Slashdot Mirror
Ohio Audit Reveals More Diebold Problems
armb writes with a link to a Wired Blog entry about irregularities found in Diebold databases from the state of Ohio. The election in question here is November 2006, and the corruption of the entries may raise doubts about accurate tabulations. "Vote totals in two separate databases that should have been identical had different totals. Although Diebold explained that this was part of the system design for separate vote tables to get updated at different times during the tabulation process, the team questioned the wisdom of a design that creates non-identical vote totals. Tables in the database contained elements that were missing date and time stamps that would indicate when information was entered. Entries that did have date/time stamps showed a January 1, 1970 date. The database is built from Microsoft's Jet database engine. The engine, according to Microsoft, is vulnerable to corruption when a lot of concurrent activity is happening with the database, such as what occurs on an election night when results are uploaded and various servers are interacting with the database simultaneously."
Jet Database Engine, a.k.a. Microsoft Access.
There are 0x40000000 types of people: those who understand 32-bit IEEE 754 floating point, and those who don't.
Now, I'd never think about developing this on a Microsoft Jet DB, since it's been somewhat deprecated for the MS Desktop SQL Server (MSDE) and SQL Server 2005 Express, which are much better and lightweight enough for a current desktop.
Nonetheless... what MS probably stated is that basically access to a JET Db is not thread safe, which means that concurrent access will cause corruption with a probability directly proportional to the amount of activity. YET if you serialize access to a Jet Db (which is a necessary and basic requirement given that it's not thread safe) there shouldn't be a fear of corruption, unless the API is buggy. If each voting station has a Jet Db and they all get exported to a central (thread safe) db then there's no need for concurrent access to any of the individual Jet DBs, and there shouldn't be a big fear of data corruption (which, anyway, can be verified somewhat easily).
As a Slashdot discussion grows longer, the probability of an analogy involving cars approaches one.
Reading this made me think about my time doing safety critical systems (it fails, someone dies) and its really stunning to think that something like voting in a democracy isn't considered mission critical to the country.
There really is no excuse for voting to not be done on a comparative basis e.g. every vote to be checked via 3 different software lines (this isn't rocket science) and a voting system to then confirm that the vote is being applied correctly. This vote should then be written to two (at least) data sources to enable reconciliation at the end.
This is a freaking implementation of a check-box system where is the sodding complexity that means its expensive to be professional.
Voting in a democracy is mission critical, to not consider it that way is to say that voting doesn't matter.
An Eye for an Eye will make the whole world blind - Gandhi
Diebold makes ATM's, but the banks keep all the records.
That's a bit like saying Panasonic would make a great a telephone carrier because they make phones.
"Learning is not compulsory... neither is survival."
--Dr.W.Edwards Deming
It was always thus... Two spaces after a period is only appropriate in circumstances where all characters are the same width, such as an old-school typewriter. So nobody “decided” that it would be that way “on the Internet;” we just stopped using the special-case rules that sprung up a few decades prior when we were using technology that wasn’t capable of proportionally spaced type.
Quantum materiae materietur marmota monax si marmota monax materiam possit materiari?
Diebolds electronic voting division was purchased wholesale from Global Election Systems in 2002. GES produced crap back then and it is no suprise that they continued to produce crap under new management. Their incompetence shouldn't reflect poorly on the ability of the engineering staff in the ATM division, although it does say quite a bit about the top-level management.
And never attribute to incompetence what is clearly not. Diebold makes TONS of other electronic transaction machines (include, probably, your bank's ATM machines) and they don't have these kinds of problems. Perhaps they do it on purpose to give them a cover of incompetence. Perhaps it really is incompetence on the part of the guy they get to write this stuff since whoever is hiring him doesn't care if he's incompetent because they're going to fool with the results anyway and it will only add cover for them.
But CLEARLY this kind of stuff is not because Diebold isn't capable of doing it properly. It's because they explicitly don't want to do it properly.
If we're going to have electronic voting machines, and I don't think that we should (not even optical scan), they should be developed, owned, and maintained by the government. Period.
The Farewell Tour II