I read a book with printed links in it, but I couldn’t click on them 🤷🏻‍♂️

So I made an alpha version of the 📖👈🏻 Links Detector app that makes printed links clickable.

Check it out 👉🏻 https://trekhleb.github.io/links-de...

awesome! what mobile browser does it work on?

people don't usually put links in books, they put QR code if necessary

Based kn the fact how many emojis you have used, you’d be the kind who writes an medium article about this

🐏🐏🐏🐏🐏🐏🐏

@theabbie it seems u don't really read much

@coffeeholic Even if they do, that's bad practice, atmost, they are shortened URLs, and the best way is to use QR codes.

@coffeeholic Even the text detection of the Vista era would do a better job, that at least didn’t crash.

@theabbie puting shortend url into a book? r u out of ur mind?
1. tracking readers
2. their lifespan migth be shorter than actual url
3. hides readbility
4. hides the REAL source they are referring

@coffeeholic shortened URLs have larger lifespan, also, they are more useful because even if the original URLs change, the link can stay valid, you just have to change the URL it's pointing to. Readability can be achieved by describing what the link is for rather than having a long URL, use self-hosted URL shortener if they care about privacy. Hiding original source is not a point because it's in a book, must be reliable source.

@trekhleb I think it's awesome that you made this -- people don't seem to understand "alpha", and you are already experiencing disgruntled customers who demand you keep supporting this.

I think for those people, there's Google Lens.

If your app can become an open alternative, that would be awesome.

Tf is that pic lmao

@theabbie in a plague-ridden world full of uncertainties there's a certain beauty to knowing that whenever you open a thread on devrant, you (clarification: as in you, theabbie) will have been there, shat all over it with your entitled dick "logic" and left everyone in a sour mood. I thank you from the bottom of my heart for your non-contributions to the platform.

@Elyz Thank you so much for your constructive feedback, unfortunately, I didn't find anything constructive in your opinion. If you don't have anything constructive, don't say it, also, I think you have lost any privilege of giving opinions to me. Feel free to Refrain replying to me. Thanks, have a nice day.

@Elyz too bad I cannot fav a comment. this. love it.

@Elyz please don’t touch Abbie, he’s our mascot. Yes he’s a total cringe and unbelievably annoying but that’s who he is. There is no need to shame him

@kiki wouldn't touch that with a 10 meter pole, don't worry.

@kiki it’s like having a retarded mole as a mascot that thinks it’s a raccoon. It isn’t cute, it’s just really sad and kind of pathetic.

@Root I agree with you, but we still can at least keep it all civilized

@kiki Depending on your definition, devRant may have been more civilized before 😉

@coffeeholic it should be Safari on iOS or Chrome on Android. Also if it is opened in a WebView there is also might be some issues.

@theabbie I've actually seen many links in modern books (about ML for example)

@trekhleb The project is useful, I agree to that, I was just saying why putting long URLs in printed books is a bad practice.

@trekhleb You could argue that if you read books about ML, the chances of being read as an e-book would be quite high

@bittersweet thanks! Yeah, Google Lens is way to go. The Links Detector app is indeed in alpha stage and has many issues so far. The one thing that this app gave me that Google Lens couldn't is all the fun creating the model and UI for it :D

@trekhleb That's what I'm talking about!

Sometimes you don't make an app because it must become a billion dollar startup, you make it just because you want to learn stuff.

You share it with the world because others might find it interesting, educational or useful -- and that's awesome.

@Root I think devRant used to be less civilized towards the outside world.

Sadly, it's now less civilized internally, with people ranting more about each other's rants than about their own lives or jobs.

looks amazing man. just a few points.

i wasnt able to get it to read any urls. (i tried a url in a browser address bar, and a url without https://. not sure if you are searching for a prefix or something.

And it seems it's refreshing the camera every few seconds or maybe searching for links? it's just a bit laggy.

but the fact that it's a web app is fuking awesome! the idea itself is really dope.

@codecrow yes you're right the app searches specifically for the prefix https:// and if it doesn't see it it won't recognize the link. Also, the dataset that I used for the training consisted only of the images of just one book, with white background and dark text, so the model is very immature now (thus there is an "Alpha" prefix on the app logo and the link to the Issues in the app, just to show that this is just a first experimental launch). But thanks for checking it anyways!

> Even if they do, that's bad practice, atmost, they are shortened URLs, and the best way is to use QR codes.

@theabbie afdsghajrgasrtgsadgahaeh holy crap kid srsly? What kind of brainf- actually never mind. Just know this: best practices != what is actually being done. EVER.

@trekhleb what did you use for training your network? Or did you take a pretrained one? Asking for my own future reference.
It seems dope, btw!