4 ms·

Tape as backups are arguably still valuable, but any important information being passed by Excel is a travesty. I've worked on the quant side of things and trad

Tape as backups are arguably still valuable, but any important information being passed by Excel is a travesty. I've worked on the quant side of things and trading securities has standard protocols (primarily FIX, but more esoteric protocols are used by some MM's). I've worked with insurance conglomerates - health information has HL7 for EHR. Purchase orders between Walmart and their suppliers use EDI. All of these aren't 'magical' but they sure are global standards which are transparent and relatively perfect (you have XML schemas to validate against). Those are just the industries I've worked in professionally (I'm sure any other industry - oil, heavy machinery use something similar.)

dvcc 10y ago
A working vendor-compatible HL7 implementation is certainly magical.
collyw 10y ago
In my last job I ended up wasting around half of my time fixing Excel / user errors (despite complaining to management multiple times about how error prone it was).
thwarted 10y ago
any important information being passed by Excel is a travesty I think that's putting it mildly. A major issue with using Excel files for data transfer is, to put it bluntly, the lackadaisical way people treat Excel files. They are passed around and edited. Passed back once edited. Passing them around creates copies of them. They get goofy names as people try to manage the multiple editors and editions. There's no verifiable authority for the source or latest version of the data. None of the data or its state is externally verifiable. We've all seen where this goes south very quickly; we even have a meme for it: the FINAL FINAL FINAL suffix that gets appended to the filename. I suspect this is why/how zombie debt, as described by Oliver, gets resurrected: data quality issues: debt that is accidentally (or on purpose, to make another/recover a quick buck) sold again when it shouldn't have been. And the original owner of the debt, once sold, isn't interested in keeping track of it (they've already written it off). Some of the issue is that there are multiple parties involved. Real estate has title records as one way to solve this, which isn't necessarily a final solution (thus the existence of title insurance), but without even something resembling a centralized and agreed upon record of the history of the debt changing hands, it's impossible to reliably know anything about it. Excel files exacerbate that. Further down in these comments, AstroJetson says that to forgive debt, they "they just delete the Excel file it's in". I find this to be a reasonable possibility for how this is handled. But deleting a file received from some other party doesn't make the file disappear. It's perhaps still in an attachment somewhere, in the sent mail folder, on the computer of the sender (perhaps moved to a "Sold Debt" folder). None of this is necessarily solved by transfer methods like EDI or HL7, except for what amounts to a barrier to entry to access the data and make it available to begin with. EDI is a solution for purchase orders between Walmart and suppliers is because 1) there's often a limited number of parties (two) involved with the data, 2) Walmart is always one of the parties (so it is effectively centralized), 3) the data itself is verifiable against other, trusted sources like accounts payable and accounts receivable. It's also going into a system that has automation components. The people who are buying debt information and receiving it via Excel spreadsheets are just going to go over it by hand and call/contact people on it. It's going into some massive computer based process and generate reports for dozens of different departments like the data being transferred via EDI at Walmart is. This is where someone mentions using a blockchain or git to track the way the debt changes hands and its current state. This would be a fine suggestion if it wasn't in the interest of the debt buyers/sellers to keep the system shitty with little visibility/oversight so they can take advantage of debtors and other debt buyers any way they can.
- AstroJetson 10y ago
  I checked with the company, and they do track the debt they "forget". They also send a letter to the debtor to say that the debt has been cleared. That takes care of the case where it comes back to life, the debtor has some level of proof that the debt was taken care of. I didn't ask if they tracked all of it in an Excel file ;-)
  - thwarted 10y ago
    Figured it was something like this. This is great news if you're trying to get rid of bill collectors. Just make some "proof" on "company letterhead" that the debt has been cleared. It's no more scummy than harassing collectors.
    - eonw 10y ago
      sounds like a business model almost as slimy as those collecting expired debt, but at least if i was selling forged documents i could sleep at night.
      - thwarted 10y ago
        No need for a slimy business model, the debtor can generate these documents themselves.
bko 10y ago
I agree a standard protocol would be beneficial to everyone and is probably why those conferences are so useful. The problem is the barrier to entry is so low (for better or for worse) and the absolute amounts traded are so low (often fractions of a cent to the dollar) that the cost of implementing and following a protocol are probably high enough that its not worth it.

dvcc 10y ago

A working vendor-compatible HL7 implementation is certainly magical.

collyw 10y ago

In my last job I ended up wasting around half of my time fixing Excel / user errors (despite complaining to management multiple times about how error prone it was).

thwarted 10y ago

any important information being passed by Excel is a travesty I think that's putting it mildly. A major issue with using Excel files for data transfer is, to put it bluntly, the lackadaisical way people treat Excel files. They are passed around and edited. Passed back once edited. Passing them around creates copies of them. They get goofy names as people try to manage the multiple editors and editions. There's no verifiable authority for the source or latest version of the data. None of the data or its state is externally verifiable. We've all seen where this goes south very quickly; we even have a meme for it: the FINAL FINAL FINAL suffix that gets appended to the filename. I suspect this is why/how zombie debt, as described by Oliver, gets resurrected: data quality issues: debt that is accidentally (or on purpose, to make another/recover a quick buck) sold again when it shouldn't have been. And the original owner of the debt, once sold, isn't interested in keeping track of it (they've already written it off). Some of the issue is that there are multiple parties involved. Real estate has title records as one way to solve this, which isn't necessarily a final solution (thus the existence of title insurance), but without even something resembling a centralized and agreed upon record of the history of the debt changing hands, it's impossible to reliably know anything about it. Excel files exacerbate that. Further down in these comments, AstroJetson says that to forgive debt, they "they just delete the Excel file it's in". I find this to be a reasonable possibility for how this is handled. But deleting a file received from some other party doesn't make the file disappear. It's perhaps still in an attachment somewhere, in the sent mail folder, on the computer of the sender (perhaps moved to a "Sold Debt" folder). None of this is necessarily solved by transfer methods like EDI or HL7, except for what amounts to a barrier to entry to access the data and make it available to begin with. EDI is a solution for purchase orders between Walmart and suppliers is because 1) there's often a limited number of parties (two) involved with the data, 2) Walmart is always one of the parties (so it is effectively centralized), 3) the data itself is verifiable against other, trusted sources like accounts payable and accounts receivable. It's also going into a system that has automation components. The people who are buying debt information and receiving it via Excel spreadsheets are just going to go over it by hand and call/contact people on it. It's going into some massive computer based process and generate reports for dozens of different departments like the data being transferred via EDI at Walmart is. This is where someone mentions using a blockchain or git to track the way the debt changes hands and its current state. This would be a fine suggestion if it wasn't in the interest of the debt buyers/sellers to keep the system shitty with little visibility/oversight so they can take advantage of debtors and other debt buyers any way they can.

AstroJetson 10y ago
I checked with the company, and they do track the debt they "forget". They also send a letter to the debtor to say that the debt has been cleared. That takes care of the case where it comes back to life, the debtor has some level of proof that the debt was taken care of. I didn't ask if they tracked all of it in an Excel file ;-)
- thwarted 10y ago
  Figured it was something like this. This is great news if you're trying to get rid of bill collectors. Just make some "proof" on "company letterhead" that the debt has been cleared. It's no more scummy than harassing collectors.
  - eonw 10y ago
    sounds like a business model almost as slimy as those collecting expired debt, but at least if i was selling forged documents i could sleep at night.
    - thwarted 10y ago
      No need for a slimy business model, the debtor can generate these documents themselves.

bko 10y ago

I agree a standard protocol would be beneficial to everyone and is probably why those conferences are so useful. The problem is the barrier to entry is so low (for better or for worse) and the absolute amounts traded are so low (often fractions of a cent to the dollar) that the cost of implementing and following a protocol are probably high enough that its not worth it.