5 ms·

You're only considering one use case. There are others: > It is feared that wills of ordinary people, some of whom may become historically significant in the

by tivert 3y ago

You're only considering one use case. There are others: > It is feared that wills of ordinary people, some of whom may become historically significant in the future, risk being lost. > Wills are considered essential documents, particularly for social historians and genealogists, as they capture what people considered important at the time and reveal unknown family links. Digitization is not a problem, but the destruction of the originals is.

mytailorisrich 3y ago
I would think that for historical or genealogical research it actually makes no difference if the paper original is destroyed because what matters is the text: the assets the person had and how they wanted them dispersed, and the names of the people mentioned, maybe the wording.
- toyg 3y ago
  Only if you trust digitalisation to be a perfectly faithful process, and digital artefacts fundamentally unmodifiable or falsifiable. Which we all know is totally the case... oh wait
- simonw 3y ago
  From the article: > “You can see the indent of the pen and if the writer is excited or tense. There are minute details on the page which digitisation [can’t capture]. There is a thrilling sensation that you are looking at a document that a real human being wrote on. You get a connection to the past that digitised versions won’t give you.” More importantly: the digitization process is prone to errors and mistakes. Anyone who's spent any time browsing digital collections will have encountered pages that were scanned out of focus, or where a part of the page was illegibly crumpled, or a page was missed or miscategorized.
  - mytailorisrich 3y ago
    >There is a thrilling sensation that you are looking at a document that a real human being wrote on I think this is it. This is about emotions and sentimentality, not fundamental practical concerns. For the purpose of this discussion we of course assume that digitalisation is done correctly because this obviously has to be a comparison of the value of the original over a correct digital copy.
    - simonw 3y ago
      "You can see the indent of the pen and if the writer is excited or tense. There are minute details on the page which digitisation [can’t capture]". Plus my notes about digitization being error-prone.
      - mytailorisrich 3y ago
        This applies only if the will is handwritten by the person and one has to wonder how important an artifact the indent of the pen is (it does not matter at all for genealogical research and it's probably not that important for historical research). I'm hoping that there are stronger arguments.
        pmontra 3y ago
        My bet is that in 2124 we will be able to use our eyes to read the original paper documents but we won't be able to read the 2024 digital copies anymore, unless we keep migrating them to the next standard. Furthermore the servers (hw+sw) are not for free and all government projects use to cost more than market prices. And finally, why does storing paper documents on shelves cost 4.5 M per year?
        maxerickson 3y ago
        Computing environments haven't changed a whole lot in the last 25 years, and current environments aren't that much bother to emulate. The ongoing integrity of the storage seems like a much larger concern than displaying the documents.
        pmontra 3y ago
        I have the floppy disks of my first Slackware Linux installation. I'm not sure if I can still read them. I should have a Toshiba laptop from year 2000 with a floppy disk reader but will it work? Of course if I cared about those files (I don't) I should have copied them to a HDD, then to a SSD like I did with everything else. Think of archives on tapes and the plans to keep moving them across storage technology. That costs too and if you fail at that eventually all you have is an unreadable piece of plastic and silicon. Maybe you should save copies to at least two of the major could storages and let them do the job of keeping current. I won't destroy the original though.
  - helsinkiandrew 3y ago
    But having paper wills after 25 years in case one becomes significant costs £4.5M a year! I’m sure any historian could come up with much better uses for those resources. Issues with scanning can be done by having the scan done at probate time and the person handling the estate validating the scan.
    - mytailorisrich 3y ago
      Yes. And an additional point is that, since there is no time limit the number of wills kept on record increases continuously forever. They said they currently have 110 million and that they started only mid-19th century. I don't know what they'll ultimately decide but it is right that they at least bring the issue up to debate it.
    - halostatue 3y ago
      And they think that the digitization and digital storage (as well as regular updating) will be cheaper than that? I don't. I asked ChatGPT 3.5 for some rough estimates on the storage costs. Ideally, we would be using archival quality scans (~75MB per page of a will) and a more or less random distribution of will length resulted in an estimate of ~700 million pages for the archived wills, or 50,000 PB. This number would need to increase each year, but this would be the baseline for the reported 100M wills. If we assume $20/TB/month for archival quality distributed storage, the cost for the baseline storage is about $12B annually, and that's just for the storage. Indexing, retrieval, etc. adds up to 50% more per year on top of the cost to digitize and develop such a system in the first place. So we’re talking ~$18B or ~£14B annually. Depending on storage density and overall reliability, 25,000 – 50,000 racks would be required without data redundancy, so the $20/TB/month baseline cost or $30/TB/month for redundancy, indexing, and retrieval does not sound too far out of line. Are we really saying that the physical storage of wills is costing ~£20B annually? (It could be, but the likely highest cost as always is personnel. On the other hand, digital storage requires more climate control than document storage, which aside from humidity control, can mostly be kept underground without temperature controls.) The chat, for consideration. The costs could be constrained by not using archival quality scans, but that would be the purposeful destruction of history — which, given that we're talking about the tories, isn’t that surprising, really. https://chat.openai.com/share/edb569c0-30f9-45ea-8d9f-745631ef4533 https://chat.openai.com/share/edb569c0-30f9-45ea-8d9f-745631...
  - trebligdivad 3y ago
    You could ask, for new ones, that Executors verify the scans after the Probate office issues the certificate
kelnos 3y ago
The originals don't have to be destroyed; perhaps these private organizations who are so interested in preserving them can pony up the cash to store them. Why should taxpayers be footing this bill?

by tivert 3y ago

mytailorisrich 3y ago

I would think that for historical or genealogical research it actually makes no difference if the paper original is destroyed because what matters is the text: the assets the person had and how they wanted them dispersed, and the names of the people mentioned, maybe the wording.

toyg 3y ago
Only if you trust digitalisation to be a perfectly faithful process, and digital artefacts fundamentally unmodifiable or falsifiable. Which we all know is totally the case... oh wait
simonw 3y ago
From the article: > “You can see the indent of the pen and if the writer is excited or tense. There are minute details on the page which digitisation [can’t capture]. There is a thrilling sensation that you are looking at a document that a real human being wrote on. You get a connection to the past that digitised versions won’t give you.” More importantly: the digitization process is prone to errors and mistakes. Anyone who's spent any time browsing digital collections will have encountered pages that were scanned out of focus, or where a part of the page was illegibly crumpled, or a page was missed or miscategorized.
- mytailorisrich 3y ago
  >There is a thrilling sensation that you are looking at a document that a real human being wrote on I think this is it. This is about emotions and sentimentality, not fundamental practical concerns. For the purpose of this discussion we of course assume that digitalisation is done correctly because this obviously has to be a comparison of the value of the original over a correct digital copy.
  - simonw 3y ago
    "You can see the indent of the pen and if the writer is excited or tense. There are minute details on the page which digitisation [can’t capture]". Plus my notes about digitization being error-prone.
    - mytailorisrich 3y ago
      This applies only if the will is handwritten by the person and one has to wonder how important an artifact the indent of the pen is (it does not matter at all for genealogical research and it's probably not that important for historical research). I'm hoping that there are stronger arguments.
      - pmontra 3y ago
        My bet is that in 2124 we will be able to use our eyes to read the original paper documents but we won't be able to read the 2024 digital copies anymore, unless we keep migrating them to the next standard. Furthermore the servers (hw+sw) are not for free and all government projects use to cost more than market prices. And finally, why does storing paper documents on shelves cost 4.5 M per year?
        maxerickson 3y ago
        Computing environments haven't changed a whole lot in the last 25 years, and current environments aren't that much bother to emulate. The ongoing integrity of the storage seems like a much larger concern than displaying the documents.
        pmontra 3y ago
        I have the floppy disks of my first Slackware Linux installation. I'm not sure if I can still read them. I should have a Toshiba laptop from year 2000 with a floppy disk reader but will it work? Of course if I cared about those files (I don't) I should have copied them to a HDD, then to a SSD like I did with everything else. Think of archives on tapes and the plans to keep moving them across storage technology. That costs too and if you fail at that eventually all you have is an unreadable piece of plastic and silicon. Maybe you should save copies to at least two of the major could storages and let them do the job of keeping current. I won't destroy the original though.
- helsinkiandrew 3y ago
  But having paper wills after 25 years in case one becomes significant costs £4.5M a year! I’m sure any historian could come up with much better uses for those resources. Issues with scanning can be done by having the scan done at probate time and the person handling the estate validating the scan.
  - mytailorisrich 3y ago
    Yes. And an additional point is that, since there is no time limit the number of wills kept on record increases continuously forever. They said they currently have 110 million and that they started only mid-19th century. I don't know what they'll ultimately decide but it is right that they at least bring the issue up to debate it.
  - halostatue 3y ago
    And they think that the digitization and digital storage (as well as regular updating) will be cheaper than that? I don't. I asked ChatGPT 3.5 for some rough estimates on the storage costs. Ideally, we would be using archival quality scans (~75MB per page of a will) and a more or less random distribution of will length resulted in an estimate of ~700 million pages for the archived wills, or 50,000 PB. This number would need to increase each year, but this would be the baseline for the reported 100M wills. If we assume $20/TB/month for archival quality distributed storage, the cost for the baseline storage is about $12B annually, and that's just for the storage. Indexing, retrieval, etc. adds up to 50% more per year on top of the cost to digitize and develop such a system in the first place. So we’re talking ~$18B or ~£14B annually. Depending on storage density and overall reliability, 25,000 – 50,000 racks would be required without data redundancy, so the $20/TB/month baseline cost or $30/TB/month for redundancy, indexing, and retrieval does not sound too far out of line. Are we really saying that the physical storage of wills is costing ~£20B annually? (It could be, but the likely highest cost as always is personnel. On the other hand, digital storage requires more climate control than document storage, which aside from humidity control, can mostly be kept underground without temperature controls.) The chat, for consideration. The costs could be constrained by not using archival quality scans, but that would be the purposeful destruction of history — which, given that we're talking about the tories, isn’t that surprising, really. https://chat.openai.com/share/edb569c0-30f9-45ea-8d9f-745631ef4533 https://chat.openai.com/share/edb569c0-30f9-45ea-8d9f-745631...
- trebligdivad 3y ago
  You could ask, for new ones, that Executors verify the scans after the Probate office issues the certificate

kelnos 3y ago

The originals don't have to be destroyed; perhaps these private organizations who are so interested in preserving them can pony up the cash to store them. Why should taxpayers be footing this bill?