Sure, it's nice that they've found a bug and fixed it. But I want Paypal and EBay to start using SPF or some similar mail source identifier mechanism so that my mail client can discard 99% of the mail I receive that purports to be from Paypal without making me run it through Bayesian filters or read the Subject line or whatever to kill it. (My actual Paypal spam ratio is lower than that, but only because I occasionally submit phishes I've received to their spoof addresses, and those generate robo-responsegrams that are pretty recognizable.)
Your basic Shuttle PC weighs about 4kg before you add the disk and CPU, takes standard PCI cards, has an optional handle for the top, and costs about $200-300 empty, maybe $700 heavily loaded, plus you'd need an LCD monitor of whatever size is convenient and a little keyboard. That'll weigh a lot less than these misnamed monster laptops, use standard parts so there's a viable upgrade path, and cost 1/4 as much.
Most enterprise-class servers that have N network cards in them don't have N different _types_ of cards - they've got N cards of the same type, or maybe N-1 of the same and one of a different type (e.g. a GigE and N-1 100Mbps cards.) So you only need one or two different drivers.
Laptops these days usually have two types of network interfaces - one wired and one wireless. Occasionally you'll have different types of wireless cards to plug in, e.g. an 802.11a vs..11g or something.
So you've learned what RTFM means by now?:-) Ok, it's been a while since I've read up on kernel structure either.... but you _should_ do so. Linux is rather famouslynot a microkernel architecture that lets you partition off little pieces into user space - it's a big honkin' kernel plus loadable modules that let you add even more things. There are hardware-dependent and hardware-independent parts of the kernel. Device drivers inherently hardware-dependent, and sharing address space with the kernel makes it easier to do things like DMA without having to do a lot of data copying.
As far as network drivers in particular go, the layers that use them, such as IP, live in the kernel, so it's rather annoying for them to talk to drivers that are up in user space. Specific network cards, especially wireless, might have bits that live up in user space, such as user interfaces for loading in crypto keys, but the bulk data transfer applications normally belong down in or near the kernel.
Why are there a whole pile of network card drivers in the kernel when you'd normally use only one or two? Same reason there are a whole pile of drivers for other devices in the kernel, when you've normally only got one graphics card and one sound card. If you're shipping a pre-compiled kernel, you want it to support as many different users as possible, and all it costs you is some RAM to store the code you're not using, or if you can handle them as loadable modules, it only costs you the work to keep track of those. But if you want to compile your own kernel for specific machines, and leave out the drivers you don't want, and while you're at it compile all your applications programs with the level of optimization your hardware supports, get a copy of Gentoo Linux and have fun learning lots more detail about Linux internals.
Thanks - looks like I'm running DirectX 9.0{b,c} on my main machines. One of them doesn't like DirectX Input testing:-) But at least I know how to run the tests now.
Ok, maybe this is a dumb question, but I'm not a gamer so I've never worried about graphics beyond how many pixels I can get and what color depth. My home PC runs integrated motherboard graphics and sound, and my work laptop does the same. How do I tell what version of DXGI or Direct3d I'm running (if at all)? It wasn't obvious from the Control Panel. I don't have digital video output, (nor DRM, thank you), but integrated video doesn't need it for another couple of years, and I'm not particularly convinced that a 1 GHz machine needs that much video horsepower.
Ars Technica suggests that they really ought to be requiring 2.5 GHz or more, and I'm surprised they've been sucked in to the "Assume all CPUs resemble Intel main-Pentium-track CPU" - my laptop 1.6GHz Pentium M seems to be faster than my desktop 2.4GHz Celeron.
They also say that the basic Home systems won't require Premium-rated hardware, but everything else will. I'm sorry, but while my business system probably does need lots of RAM, USB2, FastEthernet, and good power management, I don't see it needing digital video output to run business apps - it needs lots of pixels on the screen to handle lots of text, and it's nice to be able to play DVDs on the airplane, but the video output doesn't need to be anything like the speed I'd use for a media player box because I don't *want* a chorus of paperclips singing and dancing at me in 6.1 audio.
When I was in college, we had a computer that had a water-cooling leak that sprayed directly at the motherboard and fried it. Unfortunately, since this was ~1975, the computer was the IBM 370 mainframe that provided most of the computing for the campus, and it was down for a few weeks.
In a smaller, more modern environment, the effect on other computers is going to depend on whether the cases let water leak from one to the next - e.g. in a stack of 1U, are there vents on the top and bottom or only sides - and on whether there's enough liquid leaking to get to the other machine before it evaporates.
I _am_ quite familiar with the T and E standards, having worked in the telecom industry for a few decades... But there are in fact countries outside the US that aren't part of Europe - Most of the world uses E1 and E3, Japan uses T1 and has a 6 Mbps J2 standard, Taiwan uses T1s, and also you can get T3s in much of Europe including parts of the UK near London. Canada uses T1, Mexico mostly uses E1.
The T1/T2/T3/T4 bundling was fairly similar to the analog transmission hierarchy that we used before getting digital transmission, and maybe some of the digital radio technology found it useful before we got fiber, but it turns out that T2 doesn't really do enough for you to make it worth building separate mux equipment that supports it. The T3 framing/signalling channels have an option called M13 that does support the bits you need for T2, but the US mostly uses the newer C-Bit options, which don't bother with it (and use those bits to give you more signalling information instead.) Unfortunately, I did recently run into some locations in Canada that still only support M13 for their T3s - the Cisco ATM switch equipment we use only supports C-Bit and not M13. By the way, if you look at Cisco routers and switches, you'll find support for T1, T3, E1, and E3, but I haven't seen anything from them that supports T2 or E2.
I've seen the term T4 in database work I did back in the early 90s, but I don't think it was actually referring to T4 - it tended to get used to refer 36xT3 format on the FT series G fibers that we used before SONET took over. (SONET's the US variand on the SDH standards used in most of the world.)
Nobody actually uses T2 - and nobody in the US implements it. There are various NxT1 solutions that carriers use to bond T1s together, often getting 6+ Mbps, but it's not T2.
Fractional T3 (except for NxT1) isn't a realistic concept either, because there aren't standards for it that are comparable to Fractional T1. Again, some carriers do things with muxes to derive NxT1 from a T3, but that's not really fractional T3, or they do Frame Relay or ATM to get intermediate speeds.
A very common pricing structure _is_ usage-dependent, defining usage by speed, though not by bytes transferred. Carriers measure bandwidth usage at some interval, typically 5 minutes, and charge based on the 95th percentile of bandwidth usage. It's fairly common for speeds above T1 - either you pay for straight 95th percentile usage, or you commit to N Mbits at some price, and pay extra if your 95th percentile is above that.
But yes, measured pricing resembling Blackberry's is totally unrealistic for data users - the amounts of data we need are different, and the quantities of data we need don't fit well into any pricing structure that also includes Blackberry-mail, because either one end is too high to be economically usable, or the other end is too low to make money so why bother.
> Why does every tech article, without fail, have more political jibes in it than tech comments?
Because almost nobody's bothering to *think* about the technology - MoveOn has a political agenda and no technical clue, and the telco guys have a knee-jerk reaction about always arguing regulation and money when anybody challenges them, rather than explaining the technical points they're making to politicians and reporters who don't have a clue about technology, and Dave Isenberg, who should know better, is arguing politics with the telcos instead of tech. Some people have portrayed this as a Netheads vs. Bellheads technology fight, but it's not - both sides are acting like boneheads, and everybody's pontificating about what the other boneheads said rather than looking at the real problem.
So here are a few technical and/or economic facts:
Recent DSL flavors can carry up to about 25 Mbps downstream, typically from your home to a green box or maybe a telco POP. Some of the press I've read indicates that Verizon's providing about that much bandwidth to customer homes on their fiber builds as well, though they've got more choices. Just because old people in Korea have 100 Mbps broadband doesn't mean that it's dedicated bandwidth - it's typically 100 Mbps to their apartment building basement and an upstream shared with all their neighbors.
HDTV needs about 9 Mbps, and digital standard-resolution TV needs about 2 Mbps, and a reasonable bundle of TV bandwidth for a home is about 15 Mbps, which leaves 5-10 Mbps of downstream for other applications.
Typical home broadband services today are 1-6 Mbps downstream.
A typical telco POP has 10K-100K users. If everybody watched TV at once during prime time, using regular unicast Internet bandwidth, that'd be about 150-1500 Gbps per telco office.That's not feasible, simply ain't going to happen, give up now and find some technical alternative.
Multicast is the obvious solution - either multicast all the way down to the user, or at least multicast down to the telco office and convert to unicast on the DSL; it doesn't make much difference except for how you program your home router. But there are still scaling issues.
A telco office might be able to afford a GigE or OC48 data feed, so 1-2.4Gbps. If that's all dedicated to video (as opposed to shared with the rest of the user traffic), that's ~111-250 HDTV channels, or 500-1200 SDTV channels, or some mixture of the two, like 50 channels of HDTV and 270 SDTV. Either that's First-Come-First-Served choice of channels, or somebody decides what channels to carry at each telco POP.
Telcos want to make money somehow, prices for wired data services keep going down, voice minutes are nearly free, and the cable TV companies are beating them up with voice-over-cable services. They think they can get money by selling TV channel access to the consumers, and the TV content industry folks are better set up for dealing with a small number of TV delivery players (cable, satellite, now add telcos) than they are for dealing direct-to-consumer.
The TV industry has ongoing big fights between the content providers and the distributors like the analog-radio-wave broadcasters, the satellite broadcasters, and the cable providers about who pays whom for what, eyeballs vs. content. Don't expect any less fighting if you add telcos as a distributor or digitize the content.
Many telco decisionmakers don't understand Internet data users, and think that the market for 25 Mbps data-only service is a very small number of geeks, and that the market's willingness to pay for extra bandwidth above ~1.5-3Mbps is pretty slim (except for television which is a special case), so they think they can satisfy their market by selling 6 Mbps "premium service", maybe up to 10 Mbps if you're at the luckier end of the bandwidth/distance curve.
Wholesale Internet Transit prices vary radically, but they're around $25/Mbp
So you need photographic copy - it's easy enough to generate. You may want to use Photoshop or GIMP to paste in the real screenshot instead of the harder-to-read photograph. If you don't already have a cat, you can obtain one at your local SPCA or Humane Society.
Patents are intended to encourage private industry and individuals to develop things that they can make money on, by preventing other people from using the technology unless they pay the inventor money. Livermore National Labs is a government-funded lab, and technology that the public pays for should not only be usable by the public, but we shouldn't have to pay them more money to use it because we've *already* paid them.
Pulver runs the Voice On The Net (VON), which is a major focus for the VOIP industry. It's not the type of show that's heavily consumer-oriented - it's mostly developers, infrastructure vendors, equipment vendors, VOIP phone vendors who are trying to sell their products to carriers, etc. Lots of Open Source - the Digium folks do a booth, and a number of Asterisk-using groups are there.
Free World Dialup didn't explode the way Skype did, but it had some good ideas, and not everybody's startup succeeds.
There are really two IM standards that matter - the better developed one is Jabber, and the emerging one is the SIP standard used for VOIP, which is a proxyable presence server that can support various media including text, VOIP, and video. Other than that, most of the choices tend to be proprietary, so you use clients like GAIM that tap into them. A number of the bigger IM services are moving from proprietary-only protocols to one of those two standards, and if you're doing VOIP anyway, it may make sense to have your VOIP presence server / directory system also support IM.
My company was recently acquired, and our new Texan overlords use an annoying homebrew IM system:-)
But before that, we used a Jabber-based system that included one server inside the firewall and one server outside the firewall (probably in a DMZ), which meant that I could use the corporate IM system from work and also from home, on or off a VPN connection, which was amazingly convenient. It also meant that I could start up the IM client when I booted my laptop, and if I was at home, that meant that it could log in without waiting for me to start my VPN client. And it had all the Jabbery goodness I'd expect, though I didn't try using a bunch of different clients just for fun.
I'm 50. It'd be really nice if they've got that Alzheimer's vaccine down solid in 10 years and seriously improved by 20 years, because I really don't want to get it. It'll also be seriously good for US society if most of the baby boomers who would have been getting it avoid it, so you younger folks don't have to spend as much taking care of us, or at least can deal with mentally competent frail old people. While we're at it, I hope the get the cancer stuff nailed down.
I've been doing various IT-like things my whole career, whether it's programming, consulting, or whatever. It's been a lot of fun, and I'm not particularly a biotech type, but I hope the tools we've built over the last few decades help the biotech folks do a much better job.
LISP, BASIC, FORTH, P-Code, Java+Netscape
on
The End of Native Code?
·
· Score: 4, Interesting
LISP was a simple, elegant language that demonstrated that almost any language written after 1961 was unnecessary, except for demonstrations of concepts like Object-Oriented programming that could then be re-implemented into LISP, and that any code written in older languages could be replaced with something better:-)
BASIC had its problems, warping a generation of programmers (including me), but it was small and light and didn't take long to learn unless you wanted to enough find tricks to get real work done.
FORTH was smaller, lighter, and faster. It was overly self-important, considering its reinvention of the subroutine to something new and radical, but if you wanted to program toasters or telescopes it was the language to use. Postscript was somewhat of a Forth derivative.
P-Code was a nice portable little VM you could implement other things on.
And then there was Java, which grew out of Gosling's experiences with NeWS, a Postscript-based windowing system. If you wonder why you're not using Netscape and maybe not using Java, and why you've probably got Windows underneath your Mozilla, it's because it became obvious to lots of people that Netscape+Java was a sufficiently powerful and easily ported environment that the operating system underneath could become nearly irrelevant - so Microsoft had to go build a non-standards-compliant browser and wonky Java implementation and start working on.NET to kill off the threat. It wasn't that conquering the market for free browsers was a big moneymaker - it was self-defense to make sure that free browsers didn't conquer the OS market, allowing Windows+Intel to be replaced by Linux/BSD/QNX/MacOS/OS9/SunOS/etc.
A stoned little dwarf throws a nasty-looking knife at you and misses. What more do you need?
Actually I'm exaggerating a bit - plug a Vax of just about any flavor into the back end of the VT100 and you can play Nethack, or get a PC with Win95 or better and run the color version.
1GB of RAM will cost you about $100 US. For that price you can get about 4GB of flash, either as USB2 or as CF.
The RAM is obviously faster, though it probably burns more power and generates more heat. On the other hand, you can install a lot of your operating system in a flash that big and avoid rotational and seek latency, which can make your machine a lot faster. And if you can get filesystem journalling to use it, that's a big big speed win.
If you've got a laptop, it's nice to use it to save to when you suspend/hibernate/etc.
256MB USB memory sticks are really cheap, and most modern PCs seem happy to boot from them, though of course you might want to spend $50 for 2 GB or $100-150 for 4GB. A Linux distro hefty enough to play Nethack should use up maybe 50MB:-)
The screen will still be burning battery, of course, but turning off the disks should help, and your data will load a lot faster because you won't have the latency of waiting for the disk to rotate or the head to seek (even if your memory stick has a lower transfer rate than the disk drives, you still should win.)
Seagate sells hard drives, and this gives them a cooler hard drive to sell. And there's enough intelligence in a typical disk controller to do most of the work here.
Intel sells chipsets, and their approach gives them a cooler motherboard chipset to sell, and hteir chipsets are flexible enough to add this kind of support.
Microsoft sells features that lock you in to Microsoft software, so if it's easy for them to support proprietary hard drives or motherboard chipsets or DRM chips that lock you in to buying and upgrading your Windows OS, that's what they'll do.
If you wanted to build a hybrid laptop for yourself, you could buy a $100 4GB USB flash memory stick, or a Compact-Flash-to-PCMCIA adapter (for mechanical safety) and some flash, install the operating system on the flash instead of the rotating disk, and use the rotating disk for backups and bulk storage. It'd be pretty easy with Linux (people fit mini-distros onto 128MB sticks, so why not on a bigger flash!), and you could probably get Windows to work instead. Seagate or Intel can do a more sophisticated job by working closely with MS to decide what types of data need to go in which caches - most Operating Systems today really only support two speeds of storage (RAM and disk-like) rather than knowing about three (RAM, Flash, Disk), (plus a few subtleties like swap space and networked filesystems), and really dealing with flash as a speed intermediate between RAM and disk will take a few years of research and some major OS reqork before it's done really really well, but even a naive version should be a big win.
Article on new Seagate products was mentioned in today's Slashdot article on hybrid flash+disk drives, but farther down the article it talks about some laptop drives with built-in encryption. While I think that the OS really ought to be doing it, farming the job out to the disk is certainly a good start, and it's probably easier to use if less flexible.
While I largely agree with Drinkypoo's view that working for the military and working for the country are much different, he's flaming the wrong flamewar for this forum. I suspect most people who join the military are less concerned about whether the DoD bureaucrats will mishandle their personnel data than about
- whether they'll get shot at, or about
- what country is going to do something stupid or dangerous enough to need attacking or defending, or
- how often they'll need to be away from their families, or
- whether their weapons will be reliable in whatever country they get sent to next (because weapons that work well in Northern Europe may suck in Vietnamese jungles or Iraqi deserts, and weapons that work fantastically well on Powerpoint presentations inside the Beltway somehow aren't the same when you take them out in the field), or
- whether they'll get the supplies they need to do the job they've got to do, or
- how to get their platoon to learn to stay alive while getting the job done, or
- whether they ought to frag their bonehead lieutenant before he gets them killed, or
- how to tell the REMFs back at the Pentagon that they don't have a clue what's happening down on the ground, or
whether the recruiters who said they'd learn valuable new skills thought they'd get jobs improvising truck armor when they got back home to LA.
That doesn't mean that they don't care about their personnel data, especially if it affects their paychecks or promotions or pensions, but when they join the Army they pretty much understand they're joining the Army.
And it's not like working for a bank or the Phone Company or a factory or a university instead gives you a lot of reassurance that your data won't get mishandled (or if it does, think again.)
First of all, you should be encrypting the disks (or file systems), not just the files that you expect to be sensitive. When CPUs are running at N GHz these days, there's really no excuse not to, and laptops aren't the only computers that get stolen or cracked or spywared. Start secure by default, and only make things less secure if you need to, not the other way around.
But far more important is thinking about what data needs to be used together and what data can be kept separate - that Customer SSN data of yours should be in a separate database, only used to generate tax paperwork, and not accessible to other applications (unless of course you're investigating fraud, which wouldn't be a totally surprising problem for a casino to encounter.) Not only should you not be taking it home, you shouldn't be keeping it near the less sensitive marketing stuff.
Laptops get lost and stolen, desktops get cracked, spyware gets run, viruses get spread, files get shared, big databases become pocketsized - get used to it. Encrypting file systems on laptops ought to be a no-brainer decision these days, for anybody who's dealing with any sensitive data, whether it's personnel or financial or whatever.
That doesn't mean that Microsoft makes it easy (the EFS stuff in NTFS looks pretty hokey and hard to use, and apparently isn't in XP Home, only XP Pro, and it's not clear from a few minutes' reading of the documentation whether you can tell it to encrypt your My Documents folder withough causing major chaos. (And yes, I realize that that's only part of what needs to be encrypted, and I don't trust MS's current crypto given how badly broken all their earlier crypto was, but at least it's a _start_.)) Linux/BSD? Multiple solutions are available and relatively easy to implement - obviously any secure data needs to be on a computer with a real operating system...
And the best security we've got in practice is that thieves or fences usually wipe the info on stolen machines to avoid getting caught, instead of realizing that it's usually worth much more than the stolen hardware. Encrypting or Multi-Level-Secure databases have been around for a while, but are still mostly researchy.
But Crypto's only a bandaid, and I say this as somebody who's been a crypto geek for a couple of decades. People who handle information need to think about what's sensitive and what's not, and design their databases so that nobody needs to touch sensitive data unless they actually need to touch the sensitive data. So Social Security Numbers (or your local government's equivalents) shouldn't be used as database keys, and Last-4-digits shouldn't be used as passwords, and Employee ID Numbers or Customer ID Numbers should be something entirely unrelated to SSN. That means you need a separate table connecting ID# and SSN that the Payroll department tax bureaucrats can use when they're reporting taxes, but which isn't accessible to anybody who's not handling taxes. And Medical Insurance account numbers shouldn't be your SSN, in spite of how convenient it is to all the large bureaucracies out there to start all conversations by asking for your Social. If HR needs to collect new hires' Citizenship ID#s when verifying that they are legally permitted to work in the country, or the Driver's License Bureaucrats need to collect it to verify that people who drive aren't "Deadbeat Dads" and don't speak Spanish, then that data needs to be kept separate from the less-sensitive data.
The Bush Administration and its predecessors in the military and civil service have put a lot of emphasis on "Know Your Customer" laws and requiring airlines and banks and employers and such to collect lots of private data and report it to them, maximizing the ability of everybody with a cheap Moore's-Law-Inside PC to do massive data mining, and it's going to be hard to undo all that infrastructure once we through them out of office - it's important to make sure that you can protect your own employees and customers and suppliers from accidental data loss, and deliberate theft, and planned or unplanned data mining.
Slashdot Mirror
Sure, it's nice that they've found a bug and fixed it. But I want Paypal and EBay to start using SPF or some similar mail source identifier mechanism so that my mail client can discard 99% of the mail I receive that purports to be from Paypal without making me run it through Bayesian filters or read the Subject line or whatever to kill it. (My actual Paypal spam ratio is lower than that, but only because I occasionally submit phishes I've received to their spoof addresses, and those generate robo-responsegrams that are pretty recognizable.)
Your basic Shuttle PC weighs about 4kg before you add the disk and CPU, takes standard PCI cards, has an optional handle for the top, and costs about $200-300 empty, maybe $700 heavily loaded, plus you'd need an LCD monitor of whatever size is convenient and a little keyboard. That'll weigh a lot less than these misnamed monster laptops, use standard parts so there's a viable upgrade path, and cost 1/4 as much.
Laptops these days usually have two types of network interfaces - one wired and one wireless. Occasionally you'll have different types of wireless cards to plug in, e.g. an 802.11a vs. .11g or something.
As far as network drivers in particular go, the layers that use them, such as IP, live in the kernel, so it's rather annoying for them to talk to drivers that are up in user space. Specific network cards, especially wireless, might have bits that live up in user space, such as user interfaces for loading in crypto keys, but the bulk data transfer applications normally belong down in or near the kernel.
Why are there a whole pile of network card drivers in the kernel when you'd normally use only one or two? Same reason there are a whole pile of drivers for other devices in the kernel, when you've normally only got one graphics card and one sound card. If you're shipping a pre-compiled kernel, you want it to support as many different users as possible, and all it costs you is some RAM to store the code you're not using, or if you can handle them as loadable modules, it only costs you the work to keep track of those. But if you want to compile your own kernel for specific machines, and leave out the drivers you don't want, and while you're at it compile all your applications programs with the level of optimization your hardware supports, get a copy of Gentoo Linux and have fun learning lots more detail about Linux internals.
Thanks - looks like I'm running DirectX 9.0{b,c} on my main machines. One of them doesn't like DirectX Input testing :-) But at least I know how to run the tests now.
Ars Technica suggests that they really ought to be requiring 2.5 GHz or more, and I'm surprised they've been sucked in to the "Assume all CPUs resemble Intel main-Pentium-track CPU" - my laptop 1.6GHz Pentium M seems to be faster than my desktop 2.4GHz Celeron.
They also say that the basic Home systems won't require Premium-rated hardware, but everything else will. I'm sorry, but while my business system probably does need lots of RAM, USB2, FastEthernet, and good power management, I don't see it needing digital video output to run business apps - it needs lots of pixels on the screen to handle lots of text, and it's nice to be able to play DVDs on the airplane, but the video output doesn't need to be anything like the speed I'd use for a media player box because I don't *want* a chorus of paperclips singing and dancing at me in 6.1 audio.
In a smaller, more modern environment, the effect on other computers is going to depend on whether the cases let water leak from one to the next - e.g. in a stack of 1U, are there vents on the top and bottom or only sides - and on whether there's enough liquid leaking to get to the other machine before it evaporates.
The T1/T2/T3/T4 bundling was fairly similar to the analog transmission hierarchy that we used before getting digital transmission, and maybe some of the digital radio technology found it useful before we got fiber, but it turns out that T2 doesn't really do enough for you to make it worth building separate mux equipment that supports it. The T3 framing/signalling channels have an option called M13 that does support the bits you need for T2, but the US mostly uses the newer C-Bit options, which don't bother with it (and use those bits to give you more signalling information instead.) Unfortunately, I did recently run into some locations in Canada that still only support M13 for their T3s - the Cisco ATM switch equipment we use only supports C-Bit and not M13. By the way, if you look at Cisco routers and switches, you'll find support for T1, T3, E1, and E3, but I haven't seen anything from them that supports T2 or E2.
I've seen the term T4 in database work I did back in the early 90s, but I don't think it was actually referring to T4 - it tended to get used to refer 36xT3 format on the FT series G fibers that we used before SONET took over. (SONET's the US variand on the SDH standards used in most of the world.)
Fractional T3 (except for NxT1) isn't a realistic concept either, because there aren't standards for it that are comparable to Fractional T1. Again, some carriers do things with muxes to derive NxT1 from a T3, but that's not really fractional T3, or they do Frame Relay or ATM to get intermediate speeds.
A very common pricing structure _is_ usage-dependent, defining usage by speed, though not by bytes transferred. Carriers measure bandwidth usage at some interval, typically 5 minutes, and charge based on the 95th percentile of bandwidth usage. It's fairly common for speeds above T1 - either you pay for straight 95th percentile usage, or you commit to N Mbits at some price, and pay extra if your 95th percentile is above that.
But yes, measured pricing resembling Blackberry's is totally unrealistic for data users - the amounts of data we need are different, and the quantities of data we need don't fit well into any pricing structure that also includes Blackberry-mail, because either one end is too high to be economically usable, or the other end is too low to make money so why bother.
Because almost nobody's bothering to *think* about the technology - MoveOn has a political agenda and no technical clue, and the telco guys have a knee-jerk reaction about always arguing regulation and money when anybody challenges them, rather than explaining the technical points they're making to politicians and reporters who don't have a clue about technology, and Dave Isenberg, who should know better, is arguing politics with the telcos instead of tech. Some people have portrayed this as a Netheads vs. Bellheads technology fight, but it's not - both sides are acting like boneheads, and everybody's pontificating about what the other boneheads said rather than looking at the real problem.
So here are a few technical and/or economic facts:
So you need photographic copy - it's easy enough to generate. You may want to use Photoshop or GIMP to paste in the real screenshot instead of the harder-to-read photograph. If you don't already have a cat, you can obtain one at your local SPCA or Humane Society.
Patents are intended to encourage private industry and individuals to develop things that they can make money on, by preventing other people from using the technology unless they pay the inventor money. Livermore National Labs is a government-funded lab, and technology that the public pays for should not only be usable by the public, but we shouldn't have to pay them more money to use it because we've *already* paid them.
Free World Dialup didn't explode the way Skype did, but it had some good ideas, and not everybody's startup succeeds.
There are really two IM standards that matter - the better developed one is Jabber, and the emerging one is the SIP standard used for VOIP, which is a proxyable presence server that can support various media including text, VOIP, and video. Other than that, most of the choices tend to be proprietary, so you use clients like GAIM that tap into them. A number of the bigger IM services are moving from proprietary-only protocols to one of those two standards, and if you're doing VOIP anyway, it may make sense to have your VOIP presence server / directory system also support IM.
But before that, we used a Jabber-based system that included one server inside the firewall and one server outside the firewall (probably in a DMZ), which meant that I could use the corporate IM system from work and also from home, on or off a VPN connection, which was amazingly convenient. It also meant that I could start up the IM client when I booted my laptop, and if I was at home, that meant that it could log in without waiting for me to start my VPN client. And it had all the Jabbery goodness I'd expect, though I didn't try using a bunch of different clients just for fun.
I've been doing various IT-like things my whole career, whether it's programming, consulting, or whatever. It's been a lot of fun, and I'm not particularly a biotech type, but I hope the tools we've built over the last few decades help the biotech folks do a much better job.
BASIC had its problems, warping a generation of programmers (including me), but it was small and light and didn't take long to learn unless you wanted to enough find tricks to get real work done.
FORTH was smaller, lighter, and faster. It was overly self-important, considering its reinvention of the subroutine to something new and radical, but if you wanted to program toasters or telescopes it was the language to use. Postscript was somewhat of a Forth derivative.
P-Code was a nice portable little VM you could implement other things on.
And then there was Java, which grew out of Gosling's experiences with NeWS, a Postscript-based windowing system. If you wonder why you're not using Netscape and maybe not using Java, and why you've probably got Windows underneath your Mozilla, it's because it became obvious to lots of people that Netscape+Java was a sufficiently powerful and easily ported environment that the operating system underneath could become nearly irrelevant - so Microsoft had to go build a non-standards-compliant browser and wonky Java implementation and start working on .NET to kill off the threat. It wasn't that conquering the market for free browsers was a big moneymaker - it was self-defense to make sure that free browsers didn't conquer the OS market, allowing Windows+Intel to be replaced by Linux/BSD/QNX/MacOS/OS9/SunOS/etc.
Actually I'm exaggerating a bit - plug a Vax of just about any flavor into the back end of the VT100 and you can play Nethack, or get a PC with Win95 or better and run the color version.
The RAM is obviously faster, though it probably burns more power and generates more heat. On the other hand, you can install a lot of your operating system in a flash that big and avoid rotational and seek latency, which can make your machine a lot faster. And if you can get filesystem journalling to use it, that's a big big speed win.
If you've got a laptop, it's nice to use it to save to when you suspend/hibernate/etc.
The screen will still be burning battery, of course, but turning off the disks should help, and your data will load a lot faster because you won't have the latency of waiting for the disk to rotate or the head to seek (even if your memory stick has a lower transfer rate than the disk drives, you still should win.)
Intel sells chipsets, and their approach gives them a cooler motherboard chipset to sell, and hteir chipsets are flexible enough to add this kind of support.
Microsoft sells features that lock you in to Microsoft software, so if it's easy for them to support proprietary hard drives or motherboard chipsets or DRM chips that lock you in to buying and upgrading your Windows OS, that's what they'll do.
If you wanted to build a hybrid laptop for yourself, you could buy a $100 4GB USB flash memory stick, or a Compact-Flash-to-PCMCIA adapter (for mechanical safety) and some flash, install the operating system on the flash instead of the rotating disk, and use the rotating disk for backups and bulk storage. It'd be pretty easy with Linux (people fit mini-distros onto 128MB sticks, so why not on a bigger flash!), and you could probably get Windows to work instead. Seagate or Intel can do a more sophisticated job by working closely with MS to decide what types of data need to go in which caches - most Operating Systems today really only support two speeds of storage (RAM and disk-like) rather than knowing about three (RAM, Flash, Disk), (plus a few subtleties like swap space and networked filesystems), and really dealing with flash as a speed intermediate between RAM and disk will take a few years of research and some major OS reqork before it's done really really well, but even a naive version should be a big win.
Article on new Seagate products was mentioned in today's Slashdot article on hybrid flash+disk drives, but farther down the article it talks about some laptop drives with built-in encryption. While I think that the OS really ought to be doing it, farming the job out to the disk is certainly a good start, and it's probably easier to use if less flexible.
- - whether they'll get shot at, or about
- - what country is going to do something stupid or dangerous enough to need attacking or defending, or
- - how often they'll need to be away from their families, or
- - whether their weapons will be reliable in whatever country they get sent to next (because weapons that work well in Northern Europe may suck in Vietnamese jungles or Iraqi deserts, and weapons that work fantastically well on Powerpoint presentations inside the Beltway somehow aren't the same when you take them out in the field), or
- - whether they'll get the supplies they need to do the job they've got to do, or
- - how to get their platoon to learn to stay alive while getting the job done, or
- - whether they ought to frag their bonehead lieutenant before he gets them killed, or
- - how to tell the REMFs back at the Pentagon that they don't have a clue what's happening down on the ground, or
- whether the recruiters who said they'd learn valuable new skills thought they'd get jobs improvising truck armor when they got back home to LA.
That doesn't mean that they don't care about their personnel data, especially if it affects their paychecks or promotions or pensions, but when they join the Army they pretty much understand they're joining the Army.And it's not like working for a bank or the Phone Company or a factory or a university instead gives you a lot of reassurance that your data won't get mishandled (or if it does, think again.)
But far more important is thinking about what data needs to be used together and what data can be kept separate - that Customer SSN data of yours should be in a separate database, only used to generate tax paperwork, and not accessible to other applications (unless of course you're investigating fraud, which wouldn't be a totally surprising problem for a casino to encounter.) Not only should you not be taking it home, you shouldn't be keeping it near the less sensitive marketing stuff.
That doesn't mean that Microsoft makes it easy (the EFS stuff in NTFS looks pretty hokey and hard to use, and apparently isn't in XP Home, only XP Pro, and it's not clear from a few minutes' reading of the documentation whether you can tell it to encrypt your My Documents folder withough causing major chaos. (And yes, I realize that that's only part of what needs to be encrypted, and I don't trust MS's current crypto given how badly broken all their earlier crypto was, but at least it's a _start_.)) Linux/BSD? Multiple solutions are available and relatively easy to implement - obviously any secure data needs to be on a computer with a real operating system...
And the best security we've got in practice is that thieves or fences usually wipe the info on stolen machines to avoid getting caught, instead of realizing that it's usually worth much more than the stolen hardware. Encrypting or Multi-Level-Secure databases have been around for a while, but are still mostly researchy.
But Crypto's only a bandaid, and I say this as somebody who's been a crypto geek for a couple of decades. People who handle information need to think about what's sensitive and what's not, and design their databases so that nobody needs to touch sensitive data unless they actually need to touch the sensitive data. So Social Security Numbers (or your local government's equivalents) shouldn't be used as database keys, and Last-4-digits shouldn't be used as passwords, and Employee ID Numbers or Customer ID Numbers should be something entirely unrelated to SSN. That means you need a separate table connecting ID# and SSN that the Payroll department tax bureaucrats can use when they're reporting taxes, but which isn't accessible to anybody who's not handling taxes. And Medical Insurance account numbers shouldn't be your SSN, in spite of how convenient it is to all the large bureaucracies out there to start all conversations by asking for your Social. If HR needs to collect new hires' Citizenship ID#s when verifying that they are legally permitted to work in the country, or the Driver's License Bureaucrats need to collect it to verify that people who drive aren't "Deadbeat Dads" and don't speak Spanish, then that data needs to be kept separate from the less-sensitive data.
The Bush Administration and its predecessors in the military and civil service have put a lot of emphasis on "Know Your Customer" laws and requiring airlines and banks and employers and such to collect lots of private data and report it to them, maximizing the ability of everybody with a cheap Moore's-Law-Inside PC to do massive data mining, and it's going to be hard to undo all that infrastructure once we through them out of office - it's important to make sure that you can protect your own employees and customers and suppliers from accidental data loss, and deliberate theft, and planned or unplanned data mining.