Iowa Caucus Debacle is One of the Most Stunning Tech Failures Ever (cnbc.com) 439
The Iowa caucus debacle represents one of the most stunning failures of information security ever. From a column: This failure was delivered by the same Iowa Democratic Party officials who have said for the last four years they were "ramping up" their technology capabilities, convening seemingly endless security task forces to ensure foreign powers did not disenfranchise voters, and collaborating with federal agencies like the Department of Homeland Security to make sure everyone was in the loop on voting security. Voters will be paying close attention to how party leaders ensure that votes going forward have clear contingency plans in place, not just to protect against hackers, but from all types of technology failures, including applications that might not work.
Iowa officials counting the results coming in Monday from the caucusing app reported irregularities that required them to switch from the app to counting votes manually. Party officials said the "underlying data" put into the app was fine, but it is unclear as of yet how they know this or even what they consider "underlying data." "Last night, more than 1,600 precinct caucuses gathered across the state of Iowa and at satellite caucuses around the world," the Iowa Democratic Party said in a statement Tuesday. "As precinct caucus results started coming in, the IDP ran them through an accuracy and quality check. It became clear that there were inconsistencies with the reports. The underlying cause of these inconsistencies was not immediately clear, and required investigation, which took time."
Iowa officials counting the results coming in Monday from the caucusing app reported irregularities that required them to switch from the app to counting votes manually. Party officials said the "underlying data" put into the app was fine, but it is unclear as of yet how they know this or even what they consider "underlying data." "Last night, more than 1,600 precinct caucuses gathered across the state of Iowa and at satellite caucuses around the world," the Iowa Democratic Party said in a statement Tuesday. "As precinct caucus results started coming in, the IDP ran them through an accuracy and quality check. It became clear that there were inconsistencies with the reports. The underlying cause of these inconsistencies was not immediately clear, and required investigation, which took time."
On the contrary! (Score:5, Insightful)
It's all going exactly as planned
Re: (Score:2)
Thinking the same thing, but a question or two still lingers...
Re:On the contrary! (Score:5, Insightful)
The Superdelegates will still decide the Dem nominee. The rule change following 2016 only says they don't get a say until the second ballot. It's clear that there probably won't be a nominee on the first ballot unless they allow Bernie to win and all indications are they're rigging the primaries against him already again. I still fully expect Hillary to be chosen in a brokered convention - that's been the Clinton Machine strategy since Nov 2016.
It's going to be a hilarious 10 months until election day. Trump might win 45+ states.
Hanlon's razor [Re:On the contrary!] (Score:5, Insightful)
A new app being rolled out for the first time, being used mostly by retiree volunteers who are, to boot, in areas with poor data access?
Yep, stupidity. Not conspiracy.
Re: Hanlon's razor [Re:On the contrary!] (Score:3)
That is a very bad heuristic in a context defined by motivated reasoning. Furthermore, it's known that the developer of this software received money from Buttgieg. It's pretty obvious at this point that the CIA just wants to apply its considerable experience to manipulating a domestic election.
Re: Hanlon's razor [Re:On the contrary!] (Score:3)
"That being said, this company and its parent has received money from many different candidates."
"And it's parent" feels like a very important distinction to me. Shadow is a new company spun off from that other one, and the only candidate I've heard of paying it is Buttgieg. Between his premature victory announcement and the fact he's CIA, this shit looks very suspicious to me.
Re:Hanlon's razor [Re:On the contrary!] (Score:4, Insightful)
The problem with your idea is that the DNC conspired before. Why wouldn't they do it again? While we're talking about what it's safest to assume, it's safest to assume that a tiger hasn't changed its stripes.
Re:Hanlon's razor [Re:On the contrary!] (Score:4, Insightful)
So there are deft, shadowy puppet masters behind the scenes able to change the outcome to whatever they want, but unable to do so without planning ahead and calling national attention to the method they're using to do it? Yeah, that seems way more likely than that they overpaid some company to build an app that was inadequately tested, had inadequate training, and was probably inadequate for the conditions in all the counties it would be used in...
Assuming there has been a conspiracy, in no way does that mean that *everything* that goes awry from that point on is a part of that conspiracy.
Re: (Score:3)
Re:Hanlon's razor [Re:On the contrary!] (Score:5, Insightful)
Was just going to say the same thing. Is there actually anyone here who couldn't see this coming? Of course it was going to fail badly, it's a large-scale IT project rushed out under time pressure, and it involves online voting. It's 100% guaranteed to fail, or at least to have serious problems, this was predestined the minute it was initiated.
And a nitpick about "one of the most stunning tech failures ever", it's nowhere near that, it's just business as usual. The really big failures cost billions or tens of billions and drag on sometimes for decades.
Also interesting to see the masses of conspiracy-theory crap around this. Every other major IT project failure ever is due to mismanagement, unrealistic schedules, constantly changing goals, etc etc, but this one very specific project is because {insert favorite conspiracy theory here}. Yeah, definitely. A conspiracy.
Re: (Score:3)
You can't. The prior precinct chair, a state party rep, or failing the two, someone else entirely, usually calls the caucus to order, but you then immediately elect the precinct chair. If you're saying you need to require anyone who stands for precinct chair election to have already installed an app and passed some test designed to show they can use it, that's fine, but it's not exactly the kind of heavy-handed on-high revision to the system
Re:Hanlon's razor [Re:On the contrary!] (Score:5, Informative)
The biggest snippet that stood out to me on this one is the "had trouble installing the app" reports. What. The. Hell. Why the shit are you waiting until the 11th hour to even install the app? You should have had it installed on every caucus leader's phone and had them test their log in credentials at least a week ago.
There was a test a week ago, although a small scale one. The software failed it:
"The problem with Iowa’s app was foreshadowed during a test in the days before the caucuses, in which precinct managers said they were having trouble using the app, according to Polk County Democratic Chairman Sean Bagniewski. So last Thursday, Bagniewski told people who couldn’t get the app to work to call in their results to the Iowa Democratic Party as a backup.(source: https://ktvz.com/lifestyle/tec... [ktvz.com] )
The problem was: the call-in reporting was supposed to be only a backup, not the main reporting method. It wasn't ready for the large volume of calls, and broke down.
More: https://www.theverge.com/2020/... [theverge.com]
https://www.vice.com/en_us/article/y3m33x/heres-the-shadow-inc-app-that-failed-in-iowa-last-night
https://www.psychologytoday.com/us/blog/mental-mishaps/202002/iowa-caucus-failures-bad-design-and-disinformation
https://www.iowapublicradio.org/post/questions-remain-about-caucus-results-app
Re: (Score:3)
Maybe my standards are too high, but...
an election software?
on the way towards the Presidency?
of the most powerful country on Earth?
Seriously? That's been given to a no-name small-time developer company with no oversight and quality control? That's too much stupidity in one place to explain away with "ah, shrugs, happens all the time"
Re:Epic fail [Re:Hanlon's razor [Re:On the contrar (Score:5, Insightful)
...a no-name small-time developer company with no oversight and quality control?
The no-name small-time developer is owned by the wife of Mr. Buttigieg's campaign manager, which might be significant.
Re:Epic fail [Re:Hanlon's razor [Re:On the contrar (Score:5, Informative)
People who work in politics are related to people who work in politics. News at 11.
No seriously. People see conspiracy theories in things like this all the time, but the reality is industries (of all kinds) tend to inbreed the people who work within them.
I'm not at all surprised when I see someone married to one of their workplace vendors, or someone who subcontracted to someone whose partner becomes a service supplier. A lot of industry contacts are made through word of mouth and the power of personal exchanges at parties / conferences / events is incredible across all industries. In some cases it's inside insight to a developing need, for example nearly all service providers to the oil and gas industry were used to work for a major company within the industry, saw a need, quit, and started developing.
Now if we can find evidence of someone rigging a competitive tender, or bypassing a procurement process requiring one, then we'd have ourselves a decent conspiracy.
Re:On the contrary! (Score:5, Insightful)
still fully expect Hillary to be chosen in a brokered convention - that's been the Clinton Machine strategy since Nov 2016.
You have a rich internal life.
However, reality doesn't follow a movie script as much as you want it to.
Re: (Score:3, Insightful)
I think the maxim (Score:5, Insightful)
"Don't attribute to conspiracy what can be explained by stupidity" works here. Attempt to use new technology (in this case software) = guaranteed failure.
Re:I think the maxim (Score:5, Interesting)
Not obvious that a technical issue would have caused this long of a delay. Long delay to me smells more like an internal fight is delaying the release of the results. We'll probably never know the truth.
Re:I think the maxim (Score:4, Interesting)
Not obvious that a technical issue would have caused this long of a delay. Long delay to me smells more like an internal fight is delaying the release of the results. We'll probably never know the truth.
I agree. I have participated in caucuses before (R not D), been a district and state delegate, and monitored caucus results for presidential campaigns. My wife and I were Parliamentarian and Secretary, respectively, for our last county caucus (so she did the reporting and all of the caucus officers watched her do it so we could attest to the results). With no tech at all, it wasn't that complicated. A complete manual count from the caucus paper records doesn't take that long and in Iowa, they allegedly had cell-phone-photos of the records to work from (no need to move paper records). We scanned and emailed ours for rapid counting and then mailed the paper for validation.
Which isn't to say we don't have math errors and mistakes that need to be checked. We certainly have. But the biggest count delays were because of contested or train-wrecked caucuses, not the process of counting itself. In 2008 in Missouri, a large number of contested delegates had to be sorted out on the floor of the R state convention and some of the District Conventions, not because of a tech error or bad math but because the results were not politically acceptable to some folks in the GOP structure (and because one of the county caucuses had the elected Chairman--- a friend of mine--- trespassed out by the local party leadership injured and in handcuffs...).
Anyway, so this stuff can get dicey, but the math isn't that hard.
Re: I think the maxim (Score:3)
I don't understand why when the new software tool failed, why their fallback solution also failed so spectacularly. They didn't have this issue 4 yrs ago, why wasn't the 2016 method their fallback?
Reports are they REFUSED to actually train precinct workers on the software before the caucus day. That's amazingly stupid (if true). Their fears of hacking all but guaranteed their system would fail.
Re:On the contrary! (Score:4, Informative)
We already know:
1. A company called Shadow Inc made the failed app.
2. Pete for America, the campaign organization for Pete Buttegieg, paid Shadow on/near $42,000.00
3. The CEO of Shadow Inc is a big Pete supporter. Look at her Twitter, she is a big fan.
4. The husband of the CEO of Shadow works on the Pete Campaign as a staffer.
NOTHING SHADY HERE AT ALL FOLKS.
SORRY, BURNARD. YOUR VICTORY AND ANY "BOUCE" WILL BE BURIED UNDER THE SCANDAL AND TRUMPS SOTU. WHAT A SHAME! /s
Re: On the contrary! (Score:4, Insightful)
You assume the goal of interference is to change the voting outcome. That is the worse case scenario, but another primary goal of interference is simply to create doubt in the election process. Similar to claims of fake news, disenfranchising voters and increasing national polarization is in the interest of many foreign and domestic political actors.
Re: (Score:3, Informative)
The intended effect was the same as the intended effect of declaring Mitt Romney the winner in Iowa back in 2012. It's to blunt the narrative of a political outsider starting with "momentum." Sure, some weeks later they back tracked and admitted Rick Santorum actually won. By that time though, Rick's campaign was dead in the water.
I doubt there was any vote tampering, as caucuses are open votes where you have to stand up and be counted in person. This simply gave them the ability, if things were looking go
Re:On the contrary! (Score:5, Informative)
Re: (Score:3, Insightful)
Don't tell that to bernie bros! They start frothing at the mouth in anger praising how great Gulags were.
Re:On the contrary! (Score:5, Insightful)
Polls are more accurate in primary states like NH than in caucus states Iowa.
Voting in a primary is easy. I just tick a box and put the ballot in my mailbox. So the participation is high and the voters are the same normal people that participated in the opinion polls.
Voting in a caucus requires a serious commitment of time and effort. You need to drive to the caucus site, often on cold icy roads, and spend half a day caucusing and negotiating. The people willing to do this are not representative of the general public, thus the polls aren't as predictive.
Re: (Score:3)
Given his speech last night. I think it is safe to say his leg hair was not standing tall glistening in the sunlight.
Re:On the contrary! (Score:5, Funny)
How did Biden do?
He did great, got >80% of the vote
Where are you located?
Kiev
Re: (Score:3)
Re:On the contrary! (Score:4, Insightful)
Either that or the Donald is praying that the Dems pick Bernie, and acts scared of him because that'll help get the Dems to pick him....
Re: (Score:3, Interesting)
Sanders also appears to have some better people working for his campaign this time around. While all of this uncertainty was in the air they put out the
Re: (Score:3)
Yes, I agree that the biggest benefit to party insiders and their big money donors is that this gives cover to Biden's piss poor showing.
Thank you for your nuanced take on this. If people on both sides were as fair minded as you, politics would be a much more civil endeavor. If you ever care to engage in civil debate over Bernie's policies, just ask.
Re: (Score:2)
They were supposed to have results much faster than this. But, there were warnings yesterday that their tech wasn't going to work.
Re:On the contrary! (Score:4, Informative)
1 Officials in areas with not WIFI and poor cell reception are unable to access the internet
2. Officials with old phones are experiencing crashes and other technical issues.
3. Officials with little tech skill and training can't figure out how to report results with the app
Still inexcusable, all these issues could not only been foreseen, they were to be expected.
Re:On the contrary! (Score:5, Insightful)
Wrong.
The problems are due to one and only one thing: Lack of testing.
Before you deploy something like this, you give it a real-world test. At minimum one. Better to do several.
Months in advance you pick an evening and have everyone with the app all send you numbers you already know. Then you see if they match. And then maybe you do that a few more times just to gently unit test the thing in as close to a live setting as you can.
That would uncover all of these problems, and let you fix them well before it's go time. Not least the "not enough internet connection to download the app" problem. FFS, you could do it every Monday night for 3 months until all the bugs were shaken out.
Lack of testing AND lack of training (Score:4, Interesting)
The problems are due to one and only one thing: Lack of testing.
No, lack of training was *also* a problem. Some caucus officials were installing it for the first time last night. They got the standard Android warning about side loading an APK and mistook this warning for a malware warning. They thought they were about to be hacked and did not install it. Also there was apparently no real security designed into this system as officials were supposed to use their personal phones for reporting election results.
Re:On the contrary! (Score:4, Insightful)
The first step in a caucus is to elect the chair. So the person in charge isn't known until after the caucus starts. Which means they can't pre-install the app on that person's phone before the caucus. Nor can they train the chair before the caucus.
That being said, they could distribute official DNC phones/tablets with the app already loaded to each caucus site, and have someone from the party present who was trained on the app to answer questions.
Or just do it the old fashioned way and call in the results.
Re:On the contrary! (Score:4, Insightful)
Which means they can't pre-install the app on that person's phone before the caucus. Nor can they train the chair before the caucus.
Which is a challenge that was known before development even started, and should have been considered in the design step.
Re:On the contrary! (Score:4, Insightful)
Re:On the contrary! (Score:4, Insightful)
Installing the app and logging in for 100% of the responsible people should have been done
The first step in a caucus is to elect that responsible person. Which makes it rather difficult to pre-install the app on the phone and give them a login to test.
Re: (Score:3)
How were credentials distributed?
Apparently via side-loading the app via "TestFairy".....and there is new reporting that they forgot to upgrade from the free version that only allow 200 installs. Which might be slightly problematic for something run at 1600 locations.
Re:On the contrary! (Score:4, Interesting)
Re:On the contrary! (Score:4, Insightful)
Good luck doing this with 1600 unpaid people who have busy lives and no reason to do what you tell them.
I kind of agree with you, but kind of not. If they cared enough to volunteer they would care enough to do it when you told them it was a condition of them running a caucus location. The people I've seen that volunteer for things like this treat it as a privilege. If they aren't willing to dedicate an extra 15 minutes to it, then find someone else who will.
Re: (Score:3)
Re: (Score:3)
I'd say the root of the problem is that the app was slapped together in two months. There's simply no way to do a proper job in that short a time. Everything else stems from that.
https://www.nytimes.com/2020/0... [nytimes.com]
And I agree that this is inexcusable. It's not like this caucus sneaked up on anyone. They literally had years to get this right. When the Democratic National Committee recommended the old system be abandoned, someone should have had the internal fortitude to say "That's a great idea. We'll lo
Re: (Score:3)
I'd say the root of the problem is that the app was slapped together in two months. There's simply no way to do a proper job in that short a time.
Bullshit.
This "app" just sent numbers attached to names back to a server. If you can't build, test, and deploy that in a week, you shouldn't be doing that job. 12 names, total votes for each name. That's it. No geolocation, no graphics, no 3D, no fingerprint scanning, no using any of the hardware on the phone, nothing that would be remotely difficult. You're not building your own encryption, FFS, you're not even doing push notifications!
And in that case, why is this even an app? You could have whipped this
ageist bastards (Score:3)
Re: (Score:2, Troll)
And to think that I wasted all those mod points on the Ethiopians writing homework assignments for Western shitlib-turd children of affluence.
[Which actually is part & parcel of the same phenomenon that we're talking about here.]
Approximately 50% of the USA no longer lives in Reality, and is now being tossed around psychologically in a violent maelstrom of Frankfurt School hypnotic cultural poisoning, utterly oblivious to their own personal absence of Freedom of th
Re: (Score:2)
Re:On the contrary! (Score:4, Funny)
paving the way for a contested convention where she can swoop in and declare victory.
Do you expect that she'll do it while riding her broomstick?
Don't Die on the Cross of Cool (Score:4, Insightful)
Apps may be cool, but you should probably just stick with what works.
Re: (Score:2)
Ever since the punchcard failure in Florida in 2000, we've been trying to get this right. Appears these iPhones are not yet part of the solution.
Re:Don't Die on the Cross of Cool (Score:5, Funny)
Ever since the punchcard failure in Florida in 2000, we've been trying to get this right....
Trying? You say that as if we didn't have any valid solutions for 100+ years prior to Chad hanging onto a ballot box.
What's wrong with K.I.S.S. solutions again? Not enough corruption?
Re: (Score:3)
What's wrong with K.I.S.S. solutions again? Not enough corruption?
Too many people involved who might cry foul.
Imagine ballots were counted by hand, with observers from both parties (and anyone else who wants). Then reported upwards by phone calls or some other simple tech that is also easy to observe for people without any tech training. Again, observers are present at both ends of the phone call.
Not easy to mess with the numbers enough to make a difference without having a bunch of people involved that need to be bribed or blackmailed. Too many people for not at least on
Re: (Score:2)
It's just so damned hard to poke a hole in the ballot when it's on an iPhone. Clearly they need a sharper stylus.
Re: (Score:2)
It's Easy! (Score:3, Funny)
Just keep pointing the other direction long enough for the Clinton apparatus to steal the election from Bernie.
Re: (Score:2)
Re: (Score:2)
Consider it a metaphor. The closet right wingers still want someone who would have been a passable GOP candidate just a few years ago.
Re: It's Easy! (Score:5, Funny)
Literally f'n true - Hillary staff wrote the app!! (Score:3)
Consider it a metaphor. The closet right wingers still want someone who would have been a passable GOP candidate just a few years ago.
But its not a metaphor, its literally f'n true. Former Hillary staff started the company that wrote the app!!!
... The LA Times notes that Shadow began life as Groundbase, which was founded by former Clinton 2016 digital campaign staffers Gerard Niemira and Krista Davis."
"At the center of the confusion is an app reportedly built by a for-profit company called “Shadow Inc.”
https://techcrunch.com/2020/02... [techcrunch.com]
Re: (Score:3, Funny)
She'll be chosen as VP to an octogenarian who will unexpectedly die of suicide by a bullet in the back of the head after packing bags for a vacation.
Re: (Score:2)
One *big* problem with your "Conspiracy Theory." Clinton is not running.
...now. Wait until the convention when things get chaotic. I'm not saying there will be a Brokered Convention, but...
Re: (Score:2)
Except he said Clinton "apparatus" which in this case likely refers to the established traditional Dem base (Schumer, Schiff, Pelosi, et al) currently backing Biden. Everyone knows the DNC really wants to throw their weight behind Biden this cycle, as they did with Hillary last time.
Re: (Score:2)
Yet.
Re: (Score:2)
Re: (Score:2)
One *big* problem with your "Conspiracy Theory." Clinton is not running.
That's just what she wants you to think... :)
It's not too late to jump in, so don't rule her out quite yet... Heck, if this thing goes to the convention to decide, it's *possible* (albeit unlikely) for her to broker a deal and end up the candidate again.
No, I'm not advocating this as a likely scenario, but at this point, the field is still as wide open as Iowa's corn fields.
Re: (Score:3)
The app was made by ex Clinton supporters (Score:3, Informative)
Re: (Score:3)
What do you mean, "ex Clinton supporters" . . . ?
I betting that tomorrow CNN will report that Hillary won!
Re:The app was made by ex Clinton supporters (Score:5, Informative)
The app was developed by "Shadow, Inc". Yes, really. Apparently we now live in a bad superhero movie.
https://www.vice.com/en_us/art... [vice.com]
Shadow, Inc's CEO is apparently an ex-staffer from the Clinton campaign. Though the company has done a lot to not reveal anything about itself.
Paper Beats Electrons (Score:5, Insightful)
Such a public disaster as this should end the argument (or at least the excuse of ignorance), unless you subscribe to the alternate theory that this is all a DNC cover story to avoid or delay the announcement of a slamdunk victory for Bernie.
Re: (Score:2)
At least with paper votes and physical ballots they have to visibly intervene to make a fraud.
This is just a foreshadowing of Nov. elections (Score:5, Insightful)
There will be glitches and delays aplenty until the vote count comes up the way the party bosses of the local districts dictate. We know there is corruption in the system. Notice how many uncounted ballots are found days after the election? Electronic voting only makes the corruption more efficient.
Re: (Score:2)
Maybe they should ask India how to organize elections? They could ask some develop country but it's probably better to learn from someone at your own level.
"Tech failure" (Score:2)
As if competent engineers can't build a reliable voting system.
Maybe my assumptions are too high.
No one's saying what actually went wrong (Score:4)
So we don't know if there was any or all of:
1) One or several types of problems,
2) A basic tech develop, test, deploy, test-at-scale process fail resulting in lack of scaled performance, or deadlocks, or whatever.
3) An app tallying logic fail.
4) A hard to use for beginners UI/UX compounded by lack of a practice (with real users) and feedback and fix and retest round.
5) Some kind of user authentication debacle
6)
Re: (Score:3)
As if competent engineers
It's not particularly clear that competent engineers were involved in developing an app in two months and deploying it on election day with no large-scale testing.
Causing problems or Uncovering problems (Score:2)
They ran accuracy and quality checks which showed problems.
Did they run similar checks in prior cycles? The Iowa caucuses are a series hundreds/thousands of meetings run by people who may in many cases being doing it for the first times and then phoning their results in. Do they really think there were no mistakes in previous years?
One of this years checks was comparing photos of calculation sheets with the numbers entered in the app. Did they have such double checking in the past?
I suspect that previous ye
Double Fail (Score:4, Informative)
When many of the leaders of the various caucuses could not report using the app, they attempted to do it "the old fashioned way" and call their results in. But the people in charge apparently had far fewer people manning the phones this time around because they wouldn't be needed, thanks to the new technology.
So much stupid, plenty to go around.
Rampant Paranoia (Score:3)
Not even close to being a major tech failure. (Score:4, Insightful)
With this nothing of value was lost. The votes will still get counted and even with little petey claiming he won that will not matter.
If anything this should go down as a political failure. Previously working and tested software was dropped so that they could pay money to a company owned by high ranking democrats. That software was not tested because they wanted to keep it under their own control.
Re: (Score:3)
Getting to play a video game or not does not change anything in the world. The importance of the Iowa caucus is not how many delegates get assigned. It's about creating a buzz for the candidate that wins or does better than expected that they can then try to build off Millions of dollars were spent and thousands of volunteers for campaigns worked the phones to try to get this buzz for their candidate and they have nothing to show for it.
They had a window last night where they were going to be the story.
Iowa reporting app was terrible (Score:5, Insightful)
According to https://www.nytimes.com/2020/0... [nytimes.com] the app was a custom development, started two months ago, that only started field testing less than a month ago, with no "at scale" testing or training. So whoever decided that they should inject new, untested software into an election on short notice is an idiot. As is whoever promised that they could reliably deliver a mission-critical application on short notice on a $63,000 total budget. It was also terrible planning to have a fall-back plan of a call center staffed with only a dozen people when nearly 1,700 precincts called in at the same time, and each call took 20 minutes to work through.
https://www.nytimes.com/2020/0... [nytimes.com]
I think that the improved transparency that they agreed to four years ago, in response to the concerns raised by Sanders' campaign, probably helped reveal the issues in the process, because when you publish the raw votes for the first and 'realigned' ballots, and the estimated delegate equivalents, it becomes obvious if there are any issues. Specifically, they said that the way they knew there were problems is that the three numbers didn't align. Previously they just released the estimated delegate counts, making it easy to obscure any issues with the election process. The actual caucus process and rules have been essentially unchanged for decades.
Re: (Score:3)
They would have been better of developing for a specific device (like a $89 COBY tablet) and distributing those single purpose, locked down devices to the precin
Re: (Score:3)
Could actually be a success, not a failure (Score:3)
Let me preface this by pointing out that I haven't the vaguest idea what sort of system they have put in place, so my argument here could be complete crap.
There has been a lot of really great work done in the last decade or so by the academic security community into the area of election auditability. The idea is that, under a pretty stringent set of assumptions, you can set things up to allow you to mathematically prove that ballots weren't lost or miscounted, and this can be done without violating voter anonymity, or enabling vote-buying or voter coercion. This is far, far better than simply using paper trails to enable manual recounts, because recounts can only count the ballots that get counted. The audit systems can't, of course, count ballots that were lost, and they can't necessarily identify fake ballots that were fraudulently added... but they can identify when something like that has happened.
It may be that the Democratic party has implemented one of these systems and that the system provided evidence of manipulation. If so, this may well be short term pain for a long term improvement in the integrity of the caucus balloting process.
In the long run, we should look at requiring the states to implement auditable elections in the generals as well, though more research into how to not only identify but recover from fraud may be needed first. It would arguably be worse for democracy for us to be able to prove that our election was manipulated but not to be able to do anything about it than it would be for us not to know, or merely to suspect.
Blame the "APP" culture (Score:3)
Also from published reports, it sounds like this app was installed on volunteers personal devices, so that greatly added to the number of variables in play in deploying the app (if they didn't want to go the html form option, they might have been better off just building the whole thing on a COBY tablet and distributing 1700 $89 tables pre-configured)...Some of the Iowa Public Radio interviews that played over noon hour with caucus attendees, said some their precinct leader were installing the app for the first time during the caucus, other commonly published accounts indicate no training on the software was given to the leaders who needed it....I mean wouldn't you have training sessions on how to lead and report the caucus in the week before, as it sounds like there was confusion in the process, not just the technology?
No load testing or user training (Score:3)
Can't imagine why it failed.
Technical aside: I was surprised it took caucus captains so long to do what we did in the 1996 caucus here in Washington State, or that Bernie did in 2016 here. You text a pic of the result sheet to the appropriate person, and drive the sheet to the counting location. While doing that, another person phones in the results, even though you're in a very slow queue.
Obviously Washington State, as a more Hispanic state with more African American and Asian population, should be First in the Nation. And we'll report the results when we finish them, in 7-10 days, since we vote on paper ballots by mail.
Fixed it! (Score:3)
Bernie
"It's broken. Can you take a look?"
Ok, it's fixed now.
"Who won?"
Bernie
"It's still broken. Let me know when it's really fixed."
Not to sound like a Luddite, but... (Score:5, Informative)
Re: (Score:2)
Well the caucus format has some safety features such as "Go stand over there if you support that candidate..." which protects against ballot stuffing.
Re: (Score:2)
LOL.. It's REALLY hard to just stuff more votes in the north west corner of the room unless you have bodies to throw that way.
I suppose you could go dig up some bodies... But emptying graveyards is another whole level of ballot stuffing and you don't want to get caught with those ballots in your trunk..
Re: (Score:2)
Well the caucus format has some safety features such as "Go stand over there if you support that candidate..." which protects against ballot stuffing.
You just stuff the voters so they are fatter, making that candidate's support area look fuller.
Re:Hanlon's Razor (Score:4)
Re:Hanlon's Razor (Score:5, Insightful)
Re: (Score:3)
Actually the Iowa Democratic caucus is a type of ranked/runoff voting system. You can read about it on FiveThirtyEight [fivethirtyeight.com]. This is why the results reported by CBC are different than reported by CNN for the same point in the counting - CBC are reporting the "final alignment" tally, while CNN are reporting the "state delegate equivalents" count. Neither are reporting the straightforward first preference count of the "first alignment".
Re: (Score:3)
Before this year, they didn't report "134 votes for A, 120 votes for B". They reported on the number of delegates won, and the math behind that is surprisingly opaque. There's rules for rounding, splitting delegates, etc such that it's not a simple percentage. Which makes it easy to fudge the result a little, and the people who attended the caucus won't know.
This year they're reporting more of the raw vote totals in an attempt to reduce claims of said fudging.
So depending on your personal level of conspi
Re: (Score:3)
What truly independent entity would do it for $60k?
The total campaign costs towards the presidency are expected to be several billions.
I'm sure they needed to cut costs on this measure and couldn't afford to spend ten or twenty more for a proper piece of software. You know, in a caucasus that everyone considers one of the more important ones. Yeah, 60k or 80k or something, that really makes a huge difference if you're trying to get one of your people into the White House, your first thought will certainly be to save money on the thing that counts the friggi