3 ms·

I've been working on some tools that integrate with USPTO (both from the application side and the validation side) for quite a few years now and they've been ma

I've been working on some tools that integrate with USPTO (both from the application side and the validation side) for quite a few years now and they've been making a TON of formatting changes recently. A lot of their PDF forms have changed, they're requiring XML versions of all data we submit, they're handling classifications differently, etc. Their process always felt like it was stuck in the past and being handled manually by humans before and now it feels like they're moving everything toward automated intake and initial reviews. I imagine this change is for the same reasons, and that's a hefty fee to force it. This also likely means I will need to rework our systems to spit out docx instead of/in addition to PDFs, which will be a nightmare to do. So that's fun.

peteradio 4y ago
I'm a masochist in need of work.
MrLeap 4y ago
The consolation is that, if I remember correctly, docx is just a zip file containing xml. I made an xlsx exporter in actionscript3 (lol) years ago and it worked like this. What I ultimately did was made a "template" document, and my code just injected strings into key spots, zipped it up in memory and gave it to you as a file.xlsx. Probably took me 3 days? I didn't have the benefit of libraries so I imagine this is significantly easier in less hobbled environments, nodejs or whatever probably has a kitchen sink package to do it.
- lofatdairy 4y ago
  That's exactly right. There are definitely nodejs docx templating packages (I've worked on codebases that used them in the past), but they're certainly not required provided your documents are reasonably simple. If anything, generating a pdf from various input files/structured text has been a much harder task. We generated docx files to allow for easy modification by non-technical staff, but to generate a pdf we had to use a headless instance of libreoffice since pandoc was struggling with the rendering.
barelycompetent 4y ago
Your work sounds very interesting. I work in the space (IPR's mostly). If you have more information do share some links about your projects!
thayne 4y ago
But why choose a (effectively) proprietary format?
- hn_go_brrrrr 4y ago
  ISO/IEC 29500 is not a proprietary format.
  - notagoodidea 4y ago
    It is for the .doc format and other Office firmat until Microsoft Office 2008 included and do not provide a standard for the the post-2008 Office format (ending with the `x` in the file extension), AFAIK.
    - formerly_proven 4y ago
      You're mistaken, ISO 29500 and ECMA-376 are both (they're actually three somewhat different standards all in all, as I understand it) for "Office Open XML" (totally not intended to be confused with Open Office and Open Office XML), which is docx/pptx/xlsx etc. ODT (OpenDocument) is also XML-in-a-ZIP and is also an ISO standard (older than Microsoft's, in fact). I don't think the older variants (.doc, .xls etc.) are standardized, but iirc Microsoft did eventually release some documentation on them.
  - kuon 4y ago
    ISO is not free which is very cumbersome and why many ISO standards are not implemented correctly (like 8601). I consider ISO proprietary and bad for that reason.
    - thayne 4y ago
      And what's more, the docx ISO "standardization" was very controversial.

peteradio 4y ago

I'm a masochist in need of work.

MrLeap 4y ago

The consolation is that, if I remember correctly, docx is just a zip file containing xml. I made an xlsx exporter in actionscript3 (lol) years ago and it worked like this. What I ultimately did was made a "template" document, and my code just injected strings into key spots, zipped it up in memory and gave it to you as a file.xlsx. Probably took me 3 days? I didn't have the benefit of libraries so I imagine this is significantly easier in less hobbled environments, nodejs or whatever probably has a kitchen sink package to do it.

lofatdairy 4y ago
That's exactly right. There are definitely nodejs docx templating packages (I've worked on codebases that used them in the past), but they're certainly not required provided your documents are reasonably simple. If anything, generating a pdf from various input files/structured text has been a much harder task. We generated docx files to allow for easy modification by non-technical staff, but to generate a pdf we had to use a headless instance of libreoffice since pandoc was struggling with the rendering.

barelycompetent 4y ago

Your work sounds very interesting. I work in the space (IPR's mostly). If you have more information do share some links about your projects!

thayne 4y ago

But why choose a (effectively) proprietary format?

hn_go_brrrrr 4y ago
ISO/IEC 29500 is not a proprietary format.
- notagoodidea 4y ago
  It is for the .doc format and other Office firmat until Microsoft Office 2008 included and do not provide a standard for the the post-2008 Office format (ending with the `x` in the file extension), AFAIK.
  - formerly_proven 4y ago
    You're mistaken, ISO 29500 and ECMA-376 are both (they're actually three somewhat different standards all in all, as I understand it) for "Office Open XML" (totally not intended to be confused with Open Office and Open Office XML), which is docx/pptx/xlsx etc. ODT (OpenDocument) is also XML-in-a-ZIP and is also an ISO standard (older than Microsoft's, in fact). I don't think the older variants (.doc, .xls etc.) are standardized, but iirc Microsoft did eventually release some documentation on them.
- kuon 4y ago
  ISO is not free which is very cumbersome and why many ISO standards are not implemented correctly (like 8601). I consider ISO proprietary and bad for that reason.
  - thayne 4y ago
    And what's more, the docx ISO "standardization" was very controversial.