DoJ scrutiny... doing something so blatant might bring down some heat, I'd think} by promoting software-driven modems, software-driven printers, and so forth?
There's *already* hardware that'll only run on, say Win9X (not even on NT...), and so forth -- and that trend does not seem to be ending.
IANAL.
but:
* It might be considered interstate commerce.
* It might be considered 'necessary and proper'.
* The Feds pretty much always overstep their jurisdictional boundaries, even when it's blatantly absurd.
Er, if you think you might be, don't ask. It brings to mind the fool who called his local police, asking whether there were any warrants for his arrest (There were. They got used, pretty darn quick.).
I doubt that they'd let you know, since it'd imperil their investigation. Afterwards, they *might* have to let you know if you were tapped at some point (or at least, find out what they found out), but...
* On stocks, it wouldn't surprise me if a LOT of people held at least some -- especially given the availability of employee stock purchase plans.
Most people don't play a whole lot, nor do they have vast investments, but they may hold a couple.
People who have cable TV might have a slightly higher average income than those that don't...
* The poor have less to spend, and most likely what they do spend on, will be necessities and products w/ low profit margins (high competition). While an executive might get an Audi or BMW and not really care that much about getting the best deal, you can't say as much for groceries. Generally, the former will have enough of a margin for a heavy shotgun-style ad budget...
...Hades, sometimes ads seem to be bought just for the *station* and not because of the program. I'm not sure how many people still watching old "ST:TNG" re-runs care that much about designer clothing or cosmetics... but when margins are high, a shotgun approach is affordable.
And that's why advertising aims towards the rich. You nailed it: money.
Given that the study claims to have tracked usage by time, this must have been somewhat controlled. Either people are reporting their own estimates of usage (which is naturally going to skew things), or monitoring software of some kind (browser plugin, otherwise) is being used -- and if the people know that they are being specifically monitored, this too might skew the results. In addition, such a study would be limited either by site, or ISP, or voluntary self-selection, or some other criterion that is not mentioned. I'm curious about the particulars.
As for Yahoo!, well, doing... what? Navigating through a hierarchy of links, trying to desperately escape Yahoo! and find specific content? Checking delayed stock quotes? What, in particular? If, say, they're snagging users via online games and e-mail accounts, that may matter more than those who simply grab the latest delayed (oxymoron? eh) stock quotes and don't care for any other services -- or advertising -- provided.
As for "contraction", sure. Seems plausible; some sites are going to be famous, and others not. More users are going to stay on the "famous" sites, now that they offer more services. Given that many new users mostly want basic information and services and aren't searching for, say, an online library of statistics papers to find the latest approaches to outlier detection on small datasets with no assumed prior distribution, or copies of the Federalist Papers, that's understandable.
Well, it's on their list (first, 'coz the different OSes are in alphabetical order, and no vendor name was given for Linux), presumably Intel itself is fairly confident that it'll not make them look like utter fools. It's going to be hard for them to justify Linux remaining on that list unless it demonstrably does better than, with, say, just 4 processors.
That is, would it become illegal to...
* use obscure software and setups that investigators would have a darn hard time figuring out?
* maintain backups?
* undo any eavesdropping devices they install?
* have, say, otherwise-legit anti-tampering countermeasures protecting the machine, room, and so forth?
For instance, suppose one used a fully cryptographic filesystem requiring a specific boot disk and some pretty darn secure authorization. Furthermore, use video systems to monitor the whole place. For the heck of it, rig the box to something spew dry ice or random debris all over the electronics if opened...
Would the owner be required to actually *be* there to disable 'em, leave handy instructions... or cheerfully watch, cackling gleefully as they try and fail?
If they botch the job (either tweaking my software configuration badly, or breaking something when they plant a bug), are they liable?
Tricky. Against the really clueful and paranoid, this might be fairly difficult; it'd be easier, perhaps, just to plant hidden cameras and such rather than learn all software configs.
More frequently, police might conduct what is technically ruled an illegal search; for instance, my memory is telling me that it is an open question whether police are allowed to search *passengers* when they stop a car. The driver, sure; the passengers, maybe not. If they find contraband on the passengers, then is this evidence valid? That's a hell of a lot more frequent and grey-area, and probably was what was meant.
In the cases you cite, the violations are pretty blatant. In my book (not a law book, just opinion...) in either case, the police officers, their department, and probably their city would all get nailed with a hefty lawsuit at the very minimum; the latter would probably result in jail time, and the evidence in that case would clearly be discarded as it was obtained under duress. The evidence in the break-in case is less disputable.
The article comes dangerously close to tarring all frequent users of computers as mentally-ill nerds.
Or, more cynically:
[rant]
Psychologists and psychiatrists have a vested interest in labelling new syndromes, addictions and conditions -- the more 'conditions' a person can have, the more study is justified (and thus research grants), the more lectures that can be given and papers written, and the more exorbitantly-priced therapy to be dispensed. Unlike, say, oncology, standards of evidence need not be high. Witness the current dispensation of ADD diagnoses and associated drugs -- Ritalin, et al -- despite a continuing debate as to whether it is a *real* syndrome, or simply a convenient label for a common personality trait. This plays perfectly well with current trends towards holistic medicine, "victim" theory, and so forth, allowing people to bilk others of their time, money and dignity.
History lesson, folks: many 'syndromes' and conditions, even those with actual manifested syndromes, have been groundless. When was the last time you saw a dame suddenly faint? Yet, years ago, that was far more in vogue... Just a few years ago, it used to be 'the thing' for "therapists" to "uncover" allegedly repressed memories, usually of sexual abuse by parents. Well, that practice has been faded a tad after numerous court cases brought in which these "memories" were shown to be completely and utterly bogus, and often the results of suggestion from the therapists... and nowadays, there's a tendency to point at things and say, "That's evil."; "That's addictive"; "That's what's wrong with society today." Remember the L*******n shootings and the immediate quest for easy answers? Remember the press reporting 'bout how the suspects spent time online? And so forth. Odds are, the next 'massacre' or 'mass shooting' (in a country where shooting >2-3 people now becomes 'mass', whereas in others, terrorists beheading and otherwise mutilating scores of people with machetes is not that unusual...) that involves a 'nerd' will "explore" this even further.
Then there's the frequent argument that denial is confirmation: to paraphrase, "Thou doth protest too much". That barely even deserves a response, other than: "Have you stopped beating your wife?" Guilt does not manifest into denial automatically, and neither is evidence of the other. Heck, I'm being a rather arrogant bastard here, and I'm fully aware of that -- knowing that I can be like that when annoyed, which happens a tad too often... On the other hand, I fully deny secretly being the right-hand man for the CIA's Deputy Director of Operations -- ergo, I'm a spymaster, right? Nope.
As for their alleged evidence, a few examples do not a syndrome make. Are microwaves bad, because bozos have used them to -- messily -- dry their cats? That's not specifically 'microwave abuse' so much as 'cluelessness'. The fact that some people abuse foo does not mean that all those who use foo are "sick". Many people habitually speed: should we now list an addiction to speeding? Or jaywalking? In some of these cases, the underlying causes are simpler: pure ignorance, willful rejection of authority, and disregard of the consequences. If a marriage is happy and joyous, is somebody really going to turn to the 'net and reject his or her partner -- or is such behavior merely a manifestation of discontent? If students spend that much time online, does it not behoove us to ask -- "why"? And consider what else they might be doing, and whether it is specific to the 'net (i.e. take that away, and do you have a "normal" -- whatever that is -- individual, or does he shift his unusual focus onto something else?)
For just about any foo, there are people who will misuse it. Given that there are people so mind-bogglingly stupid enough to play a game in which they aim (pref. unloaded) automatic pistols at each other and see how fast they can pull the triggers (True. In the case I'm thinking of, the pair once used loaded weapons. One got Darwinned, and the other's paralyzed -- and facing criminal charges.) Is this the fault of the firearms, or simply a case of almost unbelievable idiocy? We've got, what, 6 billion people on the planet? Expect this.
But be sure before tarring millions with an alleged addiction.
[/rant]
Fine. Y'all can moderate this down as 'Reactionary curmudgeon's flamebait from Hell' if y'like.