3 ms·

Remember the xerox-scandal years back, that invalidated officially scanned documents. You already life in a world where "scanned before" means you can legally c

Remember the xerox-scandal years back, that invalidated officially scanned documents. You already life in a world where "scanned before" means you can legally challenge the validity of a document.

unreal6 3mo ago
Is this the scandal to which you are referring? https://www.bbc.com/news/technology-23588202 https://www.bbc.com/news/technology-23588202
- 21asdffdsa12 3mo ago
  Yes, several national archives where affected on all settings. If a institution scanned and archived documents with these- these documents are - in theory challenge-able in court- within reason..
- mmh0000 3mo ago
  Yes, that is what they're referring to. The technical details behind it are kind of cool. It's a "bug" of the JBIG2 compression algorithm. See 1, 2 [1] https://community.apryse.com/t/jbig2-compression-issue/1814 https://community.apryse.com/t/jbig2-compression-issue/1814 [2] https://en.wikipedia.org/wiki/JBIG2#Character_substitution_errors_in_scanned_documents https://en.wikipedia.org/wiki/JBIG2#Character_substitution_e...
  - xattt 3mo ago
    All I see is 1, 1
- arthurmorgan 3mo ago
  There‘s also a great conference talk [0] on this, with an english translation [1]. [0]: https://youtu.be/7FeqF1-Z1g0 https://youtu.be/7FeqF1-Z1g0 [1]: https://youtu.be/zXXmhxbQ-hk https://youtu.be/zXXmhxbQ-hk
mindslight 3mo ago
I am pretty sure I was bitten by this bug, or a similar bug, on an IRA transfer form that I uploaded to a bank. On the scan that I uploaded, one digit in the account number was a clean "5", but the customer service person said it looked like a good clean "6". I presume their document management system was an outcropping of a system originally set up for dealing with faxes, and used that or a similar compression algorithm.
- 21asdffdsa12 3mo ago
  Thus - one could assume that all transfers stored by the bank- before the update would not hold up in court.
  - lotsofpulp 3mo ago
    One should assume the outcome of one's case in court is at least partially, if not heavily, influenced by one's political (influential-ness) standing.
    - inquirerGeneral 3mo ago
      [dead]
  - mindslight 3mo ago
    I don't know what you mean by "hold up in court". This seems to be an instance of software engineers making sweeping prognostications about legal matters. "Hold up in court" as in was it duly executed and filed by a certain date, regardless of the time and extra management required for the bank's performance? Yes, why not. "Hold up in court" against someone claiming it was an unauthorized transfer? Even without the bug, that can be contested in many straightforward ways! But unless there is a specific legal claim that hinges on the bug being significant, it's pretty irrelevant "in court". I knew about this precisely because the transfer failed to go through due to redundancy - one digit in account numbers is generally a check digit, plus account titles and whatnot. In general you can always challenge the validity of scanned documents, regardless of known software bugs or not! Just like you can always challenge that a paper document is a forgery. You need a specific argument and some evidence though!
ryukoposting 3mo ago
Hah, scan-dal.

Remember the xerox-scandal years back, that invalidated officially scanned documents. You already life in a world where "scanned before" means you can legally challenge the validity of a document.

unreal6 3mo ago

Is this the scandal to which you are referring? https://www.bbc.com/news/technology-23588202 https://www.bbc.com/news/technology-23588202

21asdffdsa12 3mo ago
Yes, several national archives where affected on all settings. If a institution scanned and archived documents with these- these documents are - in theory challenge-able in court- within reason..
mmh0000 3mo ago
Yes, that is what they're referring to. The technical details behind it are kind of cool. It's a "bug" of the JBIG2 compression algorithm. See 1, 2 [1] https://community.apryse.com/t/jbig2-compression-issue/1814 https://community.apryse.com/t/jbig2-compression-issue/1814 [2] https://en.wikipedia.org/wiki/JBIG2#Character_substitution_errors_in_scanned_documents https://en.wikipedia.org/wiki/JBIG2#Character_substitution_e...
- xattt 3mo ago
  All I see is 1, 1
arthurmorgan 3mo ago
There‘s also a great conference talk [0] on this, with an english translation [1]. [0]: https://youtu.be/7FeqF1-Z1g0 https://youtu.be/7FeqF1-Z1g0 [1]: https://youtu.be/zXXmhxbQ-hk https://youtu.be/zXXmhxbQ-hk

mindslight 3mo ago

I am pretty sure I was bitten by this bug, or a similar bug, on an IRA transfer form that I uploaded to a bank. On the scan that I uploaded, one digit in the account number was a clean "5", but the customer service person said it looked like a good clean "6". I presume their document management system was an outcropping of a system originally set up for dealing with faxes, and used that or a similar compression algorithm.

21asdffdsa12 3mo ago
Thus - one could assume that all transfers stored by the bank- before the update would not hold up in court.
- lotsofpulp 3mo ago
  One should assume the outcome of one's case in court is at least partially, if not heavily, influenced by one's political (influential-ness) standing.
  - inquirerGeneral 3mo ago
    [dead]
- mindslight 3mo ago
  I don't know what you mean by "hold up in court". This seems to be an instance of software engineers making sweeping prognostications about legal matters. "Hold up in court" as in was it duly executed and filed by a certain date, regardless of the time and extra management required for the bank's performance? Yes, why not. "Hold up in court" against someone claiming it was an unauthorized transfer? Even without the bug, that can be contested in many straightforward ways! But unless there is a specific legal claim that hinges on the bug being significant, it's pretty irrelevant "in court". I knew about this precisely because the transfer failed to go through due to redundancy - one digit in account numbers is generally a check digit, plus account titles and whatnot. In general you can always challenge the validity of scanned documents, regardless of known software bugs or not! Just like you can always challenge that a paper document is a forgery. You need a specific argument and some evidence though!

ryukoposting 3mo ago

Hah, scan-dal.