Yes, there are other approaches to it, but it's hard to beat the vinyl iPad cover/stand/keyboard things. Of course, it iPads had the decency to provide USB ports on the side, you could also plug a better keyboard into them when you wanted, without messing with non-standard docking cable adapters.
Yes, you usually ported it on a cart, but you could easily move it from your lab to your office, or your office to the lecture hall, or to the computer room where your grad students could use it, and you didn't have to disassemble/reassemble it the way you did with a PDP-11 or mainframe.
And this meant that when I was taking a number-crunching course in grad school, and our professor didn't want us to waste valuable mainframe time doing graphics with computers when we could learn more doing them by hand, I could learn more trying lots of different time series on the 5100 and print them out on the little thermal printer, and _then_ take the ones that got the results I wanted and graph them by hand:-) If the 5100 hadn't been portable, I wouldn't have had access to it except during limited hours. (And no, I didn't walk uphill both ways in the snow to the computer center - I went to grad school in California. Uphill both ways in the snow was undergrad.)
The original article didn't say that no other form factors have been used, or that the clamshell was the first format. (And the IBM 5150 (IIRC) that I was using in 1978-1979 was also portable, though you usually did the porting on a rolling cart.)
It said that the Clamshell format won, and everything else uses that format these days. Except, of course, for the Palm Pilot, Blackberry, iPhone, and iPad, which also are portable computers that won the Format Wars. (Yes, the Newton used that format before the Palm Pilot, but didn't win, and there were lots of tablet computers before the iPad, but they all choked. And there were other screen+keyboard handhelds before and after the Crackberry, but it was one of the most successful, though the iPhone has mostly killed it.)
I'm a vegetarian for ethical reasons. Our bodies may do just fine eating animal parts, considering that we've done it since at least the time we became apes, and we've been eating cooked animal parts since we figured out fire a million years or so ago, but it's a pretty hostile way to treat animals, just because they happen to be tasty and easy to catch. I'm also fat, and could stand to lose weight, but my blood sugar is just fine, stays nicely in the middle, not too high or too low.
Most of the scientific studies of Atkins have found that it really causes weight loss by reducing your calorie intake, while keeping you from getting hungry while you do so, by tricking your metabolism into not doing all the things that tell you it's time to eat a lot. (And your chart left out the body's glycogen stores, BTW, which get used up before burning body fat.) All nice in theory.
So back during one of the previous resurgences of the Atkins fad, I looked at what I'd be able to eat if I tried it. Eggs, cheese, yogurt, no grain or fruit, no beans, lots of green veggies and rabbit food. BORING! A lot more monotonous than eating all those with meat. And I'm mildly allergic to eggs and dairy, so I'd be sneezing more.
No thanks. I'd rather eat fewer calories and substitute more garlic and peppers for the butter and salt. What I really need to do is step away from the computer and go do some cardio exercise; tai chi just doesn't count.
That's a good start, and Android's tools for doing speech-to-text translation (not only for texting, but for most applications) are also a good start, but it's not the same as having a phone UI that's voice-centric, rather than a screen-centric UI which also has voice support.
Some friends of mine were working on that back during the boom (a few grad students, and a bad entrepreneur you and I know), but it didn't really take off. It's probably a lot more practical now that we're carrying computers with another decade of Moore's law speedup. Ideally, for blind people, you'd want a system that could be entirely driven from a bluetooth headset, only getting the phone out of your pocket if you need to take a picture of something and have it read it to you or whatever.
Dextr's kind of cute, though most studies have found that for two-handed typing, there isn't actually much performance difference between layouts, once people have time to adjust to them. And while Dvorak may have been a bit better designed, "alternating hands" was pretty much the goal of QWERTY as well, because that kept keys from jamming.
FITALY is a keypad design intended for one-finger (or one-stylus) usage. The design paid attention not only to what letters are used most often, but also what letters are often used near each other, and seems to be fairly efficient. Unfortunately, due to the magic of Software Patents, you may never get to use it, but there are a few implementations out there. (It was patented in 1996, back in Palm Pilot days, but could probably be easily adapted for modern touchscreen smartphones, if the patent holder is still actively developing it.)
It's not a manual typewriter. It's an input device for an operating system that can support a wide range of keyboards. So if you know how to touch-type for an AZERTY or Swiss-German keyboard, tell the OS that that's what you're using, don't look at the keys, and just touch-type. (And remember to set it back to the original when you're done:-)
My mother's manual typewriter didn't have 1 or 0 on it, but it was a QWERTY which also had accent marks and a couple of extra characters such as ç and ñ, for writing French and Spanish. I don't remember if it had any of the Scandinavian accents and weird letters.
Actually, the odds are that 10 of them won't understand me if I talk Flemish to them, because I'm an American:-) (To me, Flemish is roughly Dutch, which is badly spelled German, and the fact that I spell German badly doesn't mean that my LOLdeutsch is comprehensible to anybody, even if it were possible for non-Dutch-speakers to get the vowels right.) And my French is worse, once you get past questions about where the train station is.
Yes, of course the article's mostly confused. But there are a few issues that aren't fictitious, to go along with the ones that are.
Inter-VM traffic is a problem, because it stays inside the server instead of getting out to where a firewall or intrusion detection system might see it, so there are cases where you might want either a virtual machine firewall or IDS, or need to move some of that traffic out of the server's virtual networks onto a physical Ethernet. I've been starting to work with Sourcefire's IDS (which is the commercialized version of Snort), though I haven't done any serious traffic measurement yet (and Sourcefire's very upfront about "Predicting performance in a VM without specific configuration detail is really hard so we're not making promises.") And maybe you want to run Checkpoint firewall software on a VM instead of a dedicated box, or maybe you want to run OpenBSD as a firewall. A couple of differences between the VM and appliance environments is that the appliance might have specialized ASICs or FPGAs to do pattern matching, as opposed to just using the CPU (and its vector processors etc.), and also that the virtual firewall or IPS is competing for CPU and memory resources with the application servers, so you need to pay attention to performance.
Also, because VMware can move processes between servers, you do need to size any external firewalls and network connections to accommodate the maximum load, not just an average load. For instance, if you've got two data centers for redundancy, and you're running active/active as opposed to active/backup, the load at one center might double if the other one fails. No big surprises there, though VM environments do give you a lot more flexibility about load balancing. You can get into situations with firewalls (or to a lesser extend, IDS), where a session gets started on one firewall, the VMware system moves the application process from one hardware server to another, which wants to use a different firewall, and so the second firewall might or might not know enough to pick up the session in the middle. Some firewalls let you configure a high availability pair that exchanges state information, some don't, and so sessions that were active when you moved the application between servers have a risk of breaking. (For IDS, that's less of a concern, though it's possible you could miss a malware event if it moved at just the wrong time.)
Obviously an event of this magnitude means you need to turn on your TV, which will be playing old reruns. Next thing you know, the electric company will be trying to muscle out the Nielsen Rating people - oh, noes!
There are different layers where you can run into problems. One of them is the ASCII value a time server hands a a time client - if it's 23h 59m 60s and the client chokes, that's a client problem. If the client tries to set an ASCII clock in the kernel to 23h 59m 60s, and the kernel chokes, that's a kernel bug. If a Unix application library can't cope with the interesting values, that's a library problem.
One obvious workaround is for the NTP server to never answer 23h 59m 60s, and for NTP clients to never tell the Unix kernel that that's what time it is. The way you implement this is simply to detect that it's about to be a leap second and Don't Respond until after it's over. (If your client can't cope with not getting a response from the server, the client's broken anyway.)
On the other hand, if the Unix kernel can't cope with a timeclock being set to 23h 59m 60s, that's a bug that should have been fixed years ago - it's not like leap seconds are a new thing.
There's a reason for leap seconds, and Astronomers will go postal on you if you try to make your clocks too dumb so they no longer track the stars. If you can't implement leap seconds out of respect, do it out of fear, okay?
I blame the people who also thought having a caps lock key where the Ctrl should be and ~` where the ESC belongs on a computer keyboard was a good idea. But typewriters had a lot of variation even back with manual typewriters. My mom's old manual typewriter had a cents key, but it was made for Romance languages, so it also had accent marks, cedilla-ç and N-tilde ñ, and maybe one or two others. (I don't remember if it had numerical 0 and 1 or if you had to use letters l and O, or whether there was a degree mark, but it didn't have the Scandinavian vowel marks.)
From a functional perspective, Bitcoin's mainly a transactional token system, not a value storage system. So while there's a bit of Ponzi / Pyramid scheme to it, that doesn't really affect most day to day use, because you're not trying to accumulate lots of Bitcoins as an investment, you're keeping a few around to buy drugs online or similar contraband, and the convenience of a hard-to-trace electronic currency is more important than losing or gaining a few percent a week on your stash. If you're the drug seller, you're going to sell your bitcoins for cash, if you're the buyer, you're going to buy a few to pay your online pharmacist.
The 7204VXR is not only safer and faster, but it functions as a space heater as well!
I forget if it could handle IPv6 very well, though - probably depends on what interface cards you're using. My Linksys won't do v6, but it sounds like Iike I'm not going to want to upgrade its firmware any time soon.
They may not be as complicated in most places as they are here in California, where we typically have half a dozen ballot questions in addition to the candidates, but for instance in the most recent election (which was a primary, not the general election), my ballot had President, Senate, Congress, State Senate, State Assembly, County Supervisor, a couple of judges' districts, two state ballot questions, and a couple of local ballot questions and bond issues. (Local governments and school boards are non-partisan, so they weren't in the primary, and while the judges are non-partisan, there were a couple of vacancies that needed to be filled.)
So yeah, the system used by most first-world parliamentary governments is much simpler and more trustable, but it's not likely to work here.
But somehow in the 2004 elections in Ohio, the "unreliable" bit was much more significant in the poor Democrat-leaning inner-city polling places, which opened late and had 1-2-hour-long lines out the door, than in the Republican-leaning suburban polling places, which had all the right parts to get their machines started and working. And Ohio's secretary of state had made a speech telling the Republicans that Ohio would be a solid win for the Republican Party.
The problem wasn't just the incompetence - it was that the "Change the Vote to Republican if Nobody's Looking" feature was designed for the US, and "Republican" means something much different in Ireland.
Technically, both of them can. "affect one polling place" works a lot better, of course, and is almost certainly what the original poster meant.
But in a sufficiently badly designed electronic voting system, one fraudster could in fact "effect one polling place", instantiating it from scratch, creating fictitious voters and audit records, and get its votes included in the district-wide totals. Back in the old paper-ballot days, you'd have to show up at the Elections Clerk's office with a stuffed ballot box and paperwork showing that it was from South Nonexistentville or from Precinct N+1, with a lock on it that would open with the Elections Clerk's key (or probably the Any Key), and it would be easier to do if several people were colluding, but an insider with credible fake paperwork might get away with it.
A friend of mine had a voting machine startup company in the early 90s. It failed, partly because of the usual reasons startups fail, but largely because they couldn't find enough of a market for highly secure tamperproof voting machines. (The Republicans fixed that problem in the early 2000s, creating a market for insecure tamper-friendly non-auditable voting machines developed by politically well-connected big companies, but a decade earlier there just wasn't a market.)
I don't know how secure his company's machine was - he was much better at sales than at technology or management, and he wasn't well-connected to the crypto community who did a lot of research into voting machine issues after the 2000 election, but it was probably at least comparable to the old lever-based machines we used back East for decades, and in spite of having some connections to the political establishments in a couple of states, he couldn't sell them.
The crypto researchers identified a bunch of requirements for secure voting, many of which are hard to do simultaneously. How do you combine auditability with voter anonymity? Can you do it in a way that ordinary non-math-geeks can understand or trust? You need a paper receipt so the voter can check it themselves later - but you need it to not show the votes so a briber or bully who told the voter how to vote can't verify it themselves. Can you audit the vote totals without having all the receipts from the voters? If there was fraud, can you fix it without violating voter anonymity?
Sorry, Mr. Republican Talking Point, but that's a bogus issue, which the Republicans have been using to exclude poor and old legitimate voters who tend to be Democrats. You've probably noticed that in most of the US, you have to sign your name in the book when you vote, and there are poll watchers from both major parties watching the voters sign in, and if more than one person shows up claiming to be a given registered voter, the second one notices it.
Yes, fraud is possible, and some dead people probably still vote in Chicago, because it's Tradition, but your party's poll watchers haven't identified more than a few cases a year in the decades they've been doing this. Furthermore, it's easy enough to audit after the election - you can take a random sample of records from people who showed up to vote, contact the registered voter and ask whether they voted, and what polling place they voted at, and if they say they didn't, check whether it's their signature.
Also, you Republicans are supposed to be the party that opposed big intrusive government, so why are you trying to make everybody carry ID?
The US voting machine contracts were given to politically connected companies after the Republicans got a lot of flack for the 2000 election vote counting failures. Accountability and Auditability were very much not requirements - they didn't want paper trails that could be audited and recounted. Speed of deployment was a requirement, and sloppiness wasn't viewed as a problem.
And while Las Vegas slot machines have a strong house advantage, the way the Republicans provided a house advantage in Ohio in 2004 was to use complicated voting machines that had to have a lot of parts to be useful, and somehow the machines in the Democrat-leaning inner city precincts didn't have enough parts to open the polls on time or keep up with the 1-2 hour-long line of voters on a rainy election day, while the Republican-leaning suburban districts had lots of working machines and no waiting. (Oh, bummer, we can't find the customized power cord! Or the cable that connects the monitor to the base, or the part of the stand that shields the voting from view when you're using it,. or we don't have a long enough extension cord to plug half the machines in to the other side of the school cafeteria because the nearby electric socket isn't rated for enough amps, or whatever.)
Also, with Las Vegas slot machines, if the casino decides that the machine malfunctioned and gave somebody a huge jackpot when it wasn't supposed to, they can declare the result to not count and not pay the sucker\\\\\\ customer.
It's not so much that they need to have English and Gaelic instead of English and Spanish - it's that the US machines' "Change to vote to Republican when nobody's looking" option means something a lot different there.
I still drove it, but for five or ten years I didn't usually drive it to an office (or if I did, it was usually to one of my customers' offices, rather than my own.) Grocery stores, restaurants, general shopping, etc. don't go away just because you're not driving to work.
Slashdot Mirror
Yes, there are other approaches to it, but it's hard to beat the vinyl iPad cover/stand/keyboard things. Of course, it iPads had the decency to provide USB ports on the side, you could also plug a better keyboard into them when you wanted, without messing with non-standard docking cable adapters.
Yes, you usually ported it on a cart, but you could easily move it from your lab to your office, or your office to the lecture hall, or to the computer room where your grad students could use it, and you didn't have to disassemble/reassemble it the way you did with a PDP-11 or mainframe.
And this meant that when I was taking a number-crunching course in grad school, and our professor didn't want us to waste valuable mainframe time doing graphics with computers when we could learn more doing them by hand, I could learn more trying lots of different time series on the 5100 and print them out on the little thermal printer, and _then_ take the ones that got the results I wanted and graph them by hand :-) If the 5100 hadn't been portable, I wouldn't have had access to it except during limited hours. (And no, I didn't walk uphill both ways in the snow to the computer center - I went to grad school in California. Uphill both ways in the snow was undergrad.)
The original article didn't say that no other form factors have been used, or that the clamshell was the first format. (And the IBM 5150 (IIRC) that I was using in 1978-1979 was also portable, though you usually did the porting on a rolling cart.)
It said that the Clamshell format won, and everything else uses that format these days. Except, of course, for the Palm Pilot, Blackberry, iPhone, and iPad, which also are portable computers that won the Format Wars. (Yes, the Newton used that format before the Palm Pilot, but didn't win, and there were lots of tablet computers before the iPad, but they all choked. And there were other screen+keyboard handhelds before and after the Crackberry, but it was one of the most successful, though the iPhone has mostly killed it.)
I'm a vegetarian for ethical reasons. Our bodies may do just fine eating animal parts, considering that we've done it since at least the time we became apes, and we've been eating cooked animal parts since we figured out fire a million years or so ago, but it's a pretty hostile way to treat animals, just because they happen to be tasty and easy to catch. I'm also fat, and could stand to lose weight, but my blood sugar is just fine, stays nicely in the middle, not too high or too low.
Most of the scientific studies of Atkins have found that it really causes weight loss by reducing your calorie intake, while keeping you from getting hungry while you do so, by tricking your metabolism into not doing all the things that tell you it's time to eat a lot. (And your chart left out the body's glycogen stores, BTW, which get used up before burning body fat.) All nice in theory.
So back during one of the previous resurgences of the Atkins fad, I looked at what I'd be able to eat if I tried it. Eggs, cheese, yogurt, no grain or fruit, no beans, lots of green veggies and rabbit food. BORING! A lot more monotonous than eating all those with meat. And I'm mildly allergic to eggs and dairy, so I'd be sneezing more.
No thanks. I'd rather eat fewer calories and substitute more garlic and peppers for the butter and salt. What I really need to do is step away from the computer and go do some cardio exercise; tai chi just doesn't count.
That's a good start, and Android's tools for doing speech-to-text translation (not only for texting, but for most applications) are also a good start, but it's not the same as having a phone UI that's voice-centric, rather than a screen-centric UI which also has voice support.
Some friends of mine were working on that back during the boom (a few grad students, and a bad entrepreneur you and I know), but it didn't really take off. It's probably a lot more practical now that we're carrying computers with another decade of Moore's law speedup. Ideally, for blind people, you'd want a system that could be entirely driven from a bluetooth headset, only getting the phone out of your pocket if you need to take a picture of something and have it read it to you or whatever.
My hovercraft is, in fact, full of eels...
Dextr's kind of cute, though most studies have found that for two-handed typing, there isn't actually much performance difference between layouts, once people have time to adjust to them. And while Dvorak may have been a bit better designed, "alternating hands" was pretty much the goal of QWERTY as well, because that kept keys from jamming.
FITALY is a keypad design intended for one-finger (or one-stylus) usage. The design paid attention not only to what letters are used most often, but also what letters are often used near each other, and seems to be fairly efficient. Unfortunately, due to the magic of Software Patents, you may never get to use it, but there are a few implementations out there. (It was patented in 1996, back in Palm Pilot days, but could probably be easily adapted for modern touchscreen smartphones, if the patent holder is still actively developing it.)
It's not a manual typewriter. It's an input device for an operating system that can support a wide range of keyboards. So if you know how to touch-type for an AZERTY or Swiss-German keyboard, tell the OS that that's what you're using, don't look at the keys, and just touch-type. (And remember to set it back to the original when you're done :-)
My mother's manual typewriter didn't have 1 or 0 on it, but it was a QWERTY which also had accent marks and a couple of extra characters such as ç and ñ, for writing French and Spanish. I don't remember if it had any of the Scandinavian accents and weird letters.
Actually, the odds are that 10 of them won't understand me if I talk Flemish to them, because I'm an American :-) (To me, Flemish is roughly Dutch, which is badly spelled German, and the fact that I spell German badly doesn't mean that my LOLdeutsch is comprehensible to anybody, even if it were possible for non-Dutch-speakers to get the vowels right.) And my French is worse, once you get past questions about where the train station is.
Yes, of course the article's mostly confused. But there are a few issues that aren't fictitious, to go along with the ones that are.
Inter-VM traffic is a problem, because it stays inside the server instead of getting out to where a firewall or intrusion detection system might see it, so there are cases where you might want either a virtual machine firewall or IDS, or need to move some of that traffic out of the server's virtual networks onto a physical Ethernet. I've been starting to work with Sourcefire's IDS (which is the commercialized version of Snort), though I haven't done any serious traffic measurement yet (and Sourcefire's very upfront about "Predicting performance in a VM without specific configuration detail is really hard so we're not making promises.") And maybe you want to run Checkpoint firewall software on a VM instead of a dedicated box, or maybe you want to run OpenBSD as a firewall. A couple of differences between the VM and appliance environments is that the appliance might have specialized ASICs or FPGAs to do pattern matching, as opposed to just using the CPU (and its vector processors etc.), and also that the virtual firewall or IPS is competing for CPU and memory resources with the application servers, so you need to pay attention to performance.
Also, because VMware can move processes between servers, you do need to size any external firewalls and network connections to accommodate the maximum load, not just an average load. For instance, if you've got two data centers for redundancy, and you're running active/active as opposed to active/backup, the load at one center might double if the other one fails. No big surprises there, though VM environments do give you a lot more flexibility about load balancing. You can get into situations with firewalls (or to a lesser extend, IDS), where a session gets started on one firewall, the VMware system moves the application process from one hardware server to another, which wants to use a different firewall, and so the second firewall might or might not know enough to pick up the session in the middle. Some firewalls let you configure a high availability pair that exchanges state information, some don't, and so sessions that were active when you moved the application between servers have a risk of breaking. (For IDS, that's less of a concern, though it's possible you could miss a malware event if it moved at just the wrong time.)
Obviously an event of this magnitude means you need to turn on your TV, which will be playing old reruns. Next thing you know, the electric company will be trying to muscle out the Nielsen Rating people - oh, noes!
There are different layers where you can run into problems. One of them is the ASCII value a time server hands a a time client - if it's 23h 59m 60s and the client chokes, that's a client problem. If the client tries to set an ASCII clock in the kernel to 23h 59m 60s, and the kernel chokes, that's a kernel bug. If a Unix application library can't cope with the interesting values, that's a library problem.
One obvious workaround is for the NTP server to never answer 23h 59m 60s, and for NTP clients to never tell the Unix kernel that that's what time it is. The way you implement this is simply to detect that it's about to be a leap second and Don't Respond until after it's over. (If your client can't cope with not getting a response from the server, the client's broken anyway.)
On the other hand, if the Unix kernel can't cope with a timeclock being set to 23h 59m 60s, that's a bug that should have been fixed years ago - it's not like leap seconds are a new thing.
There's a reason for leap seconds, and Astronomers will go postal on you if you try to make your clocks too dumb so they no longer track the stars. If you can't implement leap seconds out of respect, do it out of fear, okay?
I blame the people who also thought having a caps lock key where the Ctrl should be and ~` where the ESC belongs on a computer keyboard was a good idea. But typewriters had a lot of variation even back with manual typewriters. My mom's old manual typewriter had a cents key, but it was made for Romance languages, so it also had accent marks, cedilla-ç and N-tilde ñ, and maybe one or two others. (I don't remember if it had numerical 0 and 1 or if you had to use letters l and O, or whether there was a degree mark, but it didn't have the Scandinavian vowel marks.)
From a functional perspective, Bitcoin's mainly a transactional token system, not a value storage system. So while there's a bit of Ponzi / Pyramid scheme to it, that doesn't really affect most day to day use, because you're not trying to accumulate lots of Bitcoins as an investment, you're keeping a few around to buy drugs online or similar contraband, and the convenience of a hard-to-trace electronic currency is more important than losing or gaining a few percent a week on your stash. If you're the drug seller, you're going to sell your bitcoins for cash, if you're the buyer, you're going to buy a few to pay your online pharmacist.
The 7204VXR is not only safer and faster, but it functions as a space heater as well!
I forget if it could handle IPv6 very well, though - probably depends on what interface cards you're using. My Linksys won't do v6, but it sounds like Iike I'm not going to want to upgrade its firmware any time soon.
They may not be as complicated in most places as they are here in California, where we typically have half a dozen ballot questions in addition to the candidates, but for instance in the most recent election (which was a primary, not the general election), my ballot had President, Senate, Congress, State Senate, State Assembly, County Supervisor, a couple of judges' districts, two state ballot questions, and a couple of local ballot questions and bond issues. (Local governments and school boards are non-partisan, so they weren't in the primary, and while the judges are non-partisan, there were a couple of vacancies that needed to be filled.)
So yeah, the system used by most first-world parliamentary governments is much simpler and more trustable, but it's not likely to work here.
But somehow in the 2004 elections in Ohio, the "unreliable" bit was much more significant in the poor Democrat-leaning inner-city polling places, which opened late and had 1-2-hour-long lines out the door, than in the Republican-leaning suburban polling places, which had all the right parts to get their machines started and working. And Ohio's secretary of state had made a speech telling the Republicans that Ohio would be a solid win for the Republican Party.
The problem wasn't just the incompetence - it was that the "Change the Vote to Republican if Nobody's Looking" feature was designed for the US, and "Republican" means something much different in Ireland.
Technically, both of them can. "affect one polling place" works a lot better, of course, and is almost certainly what the original poster meant.
But in a sufficiently badly designed electronic voting system, one fraudster could in fact "effect one polling place", instantiating it from scratch, creating fictitious voters and audit records, and get its votes included in the district-wide totals. Back in the old paper-ballot days, you'd have to show up at the Elections Clerk's office with a stuffed ballot box and paperwork showing that it was from South Nonexistentville or from Precinct N+1, with a lock on it that would open with the Elections Clerk's key (or probably the Any Key), and it would be easier to do if several people were colluding, but an insider with credible fake paperwork might get away with it.
A friend of mine had a voting machine startup company in the early 90s. It failed, partly because of the usual reasons startups fail, but largely because they couldn't find enough of a market for highly secure tamperproof voting machines. (The Republicans fixed that problem in the early 2000s, creating a market for insecure tamper-friendly non-auditable voting machines developed by politically well-connected big companies, but a decade earlier there just wasn't a market.)
I don't know how secure his company's machine was - he was much better at sales than at technology or management, and he wasn't well-connected to the crypto community who did a lot of research into voting machine issues after the 2000 election, but it was probably at least comparable to the old lever-based machines we used back East for decades, and in spite of having some connections to the political establishments in a couple of states, he couldn't sell them.
The crypto researchers identified a bunch of requirements for secure voting, many of which are hard to do simultaneously. How do you combine auditability with voter anonymity? Can you do it in a way that ordinary non-math-geeks can understand or trust? You need a paper receipt so the voter can check it themselves later - but you need it to not show the votes so a briber or bully who told the voter how to vote can't verify it themselves. Can you audit the vote totals without having all the receipts from the voters? If there was fraud, can you fix it without violating voter anonymity?
Sorry, Mr. Republican Talking Point, but that's a bogus issue, which the Republicans have been using to exclude poor and old legitimate voters who tend to be Democrats. You've probably noticed that in most of the US, you have to sign your name in the book when you vote, and there are poll watchers from both major parties watching the voters sign in, and if more than one person shows up claiming to be a given registered voter, the second one notices it.
Yes, fraud is possible, and some dead people probably still vote in Chicago, because it's Tradition, but your party's poll watchers haven't identified more than a few cases a year in the decades they've been doing this. Furthermore, it's easy enough to audit after the election - you can take a random sample of records from people who showed up to vote, contact the registered voter and ask whether they voted, and what polling place they voted at, and if they say they didn't, check whether it's their signature.
Also, you Republicans are supposed to be the party that opposed big intrusive government, so why are you trying to make everybody carry ID?
The US voting machine contracts were given to politically connected companies after the Republicans got a lot of flack for the 2000 election vote counting failures. Accountability and Auditability were very much not requirements - they didn't want paper trails that could be audited and recounted. Speed of deployment was a requirement, and sloppiness wasn't viewed as a problem.
And while Las Vegas slot machines have a strong house advantage, the way the Republicans provided a house advantage in Ohio in 2004 was to use complicated voting machines that had to have a lot of parts to be useful, and somehow the machines in the Democrat-leaning inner city precincts didn't have enough parts to open the polls on time or keep up with the 1-2 hour-long line of voters on a rainy election day, while the Republican-leaning suburban districts had lots of working machines and no waiting. (Oh, bummer, we can't find the customized power cord! Or the cable that connects the monitor to the base, or the part of the stand that shields the voting from view when you're using it,. or we don't have a long enough extension cord to plug half the machines in to the other side of the school cafeteria because the nearby electric socket isn't rated for enough amps, or whatever.)
Also, with Las Vegas slot machines, if the casino decides that the machine malfunctioned and gave somebody a huge jackpot when it wasn't supposed to, they can declare the result to not count and not pay the sucker\\\\\\ customer.
It's not so much that they need to have English and Gaelic instead of English and Spanish - it's that the US machines' "Change to vote to Republican when nobody's looking" option means something a lot different there.
I still drove it, but for five or ten years I didn't usually drive it to an office (or if I did, it was usually to one of my customers' offices, rather than my own.) Grocery stores, restaurants, general shopping, etc. don't go away just because you're not driving to work.