5 ms·
The big second-hand bookseller in my town has a huge collection and several locations in multiple metro areas. Yet they have no way to search for a specific tit
by sbuccini 6y ago
The big second-hand bookseller in my town has a huge collection and several locations in multiple metro areas. Yet they have no way to search for a specific title!
I was shocked. If you have no idea of your current inventory and sales history, how do you know how much to pay for people's books? Their books are about $0.50 a title -- I would gladly pay that price for a book I would otherwise get from the library if I knew it was available. They would be able to do a serious business selling required titles to local college students. And they could have a robust online presence through 3rd party sellers like Amazon and Ebay.
Yes, some beloved used bookstores are falling victim to "market forces". But if those same market forces inject some modernity into these businesses that badly need it, I'd be very grateful.
- throwawaysea 6y agoThere’s https://bookshop.org https://bookshop.org for injecting modernity into supporting independent bookstores. It doesn’t help specifically for used book stores however.
- gxqoz 6y agoI find the bookshop.org model a bit weird. There's no real reason that independent bookshops need to be part of it, apart from a marketing angle to court people who want to "help local bookstores" over Amazon.com. From a recent NYT article on the site: "Orders are fulfilled through Ingram, a large book distributor, and mailed directly to customers, so stores don’t have to have the books in stock or process inventory. Bookstores get 30 percent of the list price — less than they would typically make from a direct sale — but don’t have to pay for inventory or shipping." https://www.nytimes.com/2020/06/16/books/bookshop-bookstores-coronavirus.html https://www.nytimes.com/2020/06/16/books/bookshop-bookstores...
- throwawaysea 6y agoYea I don't fully understand it either, to be honest. Also it's interesting because in a recent HN discussion about Jeff Bezos courting early investors, the second employee of Amazon showed up in the comment section (https://news.ycombinator.com/item?id=24245909 https://news.ycombinator.com/item?id=24245909) and he mentions Ingram: > I remember that not long after we "opened the doors" to the public, we had a visit from some reps from either Ingram or Baker & Taylor, the US's two biggest book distributors (I forget which). Part of the reason Jeff had picked Seattle was that it was within the 1 day delivery radius for both companies. These guys came over to see what we were doing and they were completely flabbergasted. They could not believe that a few people in a small commercial building in Seattle had built what we had already done by that point. They had no idea of the technologies involved, they had no grasp of the vision. But we never had to convince companies like this - we just ordered books from them, as their customers, and then sold them to ours. In some ways it's like Ingram is striking back at Amazon, and sharing part of the proceeds with independent bookstores. My current feeling is that it has its place. I don't want Amazon to accumulate even more control and kill the diversity of businesses that make up our society.
- contingencies 6y agoAll books in the last ~50 years have a barcode already printed on them, and an ISBN printed inside. Any phone camera can read these. If you assume 15 seconds per book at pace, 40 books per shelf, 6 shelves per bookcase, and 30 bookcases in the average store, complete digitization of a collection should take only ... Edit: Responding comment is quite right, original calculation mistaken. It's 4AM here. 40 × 6 × 30 = 7200 books × 15 seconds ÷ 60 ÷ 60 = 30 hours, so <4 days assuming 8 hour days. Or 1 day if you can speed up to 5 seconds per book.
- klelatti 6y agoI get 7200 books and 30 hours. Am I missing a factor somewhere?
- bjo590 6y agoAssuming ~600 man hours (your numbers) to digitize an entire used book store, at $15/hr cost for labor it would cost ~$900 to digitize a bookstore. I think you're shy a factor of 3 or 4 on the amount of labor it would take to digitize a book store, given my experience with running inventory at retail shops. Maybe it could be worth it if there was an inventory management system that could automatically post the books for sale on an internet platform, but if every used book store posted their entire inventory online the market for many titles would be flooded and it would be a race to the bottom. Ultimately all physical retails spaces have the same common problem -- why should someone go into your store when they can get the same products online. This problem is even bigger during covid-19. I personally think there are two answer, one is to move the core retail business online, and the other is to create experiences people will keep coming back for. If you're making unique products it might be best to pivot online. If you're into the classic buy wholesale sell retail business than the online market can be very competitive. A used book store can be a great place to have experiences. Authors can come and talk, you can have children story time, book clubs can meet, coffee shops pair well with book stores. You have to clear inventory to make space for experiences, but you can use it as an opportunity to remove inventory that wasn't selling anyway.
- ninjinxo 6y ago
- xaqfox 6y agoThis is similar to the use case that OPDS is marketing itself with. It was featured on HN a week ago: https://news.ycombinator.com/item?id=24178548 https://news.ycombinator.com/item?id=24178548 https://opds.io https://opds.io