Different people have different goals and different thought processes. Some web designers think that a layout that makes sense to them, when they've got the entire site map in their head and know what all the parts mean, is useful to other people. A few of those designers are even correct, but mostly for small simple sites without automation:-) In general, some people who make badly designed web sites know and don't care, but most bad web designers don't know that they're bad, or who they're bad for, but they do care. A site map is really easy to add to any site that isn't totally badly designed, and doesn't interfere with a good site. You might as well use one, in case you're one of those bad web site designers who doesn't know it.
Site maps are useful for some readers, search engines are useful for other readers, good well-designed links are best, good FAQs are useful for other readers. It's obvious that a good search engine is hard to implement, but if you make your pages easily searchable by Google, that's at least a start.
" But this format will (of course) only represent the appearance, not any structure."WHAT!?!?! Do they still not bloody get the bloody concept, or are they deliberately trying to make interoperability unusable? They did this in earlier versions of Office with their save-as-html modes, which did stupid things like saving a "Header Type 2" as "14-point-boldface-text" or whatever your current style was rather than saving it as an HTML "H2", but at that point it could be attributed to stupidity and/or incompetence, since some people think for some reason that HTML is an appearance description language rather than an specific implementation instance of a content description metalanguage, which is a bit too abstract for some people. But XML is much more explicit about being a content description metalanguage, and if you've got enough of a clue about it to output your material as XML, you've got to get that much of the concept. I'd attribute this one to malice.
Is the cap on your upload or your download? Many of the US cable modem companies cap everybody's upload at 128kbps (if their equipment is new enough) or something like 384-768 with older hardware, so a 128kbps cap is basically no change.
Capping your *download* speeds to 128kbps would get your attention, though:-) Some cable companies, who are either "ahead of the curve" or "even more terminally clueless than most about why people buy broadband", have monthly download quotas of a couple of GB and drop your bandwidth to 56kbps or so once you exceed them. That's fine for regular web viewing, probably marginal for average Napster use, and totally useless if you want to download the latest Linux release. (On the other hand, some cable companies start surcharging you by the GB if you exceed the limit, so you can end up with megabuck charges if you weren't paying attention; I'd rather get slowed down than risk that.)
DSL is available in a fairly high fraction of non-rural residences in the US; your distance from the telco office determines what speeds you can get, and it of course depends on what DSL technology is being used, but typical limits are 12-18000 feet for 384kbps or various ADSL speeds, and about 30000 feet for IDSL (which uses ISDN technology, so it's a boring 128/144kbps, but better than modems.) If you want the newer up-to-8Mbps DSL flavors, you need to be much closer, and there are a lot fewer POPs supporting, and the pricing is usually a lot higher.
I'm getting about 800-900kbps ADSL at a distance I've been told is somewhere between 12000 and 16000 feet, depending on who measured it (I think it's really about 12-14000 feet but bad wiring...). A few years ago, I had 384kbps SDSL here, but 768kbps SDSL didn't work.
The advertising biz has been trying to understand the Internet, and the web, and email, since the mid-90s. It's a really new field, and unlike television, the interactions between parties are much more complex. One of the big contributors to the dot-com boom and bust was the market's exploration of banners and similar advertising
Will banners reach consumers?
What kinds of messages are they good for?
What's the right activity to measure and price? Banner impressions? Clickthroughs? Something else?
How much is a banner impression worth? How long will it stay worth that much?
How can you measure advertising success to sell to ad-buyers?
How do you measure it without somebody's robo-clicker tricking you into paying them lots/no money?
Can advertising generate enough money to fund web sites?
How much content can you afford to build with your advertising revenue?
How much new fresh content do you need to develop to get enough readers to make the ads pay for your fixed and variable costs?
Hey, this was a great business model last year! What do you mean the market's saturated?
What are the supply and demand curves like?
How much is it worth to be Cool Site Of The Day/Week/Month/Year/Decade? Can your site stay cool for more than 15 minutes?
How much is it worth to be Cool Advertising Technique Of The Day? Are you cool because of fundamentals, or is it just your lucky 15 minutes?
If the social phenomena that your business model depends on are going to be short-term, can you grab the wave long enough to make back your fixed costs before something else is cool?
Does deeply thought out advertising get 30 minutes of fame instead of just 15, and does it cost a lot more than a quick+dirty job that only gets you 7.5 minutes?
What tools can we sell to the people who are doing the advertising, the content building, the ad-blocking, etc?
How long will the customers for those tools be able to afford them?
If the ad-buyers are only selling Internet-related products, as opposed to cars and movies and consumer electronics toys, is this just a bubble that'll disappear once we've burned all the VC money? This is a rapidly changing business, and the one thing that everybody's sure of is that
It's not your machine - it's the one at the other end of your loopback network:-) I tend to use it to indicate sites I'm killing off.
The main problem with 127.0.0.1 is that if you're also running a real web server on your machine, the requests for blocked sites will be sent to it, so your system will have to respond in some appropriate manner, and your browser will have to display the response appropriately. Some operating systems don't seem to have the clue that 127.0.0.2 is different from 127.0.0.1:-) I haven't installed a web server on my main work machine since it got upgraded (?) to Win2000, but older Windows used to be a bit fuzzy about the distinction.
This isn't a popup blocker you're installing - those are for those poor suckers using IE who need all the protection they can get. This is just choosing the option that implements or doesn't implement popup windows, telling it you don't want the things. Works real fine; the last time I was on a machine that didn't have Mozilla, I was really appalled at what IE users have to put up with.
Phoenix Mozilla Browser doesn't load them
on
Next-Gen Pop-up Ads
·
· Score: 2
Phoenix is a Mozilla derivative, designed to be fast and lightweight. It has the Mozilla feature of "decide whether to allow scripts to open windows without permission", which of course I set to not let them. (I wouldn't call this a popup-blocker - it's more of a popup-non-supporter...) Whatcar.com didn't pop up anything, even though I tried a few pages, and I never see X10 popups either.
Now, it's possible that I don't have quite all of the Flash/Schlockwave plugins reinstalled correctly since the last time I installed a new rev of Phoenix:-) I'm running 0.5, and I've had some problems with some plugins not working, since their installers seem to want Real Netscape, but most are ok, and about:plugins claims that I've got Shockwave 6.something installed.
Eighteenth-Century Britain had a popular investigative technique called "We'll keep torturing you until you confess". One of the traditional methods was to keep piling heavy rocks on the accused until he either confessed or died; I've forgotten which defiant holdout's last words were "More weight!". This sort of thing wasn't a new invention of the time, and the Brits weren't the only people who used it (nobody expected the Spanish Inquisition...), and it was more popularly used on political enemies, accused heretics and witches, and people who had "accomplices" on whom it was desired that they should rat, rather than on common criminals, who usually had the sense to confess or frame someone else, especially since they were often actually guilty of something. That's the main reason for the Fifth Amendment. That doesn't mean we've totally abandoned the practice - cops still beat people up or threaten to do things to them or their families - but it's certainly reduced the problem, and at least it's a rare illegal event in the back room rather than a common event on the courtroom floor.
American jurisprudence also has a bunch of 1960s practices like the Exclusionary Rule and the Miranda Warning which say that courts can't admit evidence that was acquired improperly, whether it was from beating prisoners until they confess, illegally searching homes without warrants, or getting warrants by lying to judges, or lying to prisoners about the law when they don't have lawyers to advise them. Again, it didn't totally eliminate abuses, but the traditional example for its effectiveness is that the year before the Exclusionary Rule, police in New York City didn't bother getting any search warrants - they just illegally searched anybody and any place they wanted to, while the year after the rule, they almost always got warrants when they needed them (even if they still lied about their evidence on occasion.)
"It might simply be an informant you've been working with for years, goes home at night, gets on the Internet and says 'you know, I think that informant we've been covering for for years while he murdered his competition and tipped off that we might be about to indict him, and that we recently had to indict some of the FBI agents who helped cover it up, I wonder if he might be a mobster? Nah, let's try to distract people by holding a news conference about Internet ads.'", US Attorney Michael Sullivan carefully refrained from saying during the news conference.
The best case is if the crook hasn't fenced your Mac yet so you can get it back. Otherwise good luck collecting. Sometimes crooks actually have money, but often the reason they're crooks is that they have expensive habits, like gambling or cocaine or rent or Mafia loans, that they can't finance legitimately.
This guy probably took the computer (or at least some of the computers from other people he'd ripped off) and sold it for half the price he "bought" it for, maybe less, so even if he hasn't poured the money up his nose, he'd only have $1200, not 2500. Maybe he's got a car, though....
I was using Win98 (I think Second Edition, though it could have been First Edition), and it worked just fine with Netscape 4.x in 24MB. But with Netscape 6 it croaked horribly. The hardware was a 300 MHz Celeron. I later upgraded the box from 24MB to 192MB, and it worked a bit better:-) One of the other machines got upgraded to 640MB, because it seemed like that ought to be enough memory for anybody.
As somebody else replied, this level of memory bloat is pretty recent - for most kinds of software, you shouldn't be designing it to require a latest-model desktop power box, but a two-year-old laptop, which has a lot less spare resources. That way it will run well for everybody and really rock on high-end machines. Some applications are obvious exceptions, like games, and scientific/engineering numbercrunching.
So what does it take to set one off? A good hard punch to the sensor? Thrown beer bottle? How embarassing is that!? Sure, they're designed to go off when they need to and not go off accidentally, but what about deliberate attempts to trigger them? If they ever become popular, will this become as good a way to start a bar brawl as walking up an insulting a guy's bike? Hey, drivin' an underpowered riceburner like that, I bet you really *need* to wear this airbag!
Netscape 6 was bad, bad, mindbogglingly amazingly bad, at least on a machine with only 24MB of RAM. 6.1 or 6.2 was much closer to usable, at least on my machines with 64MB, but by then Mozilla was working adequately. If Netscape 7.0 is fast, a large part of it is probably from using lots of memory to accelerate other functions.
I'm now using Phoenix 0.5, which came out just recently, and it's quite toasty - I think it's ready to replace Mozilla as my main browser. The main plugins work (I'd had trouble getting them installed on 0.3 and 0.4) and it's very very fast, especially since I set the startup delay to 0 (default is 1200ms, which lets it recover from slow-loading graphics that would otherwise force redraws.) The Google-search-bar extension is really convenient, though I gather than newer Mozillas also have it. I'm normally no fan of themes (why clutter up the GUI at the cost of making it larger and slower?), but the "LittlePhoenix 1.3" theme has icons that are enough smaller that I can reclaim significant screen space, and the "Linky" extension has been a good way to handle pages with lots of links (e.g. letting you leech all the pictures into a separate window or tab, or examine a page by grabbing all the URLs on it into a tab, which can be cleaner than View Source for some ugly web pages.)
Several people have pointed out that the cost of laying a big bundle of fiber isn't much different from laying a single pair of fibers, and it's dumb to use expensive optical amplifiers on fibers you don't need to use yet, plus expensive optical muxes and line-terminating equipment at the ends, especially when a pair of fibers can run up to 160 wavelengths, of up to OC192. One reason it makes financial sense to roll out lots of fibers, especially in metropolitan areas, is that you get much more flexibility about customer locations, and as long as you're not exceeding distance limitations (which depend on the type of signals you're sending) it's a lot cheaper to run extra fiber and use it inefficiently, because fiber splices are much cheaper than optical amplifiers or active electronics or wavelength-division multiplexers. It's also a lot easier for a carrier to handle selling to multiple customers (who are often other carriers, not just end-users) if you can give them their own fiber rather than wavelengths. For long-haul routes, or big metro rings, it's worth putting more active equipment in and using more wavelengths.
Another issue is that much of the Oregon build-out happened late in the boom - if you look at the Boston-NewYork-WashingtonDC routes, which were developed earlier, you'll find that there's less multiplexing and more sale of individual whole fibers, because the equipment costs were higher, but they did a Moore's Law type crash dive during the late 90s and early 00s that Oregon benefitted more from.
Meanwhile, if you're trying to get a data feed to every lottery terminal in Oregon, you don't care about 160-wavelength x 10Gbps OC192 fibers - you care about getting 4kbps worth of data on whatever kind of copper wire or wireless goes out to the convenience store, and doing the protocol conversion because lottery terminals still speak X.25 and nobody wants to pay the capital costs to teach the things IP:-)
Back in May 2001, there was an article in the NY Times(copy) and Slashdot (also mentioned on DeadMedia.org) about a project to run fiber optics through pneumatic tubes in New York and other big cities. While the meme is out there, it's not clear that anybody's actually implemented it. One problem, besides the financial issues, and the World Trade Center collapse in the most interesting market area, is that real ownership of the tubes is vary unclear, at least in New York City.
*Everything* is a horrible archival medium. The IDE drives will support you just fine while you're starting to address the hard problems. Don't bother with 100MB-Ether if you can conveniently do GigE though - it's gotten really cheap.
Data format rot and hardware medium support rot will kill your data much more thoroughly than equipment failure will.
Only one of the writable-DVD formats will still be readable in 5 years - I don't know which one, because it will be a market support issue, not a hardware reliability issue. Copy protection may make that worse.
The file system format your data is stored in won't be readable by Windows in five years, though it will probably be readble by Linux.
Any data format that isn't basically text will be obsolete - HTML and XML will be readable, but word processor files will be almost totally useless, and at least half of your non-text-like database formats will be mostly useless. You can keep a copy of MSWord and Excel, and the original CDROM installation media, but it won't run on Windows 2007XYZ, and your WordStar 5.25" floppies aren't much use today either. Closed-source will rot faster than open-source, but both will rot eventually.
That 6GB disk drive you bought three years ago isn't worth using, but the data fits nicely in a corner on that $120 120GB drive you picked up last week. The 120GB drives will be pretty boring in 5 years, whether they're IDE or SCSI, but the data will fit nicely in a corner on the old 1TB drive you'll be using for spare storage in a couple of years or the newer 8GB drive, but you may have a bit of trouble reading those tiny little 8KB blocks from the wimpy 2MB cache.
So yes, use some kind of IDE RAID system for your backups, and be sure to save some backup computers and operating systems every year or two to read the old disks.
Depending on how much of your data is files as opposed to databases, come up with some reasonable strategy for full and incremental backups, and what stays on-line, on-site-but-off-line, and off-site, and if possible, keep a computer off-site as well.
Keep a database of what you've got backed up, and where.
At least annually, copy at least some of your full backups from old disks to new ones, and depending on your incremental or journaling structure, maybe more than that.
Make ABSOLUTELY sure your financial people have their records in some format that's readable and usable, and review with them every year to make sure their backups are in a usable format. Audits of seven-year-old financial data are no fun even when you *can* find the supplies for/dev/iso-9000-clay-tablet-scriber and programs that read the old-format sound files to output carrier pigeon calls to/dev/audio.
Maybe you're buying good SCSI disks and el-cheapo IDE drives, but the real difference in them is the controller cards, not the disks themselves. If you're using the disks for archive, not for active use, they're not going to be under the same load or wear&tear as the primary service disks. And you're probably going to RAID-5 them anyway.
The real problem isn't usually hardware rot - it's data format rot and interface standards rot. IDE is a bit tired, but it'll be supportable much longer than N-1 of the current DVD-R standards, so you probably want both. In my attic, I've got a probably-good 9-track backup tape of some of my early work, and some newer probably-good Sun cartridge tapes, and I might still have a Sun tape reader, if I can find a copy of SunOS 3.5 for my Sun-2 diskless workstation to bootstrap something to read it with.... Not a high priority - anything useful I did back in those days is probably on a Usenet archive somewhere.
We had fun last week dissing Alan Ralsky, the spammer that everybody's now signing up for spam. Previous Slashdot articles on harassing spammers point out that Overture.com accepts bids from advertisers for top positions in their search results, and the top three positions get sold on Google, MSN, etc. - So if you search for "bulk email" and click the first couple of links, each one costs the advertiser (who's presumably a spammer) whatever their bid is - typically a few bucks. For some reason, "bulk email" is having a bidding war - today's prices were over $25. The system is designed to detect multiple clickthroughs by the same person (which is why I'm not providing a direct link), but once you've got all those ads for spamware on your screen, you might as well give them the name of a promising lead - like Ralsky...
Look, I'm not *advocating* that anybody go shoot Ashcroft or Poindexter, and Ashcroft even seems to get along well with the NRA. Nor do I own guns. This is a discussion about political theory, and they're some of the most recent examples of the types of people that the authors of the Bill of Rights had in mind when they wrote the Second Amendment, just as they are recent examples of the types of people the authors had in mind when they wrote the First, Fourth, Fifth, and Sixth, and Disgraced Ex-Admiral Poindexter and his henchperson Ollie North are examples of the type of person the authors of the Uniform Code of Military Justice had in mind when they wrote the part *requiring* US military personnel to disobey illegal orders.
Yeah. Shame we couldn't mark the whole thread as "flamebait", instead of just the articles:-) Some of us believe strongly in self-protection as a basic human right, others believe strongly in the opposite, some people are seriously confused about whether guns give you more or less ability to protect yourself, and you can *forget* the Europeans (:-), who seem to believe that an armed government can run a polite society, or the Americans, who don't believe in polite societies.
The revolting colonists also had rifles, mainly from Pennsylvania and Kentucky, and cannon, and one of the political issues early in the revolt was that His Majesty's Legitimate Government of North America attempted to seize cannons from some towns. Even though the Constitution was a later compromise with people who wanted more power than they'd had under the Confederation, they still remembered overthrowing the Brits. They recognized that guns were not only essential for hunting and for defending families and livestock against bears and wolves, they were needed for fighting with Native Americans who didn't like their land being stolen, and for shooting people like Ashcroft and Poindexter. The Feds defeated most of the Native Americans and stole most of their land, wolves have pretty much been wiped out in the Middle 48 states, bears can only be hunted in season, but we seem to keep coming up with more people like Ashcroft and Poindexter.
Meanwhile, when liberal gun controllers start insisting that gun control laws shouldn't be enforced by armed policemen, I'll start taking them seriously. Until then, they're just saying that I should trust them with guns but they don't trust me. For liberals, that's hypocrisy, and it's amazing that they can say it with a straight face. Conservatives say that sort of thing all the time, of course, but it's *consistent* for somebody who's in power to be honest about wanting to stay in power....
They're loud wimps, and there are a lot of them, so they've been able to slow down the gun control folks more effectively than some groups, but they've done a lot of compromise over the years on issues like registration, instant background checks (as an alternative to waiting periods), government-funded pro-hunting programs (the NRA likes them) and the like.
ACLU President Nadine Strossen has spoken about guns in the past, saying that she leans toward the position that the second amendment identifies gun ownership is an individual civil right, not merely a collectivist permission to be part of the National Guard or for Guardsmen to keep their rifles at home like the Swiss. But the ACLU isn't a monolithic organization - they have a lot of central resources, and get involved in Supreme Court cases, but their real work is done by local chapters, who come up with lawyers to defend people in most of the cases that they work on. So if you want the ACLU to defend gun rights, get involved, get your law degree (:-), and find cases that you can convince your local organization that it makes sense to work on.
Slashdot Mirror
Site maps are useful for some readers, search engines are useful for other readers, good well-designed links are best, good FAQs are useful for other readers. It's obvious that a good search engine is hard to implement, but if you make your pages easily searchable by Google, that's at least a start.
" But this format will (of course) only represent the appearance, not any structure." WHAT!?!?! Do they still not bloody get the bloody concept, or are they deliberately trying to make interoperability unusable? They did this in earlier versions of Office with their save-as-html modes, which did stupid things like saving a "Header Type 2" as "14-point-boldface-text" or whatever your current style was rather than saving it as an HTML "H2", but at that point it could be attributed to stupidity and/or incompetence, since some people think for some reason that HTML is an appearance description language rather than an specific implementation instance of a content description metalanguage, which is a bit too abstract for some people. But XML is much more explicit about being a content description metalanguage, and if you've got enough of a clue about it to output your material as XML, you've got to get that much of the concept. I'd attribute this one to malice.
Capping your *download* speeds to 128kbps would get your attention, though :-) Some cable companies, who are either "ahead of the curve" or "even more terminally clueless than most about why people buy broadband", have monthly download quotas of a couple of GB and drop your bandwidth to 56kbps or so once you exceed them. That's fine for regular web viewing, probably marginal for average Napster use, and totally useless if you want to download the latest Linux release. (On the other hand, some cable companies start surcharging you by the GB if you exceed the limit, so you can end up with megabuck charges if you weren't paying attention; I'd rather get slowed down than risk that.)
I'm getting about 800-900kbps ADSL at a distance I've been told is somewhere between 12000 and 16000 feet, depending on who measured it (I think it's really about 12-14000 feet but bad wiring...). A few years ago, I had 384kbps SDSL here, but 768kbps SDSL didn't work.
This is a rapidly changing business, and the one thing that everybody's sure of is that
The main problem with 127.0.0.1 is that if you're also running a real web server on your machine, the requests for blocked sites will be sent to it, so your system will have to respond in some appropriate manner, and your browser will have to display the response appropriately. Some operating systems don't seem to have the clue that 127.0.0.2 is different from 127.0.0.1 :-) I haven't installed a web server on my main work machine since it got upgraded (?) to Win2000, but older Windows used to be a bit fuzzy about the distinction.
This isn't a popup blocker you're installing - those are for those poor suckers using IE who need all the protection they can get. This is just choosing the option that implements or doesn't implement popup windows, telling it you don't want the things. Works real fine; the last time I was on a machine that didn't have Mozilla, I was really appalled at what IE users have to put up with.
Now, it's possible that I don't have quite all of the Flash/Schlockwave plugins reinstalled correctly since the last time I installed a new rev of Phoenix :-) I'm running 0.5, and I've had some problems with some plugins not working, since their installers seem to want Real Netscape, but most are ok, and about:plugins claims that I've got Shockwave 6.something installed.
American jurisprudence also has a bunch of 1960s practices like the Exclusionary Rule and the Miranda Warning which say that courts can't admit evidence that was acquired improperly, whether it was from beating prisoners until they confess, illegally searching homes without warrants, or getting warrants by lying to judges, or lying to prisoners about the law when they don't have lawyers to advise them. Again, it didn't totally eliminate abuses, but the traditional example for its effectiveness is that the year before the Exclusionary Rule, police in New York City didn't bother getting any search warrants - they just illegally searched anybody and any place they wanted to, while the year after the rule, they almost always got warrants when they needed them (even if they still lied about their evidence on occasion.)
Be sure to call us at 1-800-turn-u-in first so we'll have it ready for you!
"It might simply be an informant you've been working with for years, goes home at night, gets on the Internet and says 'you know, I think that informant we've been covering for for years while he murdered his competition and tipped off that we might be about to indict him, and that we recently had to indict some of the FBI agents who helped cover it up, I wonder if he might be a mobster? Nah, let's try to distract people by holding a news conference about Internet ads.'", US Attorney Michael Sullivan carefully refrained from saying during the news conference.
This guy probably took the computer (or at least some of the computers from other people he'd ripped off) and sold it for half the price he "bought" it for, maybe less, so even if he hasn't poured the money up his nose, he'd only have $1200, not 2500. Maybe he's got a car, though....
As somebody else replied, this level of memory bloat is pretty recent - for most kinds of software, you shouldn't be designing it to require a latest-model desktop power box, but a two-year-old laptop, which has a lot less spare resources. That way it will run well for everybody and really rock on high-end machines. Some applications are obvious exceptions, like games, and scientific/engineering numbercrunching.
So what does it take to set one off? A good hard punch to the sensor? Thrown beer bottle? How embarassing is that!? Sure, they're designed to go off when they need to and not go off accidentally, but what about deliberate attempts to trigger them? If they ever become popular, will this become as good a way to start a bar brawl as walking up an insulting a guy's bike? Hey, drivin' an underpowered riceburner like that, I bet you really *need* to wear this airbag!
I'm now using Phoenix 0.5, which came out just recently, and it's quite toasty - I think it's ready to replace Mozilla as my main browser. The main plugins work (I'd had trouble getting them installed on 0.3 and 0.4) and it's very very fast, especially since I set the startup delay to 0 (default is 1200ms, which lets it recover from slow-loading graphics that would otherwise force redraws.) The Google-search-bar extension is really convenient, though I gather than newer Mozillas also have it. I'm normally no fan of themes (why clutter up the GUI at the cost of making it larger and slower?), but the "LittlePhoenix 1.3" theme has icons that are enough smaller that I can reclaim significant screen space, and the "Linky" extension has been a good way to handle pages with lots of links (e.g. letting you leech all the pictures into a separate window or tab, or examine a page by grabbing all the URLs on it into a tab, which can be cleaner than View Source for some ugly web pages.)
Another issue is that much of the Oregon build-out happened late in the boom - if you look at the Boston-NewYork-WashingtonDC routes, which were developed earlier, you'll find that there's less multiplexing and more sale of individual whole fibers, because the equipment costs were higher, but they did a Moore's Law type crash dive during the late 90s and early 00s that Oregon benefitted more from.
Meanwhile, if you're trying to get a data feed to every lottery terminal in Oregon, you don't care about 160-wavelength x 10Gbps OC192 fibers - you care about getting 4kbps worth of data on whatever kind of copper wire or wireless goes out to the convenience store, and doing the protocol conversion because lottery terminals still speak X.25 and nobody wants to pay the capital costs to teach the things IP :-)
Back in May 2001, there was an article in the NY Times (copy) and Slashdot (also mentioned on DeadMedia.org) about a project to run fiber optics through pneumatic tubes in New York and other big cities. While the meme is out there, it's not clear that anybody's actually implemented it. One problem, besides the financial issues, and the World Trade Center collapse in the most interesting market area, is that real ownership of the tubes is vary unclear, at least in New York City.
So yes, use some kind of IDE RAID system for your backups, and be sure to save some backup computers and operating systems every year or two to read the old disks.
The real problem isn't usually hardware rot - it's data format rot and interface standards rot. IDE is a bit tired, but it'll be supportable much longer than N-1 of the current DVD-R standards, so you probably want both. In my attic, I've got a probably-good 9-track backup tape of some of my early work, and some newer probably-good Sun cartridge tapes, and I might still have a Sun tape reader, if I can find a copy of SunOS 3.5 for my Sun-2 diskless workstation to bootstrap something to read it with.... Not a high priority - anything useful I did back in those days is probably on a Usenet archive somewhere.
We had fun last week dissing Alan Ralsky, the spammer that everybody's now signing up for spam. Previous Slashdot articles on harassing spammers point out that Overture.com accepts bids from advertisers for top positions in their search results, and the top three positions get sold on Google, MSN, etc. - So if you search for "bulk email" and click the first couple of links, each one costs the advertiser (who's presumably a spammer) whatever their bid is - typically a few bucks. For some reason, "bulk email" is having a bidding war - today's prices were over $25. The system is designed to detect multiple clickthroughs by the same person (which is why I'm not providing a direct link), but once you've got all those ads for spamware on your screen, you might as well give them the name of a promising lead - like Ralsky...
Look, I'm not *advocating* that anybody go shoot Ashcroft or Poindexter, and Ashcroft even seems to get along well with the NRA. Nor do I own guns. This is a discussion about political theory, and they're some of the most recent examples of the types of people that the authors of the Bill of Rights had in mind when they wrote the Second Amendment, just as they are recent examples of the types of people the authors had in mind when they wrote the First, Fourth, Fifth, and Sixth, and Disgraced Ex-Admiral Poindexter and his henchperson Ollie North are examples of the type of person the authors of the Uniform Code of Military Justice had in mind when they wrote the part *requiring* US military personnel to disobey illegal orders.
Yeah. Shame we couldn't mark the whole thread as "flamebait", instead of just the articles :-) Some of us believe strongly in self-protection as a basic human right, others believe strongly in the opposite, some people are seriously confused about whether guns give you more or less ability to protect yourself, and you can *forget* the Europeans (:-), who seem to believe that an armed government can run a polite society, or the Americans, who don't believe in polite societies.
Meanwhile, when liberal gun controllers start insisting that gun control laws shouldn't be enforced by armed policemen, I'll start taking them seriously. Until then, they're just saying that I should trust them with guns but they don't trust me. For liberals, that's hypocrisy, and it's amazing that they can say it with a straight face. Conservatives say that sort of thing all the time, of course, but it's *consistent* for somebody who's in power to be honest about wanting to stay in power....
If you want an uncompromising political group, there's JPFO - Jews for the Preservation of Firearms Ownership. Or there's Gun Owners of America, though they also like to push the pro-hunting agenda.
ACLU President Nadine Strossen has spoken about guns in the past, saying that she leans toward the position that the second amendment identifies gun ownership is an individual civil right, not merely a collectivist permission to be part of the National Guard or for Guardsmen to keep their rifles at home like the Swiss. But the ACLU isn't a monolithic organization - they have a lot of central resources, and get involved in Supreme Court cases, but their real work is done by local chapters, who come up with lawyers to defend people in most of the cases that they work on. So if you want the ACLU to defend gun rights, get involved, get your law degree (:-), and find cases that you can convince your local organization that it makes sense to work on.