Thank you for your re-review.
I'm glad you are able to find the resource you were looking for, though saddened to hear of your continued disappointment with Forgotten Books. Please do reach out to csupport@forgottenbooks.org if you feel there's anything we can do to help alleviate this.
I will note that we are a for-profit enterprise - we do rely on the revenue generated from our subscriptions and direct purchases in order to continue running the company and improve our restoration software, so this is why you see some content is restricted when browsing without a subscription. We believe this to be competitively priced at $8.99/month
Original Response:
I can understand why this might at first glance appear to be the case and we appreciate you reaching out with your concern, however I'm happy to assure you that there is actually far more to the process than simply finding book scans and putting our own stamp on them!
Forgotten Books uses in-house developed restoration software - an ongoing project we're always building on and improving - to make some pretty significant and useful changes to the quality of our books and ebook files before we publish.
Some highlights:
Optical character recognition;
This allows the content of the book to be searchable by a computer - otherwise you'd have to search through every one of our 1,200,000+ titles manually in order to find books with the information you need! (This is actually incredibly difficult on these aged texts where the font it was printed in was archaic and the ink application was inconsistent due to older printing technology, not to mention other issues such as the yellowing of aged paper and the dirt and marks that may have accumulated on the page prior to scanning. We're still putting a lot of work in to improve this, particularly with the new technologies in AI becoming more accessible).
Picture detection;
Due to the age of the paper and dirt or "noise" as mentioned previously, we have to know where any pictures are on any and every page. The steps taken to clean up an image (which can be drawn in greyscale) are very different to the steps taken in order to clean up text (for which we're only interested in the pure black and pure white pixels on the scan). If we don't accurately detect where the images are, they can end up looking really poor after we remove the background noise. While a human can easily see the difference between a picture and a word, for an automatic system it's very difficult, even giants such as Google and Amazon can't do this perfectly yet (automation is essential - imagine the hours needed for a person to check 1.2 million, 300 page books! We couldn't publish many books this way) Again, this is something we're actively investing in improving.
Damage detection;
Many books, when scanned, include damage to text and pages (for example graffiti that makes the text illegible or a rip on the page). These, of course, are very frustrating to encounter when you're halfway through a book. We put a lot of work into detecting and either fixing or, if all else fails, removing these books from publication. Sometimes errors do slip through - we're still working on improving this feature too (you can see a theme emerging!) but the majority of the time, if a customer finds damage in a book they have purchased, we can fix it and send them a new copy.
There are far more features we add and improve that I could list, however they start becoming more technical and less exciting - things like rotation correction and metadata generation so that the books are discoverable on marketplaces like Amazon or library systems.
I hope this provides some reassurance - our team are always working hard to do our bit and provide at least some small amount of good in the world. Reviews really do effect our ability to do that so if you feel this information changes the star rating you're happy to give us, that would be greatly appreciated.
With regards,
Forgotten Books Management