UPDATE: In the last 24 hours, I’ve had seven different coders — other than the first one who sent me the story idea in the first place — take a crack at the survey’s page code. The results are wholly infuriating: three found the same thing my original source, where the data inputted into the survey was flushed out on “Submit.”
Three more said I was full of it (and most used the term that I abhor: ‘fake news’), and sent along screenshots of the Development Console showing the data was captured correctly and passed on to the server. Example:
Okay. Fine. There was one theory that the original survey, which was posted on GOP.com, was then hastily thrust over to DonaldJTrump.com, and all of the necessary query hooks weren’t added onto the backend. (For those less tech-inclined: the transplanted code wasn’t properly installed.) However, someone else found the survey to work fine on both servers.
Then, there was the developer who found the results are POSTed to the secure DJT server, and that the server is ignoring them and redirecting to the “contribution” page.
“So: the data is transmitted to the server. But it appears that the server is not actually recording that data (except possibly the logs). Most definitely it isn’t putting it in a database for careful consideration. It seems weird to think that they wouldn’t be holding onto that information for further mailing purposes, but maybe they just aren’t interested in anybody who isn’t willing to pony up cash the first time.
“So: Yes, the web page transmits the data. But the server is responding with HTTP 302 (REDIRECT) which basically means ‘go somewhere else.’ It usually means that the data isn’t being processed, but without the server-side code it’s impossible to tell.
“Smoking gun? Yes. Bullet hole? Nope.”
That slight pop you just heard? That was my brain, leaking out of my left ear.
Even if the data is being correctly mined, I have zero faith that dissenting opinions are being taken seriously with this survey. Katie Wohlgenant has given really solid reasoning on why we should stop taking the Survey. But when the results are really inconsistent from attempt to attempt — even when seen in the span of five minutes… I just don’t know.
That said: please don’t let it be said that I want to spread a false narrative. I’ll leave that to other people in the 202.
Here’s the original story, as published:
I was chatting with a web developer friend earlier this evening, and he stops in the middle of the conversation. “Hey, you know that Media Accountability Survey the GOP is hawking? It’s a fraud.”
I’m not surprised by the statement, but my friend isn’t given to hyperbole. “How do you mean?”
“The thing only exists to raise money for the party coffers. No one sees the polling data, because it’s never sent.”
After I stop laughing, I ask him to explain it to me. I’m technically inclined, but I have a feeling I’ll need Cliff’s Notes for this one.
He busts out a series of screenshots from his development console, which he had running while taking the survey. “You can do this yourself — fill out the survey, but don’t hit ‘Submit.’ Open Chrome’s Developer Tools, and select Network (to get the network activities). Then, back in the browser window, submit the survey.
“This shows the order of requests — the first five are the important ones.
“First, you’ll notice a curious thing — a POST (the form submission) is cancelled.
“Now, the POST before it was accepted and redirected. That POST request contains your contact information.
“The POST that was cancelled? That contains the survey data — basically, throwing it away.
“And, lastly, there’s a redirect that goes to the donate page.”
I was a little taken aback with the brazen nature of it. “You’re telling me the whole thing is a Trojan Horse to get people pissed off enough to want to donate to the GOP war chest because they don’t trust the Liberal Media?”
“Hey, man… it worked for the televangelists.”