Follow the money. The idea of research chumps had to have crossed the minds of the kernel developers. Good thought!

Nothing short of full disclosure has any chance of getting the chumps past this mess.

Then the input problem remains.

Also, if we are going to manually issue tickets and such, we might as well put the record of the vote there, call it a ballot, and get the election done in a trustworthy way.

First, I am just going to say without making our votes personally identifiable, it's not possible to vote electronically in a way the voter can trust.

Why is that?

Here is an analogy:

There is a room, with a person in it, and that room is secure. Nobody can enter or leave, except for that person, and nobody has access to the resources inside that room, except the person.

You, the voter approach and communicate your vote. The person in the room makes a record, and then communicates your vote back to you. They may even produce a piece of paper for you to remember your vote by.

Once everyone has voted, this person leaves the room and communicates the total votes. There is no recording of the voter interaction, just that person interpreting the votes they hear, keeping track of them, and communicating a final tally.

The problem is obvious, right? That person can do whatever they want and nobody is going to be able to prove otherwise.

Now, that problem can be mitigated somewhat. Exit polls can be used to analyze the statistical nature of the votes and compare against the tally. And if incremental vote counts are communicated, math can be applied to that as well. We can sort of gain confidence in the election.

Compare that to a room where people collect ballots. That room is secure, so the ballots can be aggregated with few worries, but is otherwise visible to everyone.

In case of error, the ballots can be brought out, examined in any number of ways and we can arrive at a very high confidence in the election outcome.

The basic problem we have with electronic input is that the electronic record isn't directly human readable. There is no way for the voter to know the record of the vote used for the tally actually reflects their intent.

With physical vote records, a few things are true:

1) The chain of trust between voter intent and the actual vote record is intact. The voter can understand their record matches their intent. With a touch screen, button, or some other thing, that record of intent is really just a grease smudge on the input device, while the intent walks away unrecorded.

2) Because of #1, say you press "bob" on the touchscreen. The computer responds with "yup, you pressed bob" and one thinks it's all good. But the computer can also add "Jane" to the tally, and the voter can't know. Or their vote can be ignored, fractionally counted, whatever. No chain of trust.

3) Electronics are made to state change. That's how we do computation. Paper, other physical media is very difficult to change in a way that cannot be seen. Erasing a mark, making a mark, all involve basic physical processes and the rules of the world we live in more or less make changes impossible to make transparent.

4) With physical media, the voter expression of intent is used directly to form a tally. With electronics, some interpretation of the voter intent is used, and not even used directly, as it's a chain of temporary states all the way through.

Why does this matter? Should this not be a fundamental crisis in computing if true?

Nope. Now think through the things we do with computing machines. I am writing this text. I can see the text fed back to me, and it could be modified without me knowing, but for the fact that said text is a dialog. We all have some concept of the state of things and modifications would very quickly come out of changes to the text, unless said changes were very sophisticated.

Banking always has redundant paper. Ever wonder why tellers are required to report the number of bills dispensed on a cash withdrawal, or you get a receipt and they keep one too? This is why. Same goes with merchants selling things, people buying them. Multiple, personally identifiable parties, are interested in the record and multiple, distributed often physical records are kept and reconciled regularly.

We can't do this with elections because of the following four pillars of a trustworthy election:

1) Anonymonity. Everyone's vote is private. We do not personally identify votes for fear of crime associated with votes.

2) Freedom. Everyone can vote or not as they see fit.

3) Transparency. A vote can be seen from the moment it is cast, through to final tally. (admittedly, many parts of the US do this poorly, but shouldn't)

4) Oversight. Everyone can understand the election process and this depends on transparency being in play. The law, the ballots, all of it.

When one factors in the input validity and forced voter trust problems inherent in electronic voting, the problem becomes super hard. Personally identifying voters is necessary to complete the picture and make something trustworthy.

This is why we cannot write voting software. It's not incompetence, or any other lame thing. It's an unsolvable problem, unless we want to make votes personally identifiable. Should we do that, there are lots of great solutions we can draw from.

First, hello Slashdot. It's been quite some time since I last logged in. Greets and all that jazz.

This flexible polymer with seeking type robot just oozed a comment. Elon is definitely out to see some future happen in his lifetime. I am a fan, just because he's perfectly willing to take the big risks and see what can be done. Works his ass off too.

Not for everyone, but while we've got him. Thanks! Let's hope for some goodness.

In my mind, there are two things:

One, any kind of two way interface. Yeah, there will be risks, but it's like having an extra channel. I suspect just getting it in there and figuring out how to setup the feedback loop needed for a person to be aware of it and explore is the hard part. Once they do? Look out. Pretty soon, talking to someone may not be any assurance you are talking just to them. They may be consulting all sorts of information resources, even just acting. Crazy times ahead. File all the obvious enhancements under this heading. Speculate away!

The other thing is fixing broken people. Seems like amazing possibilities here. That's exciting, if nothing else.

Microsoft is pushing a lot of testing onto early and non business users. What did they expect actually?

Secondly, Microsoft has moved to a rolling release style of development, while also pushing hard on features people aren't all that excited about. What do they expect?

If they really "demand answers", maybe they can fund the internal testing, etc... needed to get them, so their "beta" program may actually then deliver more meaningful feedback.

...an insane long standby time.

When I full charge that thing, it can take a call or text for like two weeks. Maybe using that mode is worth a thought. It's still useful for web browsing, e-mail, text, etc...

(And yes, I tested that on a long trip, no charger. Got 10 days, no problem, took a few calls, answered a few e-mails, various SMS.

And that's the usual, threats, etc...

As for the speech we don't like? It's up to them. The First Amendment applies to the government. Twitter can do whatever it wants.

Personally, I like the haters using Twitter. It's making it very clear, very quickly, just who is who. Trump is making an International ass of himself. Glorious!

Also, the First Amendment does not contain a shield. The answer to free speech we don't like is more free speech. So, that means we don't or should not ask Twitter to shut Trump down, unless it's criminal. And that also means we simply use Twitter to tell everyone just how big of an asshole Trump actually is!

Passes Popcorn Bowl to the right. Crunch, crunch....

Oh, and that is a somewhat difficult skill set to find.

I have it, and guess what? I contract out for pre-sales work all the time. Enterprise B2B people are just drooling over people who can do this. No wonder the talent isn't available to the big companies.

If they are smart, they do what I'm doing, and that is leverage that ability to understand new and old tech, create visions, back them with a strategic business alignment and value, and then help the salesperson pitch that to the execs for new sales.

(take the CIO out to lunch for bonus points)

Oh, that's a raw spanking! Well played.

Really, this piece should be titled, "Old execs can't find anyone young that is able to talk to them like their older buddies talk to them..."

Isn't the CIO the generalist who is able to articulate how the business can succeed with technology?

Ok then, get after it. Your new people and old people all have perspective. Get off your arse, talk to them, make some choices and go and sell that to management or prioritize the budget.

Delegating the budget is just fine, but even that needs a basic review. I understand how it is in very large enterprises, but I also understand companies of that size can afford to hire several CIO types too. Not all techs can be business minded, young or old. That's a specific skill set, and as a tech generalist, they would and should be expected to get what they need from the hard core techs, who will gladly give it to them too.

For smaller companies, if they even have a CIO type position, the generalist there needs to do the work to understand what the strategy actually is and what it means.

...grok programming more quickly and easily.

This all comes down to what one has to know in order to attempt some programming. BASIC requires one know very little to get something useful done. They try the PRINT statement, and that's cool. GOTO, INPUT, strings, numbers, basic math follow.

From there, you can do pretty useful programs!

EXCEL works a similar way. You see what cells do, then you find things like autosum, then you put a little bit of math in a cell, and suddenly, you can make some really useful spreadsheets. I know people with about that level of knowledge modeling businesses to great success. It's not the most advanced use of EXCEL, but it works fine, they can change it, they get the benefit of some automation and can communicate advanced ideas to others with relative ease.

Way back in the day, before EXCEL, I had used BASIC to compute a whole pile of useful manufacturing related things. Saved me a ton of time, and I sold those and some CAD system programs to get a reasonable PC. All development was done on some 8088 clunker from a thrifty store. (yes, it ran the CAD system, having exactly the minimum requirements listed on the box)

The CAD system had a BASIC like language built in. Was cake to do this. I did know something about programming, but I also was able to teach others how to make useful programs on just little nubs of knowledge. Some of them advanced, getting into IT, systems, etc... while others just used the programs they made and were happy about it.

Indeed! The print is too small.

Best use case for new programmers, is to maximize utility while minimizing knowledge dependencies. They don't need much to get the spark. Once they get it, as they progress, they will want out of whatever little environment they started in. The ones who really have aptitude will get out and do just fine. For many others, they will just use the thing and be happy, or move on and not care so much.

We really should give everybody a go. Find out who is who.

Think of this like public speaking. We make everybody do it, or most everybody. Most people experience an ordinary, "I can do this" outcome. Some of us find out it's not for us, and still others find out they are great somehow. We lose out if we don't run everyone through.

And of course, those of us who prefer humans make basic marks on physical media are right about all of this and talked about the expense and untrustworthy nature of voting machines.

Here in Oregon, we vote by mail, and are joined by WA and CO now, with some other pockets here and there in various states. It's awesome, works, can be trusted, is difficult to fraud on a scale that would impact anything, and turnout is generally higher than the poll methods in use most everywhere else today.

We can actually manually count and evaluate every last vote if needed.

Back in the Napster days, a flat license of something like 2+ Billion dollars was offered to get this same sort of thing started...

My early experiences were the old Atari VCS (2600) and VCS stood for video computer system. I was fascinated by the pixels and the idea of a TV being interactive.

I wanted control of the pixels.

Later, in school, I got to work on Apple ][ computers, and those just begged to be programmed. Gaming can initiate the desire, but so can a lot of other computer driven things these days.

It is not prep directly.

Indirectly, games can be prep. For a few friends and I, cracking copy protection got us into 6502 machine and later on, Assembly language. We would use the monitor to see what was going on. Reading the ROM listing told us a lot more.

BASIC is slow, and that too drove learning more. To get the real magic out of the old machines, one has to know stuff. We made games, played them and learned. Utility type programming was good too. One such program generated book reports with just a few picks and keyboard input.

Just playing, unless the game incorporates programming concepts, is not meaningful. The ability of games and other interactive things can spark the desire to build and control.

The latter leads to activities that do serve as prep.

C64 used a non-sequential scheme that mirrored it's character display.

8 bytes sequential on most machines means a linear series of pixels on the same scan line.

On the C64, those bytes got stacked up to form a character, each byte on a sequential scan line, assuming one starts at a character boundary.