Most active commenters
  • carlosjobim(4)
  • contact9879(3)

←back to thread

Rubio stages font coup: Times New Roman ousts Calibri

(www.reuters.com)
386 points italophil | 15 comments | 10 Dec 25 00:08 UTC | HN request time: 0s | source | bottom
Show context
WhyOhWhyQ ◴[10 Dec 25 22:35 UTC] No.46224941[source]▶
Is Calibri actually more accessible? Every step of this story seems pointless and fake.
replies(5): >>46224996 #>>46225153 #>>46225174 #>>46225199 #>>46225226 #
1. papercrane ◴[10 Dec 25 22:41 UTC] No.46224996[source]▶
One of the reasons Calibri was selected over Times New Roman was it has a lower rate of OCR transcription errors, making documents using it easier for people using screen readers.
replies(2): >>46225285 #>>46227403 #
2. blueflow ◴[10 Dec 25 23:05 UTC] No.46225285[source]▶
Link on that, as OCR should be more reliable with Times New Roman due to significant serifs.
replies(3): >>46225527 #>>46226162 #>>46227942 #
3. papercrane ◴[10 Dec 25 23:28 UTC] No.46225527[source]▶
The memo at the time said the serifs can cause OCR issues.

https://x.com/John_Hudson/status/1615486871571935232

replies(1): >>46226222 #
4. orwin ◴[11 Dec 25 00:38 UTC] No.46226162[source]▶
I don't have link on that, but the main difficulty with OCR isn't the OCR part (not anymore at least), it's the "clean up" part, and serifs are a pain in the ass, especially on sightly crumpled paper. My use case was an ERP plugin that digitalized and read to receipt to autofill reimbursement demands, and since most receipt use sans-serif fonts, it was mostly fine, but some jokers use serifed font (mostly on receipts you get when using cash, not credit card receipts) and the error rate jumped from like 1% to 13% (not sure about the 1%, it might be a story i told myself to make me feel better, it was a decade ago, before i pivoted to network from AI. I always take the best decision it seems)
5. opo ◴[11 Dec 25 00:47 UTC] No.46226222{3}[source]▶
Just because they claimed it, doesn't make it true. OCR and screen reader software in 2023 did not have problems with serifs.
replies(1): >>46226543 #
6. ◴[11 Dec 25 01:26 UTC] No.46226543{4}[source]▶
7. carlosjobim ◴[11 Dec 25 03:43 UTC] No.46227403[source]▶
That doesn't make much sense, since a typewriter will neither type Calibri nor Times New Roman. And OCR should only be needed for type written documents, because any document made with Calibri or TNR is already digital.
replies(2): >>46227546 #>>46228615 #
8. contact9879 ◴[11 Dec 25 04:08 UTC] No.46227546[source]▶
printed documents, images, horribly inaccessible pdfs, horribly inaccessible websites
replies(1): >>46227862 #
9. carlosjobim ◴[11 Dec 25 05:01 UTC] No.46227862{3}[source]▶
> Printed documents - Use the original, which is digital.

> Images - Use the original, which is digital.

> horribly inaccessible pdfs - Use the original, which has real text in the PDF

> horribly inaccessible websites - All text on any web site is digital. Nobody uses OCR on a website.

A massive paper producer like the government shouldn't adopt their type setting to people who are using technology wrongly.

replies(2): >>46227921 #>>46227938 #
10. contact9879 ◴[11 Dec 25 05:14 UTC] No.46227921{4}[source]▶
an example from today (pdf warning): https://www.ntsb.gov/news/Documents/National%20Defense%20Aut...
replies(1): >>46233382 #
11. contact9879 ◴[11 Dec 25 05:18 UTC] No.46227938{4}[source]▶
it's easier to mandate font than to excise all processes within the fed bureaucracy that result in these.

images being digital have no bearing on OCR ability

replies(1): >>46233409 #
12. nerevarthelame ◴[11 Dec 25 05:19 UTC] No.46227942[source]▶
I don't know what studies Blinken's State Department considered, but here are 2 studies on the matter.

https://www.academia.edu/72263493/Effect_of_Typeface_Design_...: "For Latin, it was observed that individual letters with serif cause misclassification on (b,h), (u,n), (o,n), (o,u)."

https://par.nsf.gov/servlets/purl/10220037: [Figure 5 shows higher accuracy for the two sans-serif fonts, Arial and DejaVu compared to Times New Roman, across all OCR engines]

13. funnybeam ◴[11 Dec 25 07:24 UTC] No.46228615[source]▶
We have a process at work where clients export information from their database as a pdf which they email to us so that we can ocr it and insert into our database.

No one else seems to think this is bat shit insane

14. carlosjobim ◴[11 Dec 25 16:24 UTC] No.46233382{5}[source]▶
God damn...

Why didn't they fax it back and forth a few times as well, just for good measure?

15. carlosjobim ◴[11 Dec 25 16:25 UTC] No.46233409{5}[source]▶
Images: use the original, which is a digital text document and not an image.

Unless they are making documents on typewriters. And in those cases neither Biden or Trump font is an option.