Thread by @naukarshah

So @ECISVEEP finally to replies to an RTI by @OfficialSauravD on my letter.

They say nothing new.

But this is good. Very good.😋

For now they state as to what they rely on to say EVM-VVPAT is safe.

And all of it can be demolished with ECI's own documents. @SpokespersonECI 1/n

Take a good look at the reply once again.

Does @ECISVEEP say that EVM is a stand-alone machine anywhere?

No right?

For they know that their lies have been exposed on that.

You won't hear that argument from @SpokespersonECI anymore. 2/n

It says "The basis of @ECISVEEP's confidence flows from a wide range of administrative protocols and procedural safeguards that protect EVMs and VVPATs against any sort of manipulation.."

And goes on to present only two key safeguards. 3/n

1. "EVMs & VVPATs are randomised twice, while being allocated to an AC after FLC & then to a Polling Booth".

True. But both randomisations are rendered useless as both are done before commissioning of the EVMs when an external device is connected to EVM.

himachal.nic.in/WriteReadData/…

This is key, as at the time of connecting an external laptop/SLU to the EVM-VVPAT through a serial port, the information as to which booth this particular EVM is going to to go is already known.

Allowing for targeted & even booth level manipulation. 5/n

And that was the reason why I gave this suggestion to @ECISVEEP and @SpokespersonECI.

To do at least the second randomisation after the commissioning and connection of external device.

https://twitter.com/naukarshah/status/1362703304212369414?s=20

6/n

If @ECISVEEP can't even grasp this simple fact, I genuinely doubt the competence of the people manning the institution. At all levels.

Now let's come to the second aspect from where @ECISVEEP gets its confidence flown from.

The three stages of mock-polls. 7/n

Okay.. But before that, let me write bit more on the randomisation first. This is from ECI's own ppt on EVM.

Notice where it says till 2nd randomisation no one knows which EVM will go to which PS.

Meaning after 2nd randomisation we will know which EVM will go to which PS. 8/n

Here, the same presentation says, 2nd randomisation is to be done just before 'candidate setting' or commissioning of the EVMs.

I hope @ECISVEEP and @SpokespersonECI will now agree with me that both the randomisation is done before the commissioning of EVMs. 9/n

And it is during the commissioning that we connect laptop/SLU to VVPAT through a serial cable & application software.

I hope you understand why both stages of randomisation are useless now @ECISVEEP, @SpokespersonECI.

If you understand, take immediate corrective measures. 10/n

Now let's come to the second aspect from where
@ECISVEEP gets its next level of confidence from.

FLC & the three stages of mock-polls.

FLC & the first mock-poll is done much before connecting EVM-VVPAT to an external device during commissioning. Hence a useless check. 11/n

2. The second mock-poll after candidate loading using laptop wherein 1000 votes on 5% of EVMs is done can also be easily by-passed now as votes are date & time stamped in the latest EVMs with RTC.

With this info mock-polls are reduced to a mere farce.
ceokarnataka.kar.nic.in/genelc_2018/st…

3. The third mock-poll of 50 votes on the day of poll is the easiest to bypass as it is known prior enough that in every EVM, the first 50 votes will be verified.

Any coder will give you a line to work around that check. Also VVPAT has session information to assist further. 13/n

So, both randomisations gone, and all three mock-polls gone. Pufff. They are no more reliable checks.

Only thing that is left now to instil confidence in that long paragraph is about the sample audit that is done post-counting @ECISVEEP, @SpokespersonECI. Rest all gone. 14/n

But here is our lack of honesty in VVPAT design that comes to bite us @ECISVEEP.

We made a jugaad that is neither fully verifiable nor has a proper design architecture. Our VVPAT sits right in the middle b/w BU & CU. At liberty to cast vote as it likes & print as it likes. 15/n

We did not even care to read what the computer scientists who proposed the VVPAT design had to say about what is a proper VVPAT.

That VVPAT & EVM should be integrated & that voter should have full agency to verify & cancel vote if not satisfied.

Courtesy: Prof Banerjee, IIT-D.

That is why as an immediate suggestion, I requested to do these two things also @ecisveep.

But even this will only be temporary. And will not yield secure elections in its true sense.

https://twitter.com/naukarshah/status/1362704171728728064?s=20

We have royally screwed up our election process, my dear seniors at @ECISVEEP.

Please don't be obstinate in denying it. You are doing a great disservice to the nation.

Try to understand the concerns. Acknowledge. Improve.

Dismiss the existing TEC for an irresponsible design.

Be transparent in your workings, and accountable in your responses.

Don't start acting like a political party trying to defend itself.

You are not fighting an election. You are conducting one.

Act like it.

👆@ECISVEEP @SpokespersonECI

Btw, please don't make these process concerns into results concerns.

Whatever be the election results, these flaws in the processes shall remain.

Parties raising this only based on elections results erodes the entire credibility of questions asked.

Please don't do that.