santiago,
@santiago@masto.lema.org avatar

Thanks U.S. Environmental Protection Agency for providing us with a PDF of “Novice Book - a user manual for the PDP 11/70 Computer”
https://nepis.epa.gov/Exe/ZyPDF.cgi/94000K6G.PDF?Dockey=94000K6G.PDF

santiago,
@santiago@masto.lema.org avatar

I always wondered what was the correct way of organizing old files.

Karen, Lisa, are you still around ?

adora,

@santiago haha I love it.

Text

ETIQUETTE FOR COMPUTER USERS

  1. Keep your directory clean of extraneous files (old versions, already processed data, out-of-date command files, etc.). If you have files that you
    would like to keep, but do not need to access frequently, talk to either Karen or Lisa in Operations about archiving it.
santiago,
@santiago@masto.lema.org avatar

@adora Thanks for the transcript!

I always wondered if it was necessary to make alt text for text screenshots ? I imagine people with sight issues would always have software that can read text in images. Isn’t that builtin in iOS / Android ? Or is it tedious compared to a proper alt text ?

adora,

@santiago I have a few blind followers who might take the time to explain, but basically:

  1. ocr isn't very good/reliable

  2. the text inside the image isn't always the important part - frequently you want other things described. By writing down just the text in the image, I'm also implicitly saying that nothing else in the image had any particular value to know.

  3. automatically trying to extract text from images and reading it requires support from whatever application you're using, and most stuff isn't designed well for blind people.

  4. when ocr fails or is run on something that has no text, it generates a ton of garbage which a screen reader would then read out for everrr

  5. it's overall just an unpleasant experience, and I love my followers and want them to be happy

santiago,
@santiago@masto.lema.org avatar

@adora Thanks ! That’s a good summary. I’ll definitely enhance future posts in that light.

adora,

@santiago you are the best:)

Another good insight on why some incorrect characters from an OCR feel bad - the subject is different but the consequences are the same
https://nojack.easydns.ca/@tantramar/110679791797196162

adora,

@santiago just as an example by the way, I created that transcript by using an OCR and then going back and manually correcting a ton of errors 😅😭😅

santiago,
@santiago@masto.lema.org avatar

@adora It would be great if native clients offered the option to quickly OCR as alt text so we could just tap on a button and then fix eventual errors for these cases.

iOS definitely has all the necessary APIs to make this and I don’t doubt Android has it too. Making this effortless will surely increase the quantity of alt text (for these cases).
https://betterprogramming.pub/enabling-live-text-interactions-with-images-in-swiftui-5dd1d7f1676

//cc @tootapp @ivory @IceCubesApp @MonaApp

adora,
bgrinter,
@bgrinter@mastodon.sdf.org avatar

@santiago handy to know 😉

nbb,

@santiago just like a Karen to be policing people’s files… MY DIRECTORY MY CHOICE!

  • All
  • Subscribed
  • Moderated
  • Favorites
  • random
  • DreamBathrooms
  • mdbf
  • ethstaker
  • magazineikmin
  • GTA5RPClips
  • rosin
  • thenastyranch
  • Youngstown
  • osvaldo12
  • slotface
  • khanakhh
  • kavyap
  • InstantRegret
  • Durango
  • provamag3
  • everett
  • cisconetworking
  • Leos
  • normalnudes
  • cubers
  • modclub
  • ngwrru68w68
  • tacticalgear
  • megavids
  • anitta
  • tester
  • JUstTest
  • lostlight
  • All magazines