196 karma · joined January 11, 2011
http://www.luminarygraphics.com/ http://outofprintclothing.com/ http://bagladiestea.com/novel-tea.html http://www.us.penguingroup.com/pages/shop/the_penguin_collec...
I've sold over 8,000 posters at an average of ~$20 each since I opened the store in September 2011 . I left my job in August to focus on Litographs full-time.
I wrote python scripts to create the posters and t-shirts and to automate most tasks around updating and keeping up my Shopify storefront (www.litographs.com). Releasing two designs each week means that automating what I can is key.
Before I started, I'd gotten about halfway through the introductory MIT OCW course. Through working on Litographs, I've used PIL, reportlab, numpy, and Scrapy, and I'm gearing up to release a customization tool using picloud and Flask. It's been a crash course in finding the right tools for the job, and learning just enough to get done what I need.
Some of the interesting challenges have been outside of software: rolling and shipping hundreds of posters a week (this was especially tricky for the 6 months I was living in Chile), sourcing artists and artwork, licensing contemporary books, comparing and choosing between printing processes, building up a large mailing list (1,500+ members) with no marketing budget.
I've mainly been a lurker on HN, but I owe a great deal of gratitude to everyone here for helping me take the leap. I'll be blogging about some of these challenges in more detail over the course of the Kickstarter campaign, and I'm happy to answer any questions here.
I do like to give something back, and usually end up buying a bottle of wine or a gift card in return. There could be some interesting opportunities for selling gifts (pet-related or not) in addition to processing cash payments.
As much as I'm sure it improves the listings, I'd recommend not making the photo required. First of all, other non-required info (such as pet type) seems more important, and second of all it requires digging through my photos. I'd be likely to add a photo when I'm looking for someone to catsit, but it was a deterrent for me to add the cat in the first place.
My question is - how far back would you go?
Not sure how common it will be for a user to add dozens of recipes at once, but I'm finding it tiring to go back and copy in all of the instructions. For a user like me, the awesome thing about YumTab is having all of my recipes in one place, so copying over the instructions is essential.
Is there any way to make it easier to add info manually, at least while you work on scraping the instructions? For now, I'm having to run the bookmarklet, refresh My Recipes on YumTab, click on the newly added recipe, click edit, go back to the original recipe, copy the instructions, go back to YumTab, and paste it.
For now, allowing manual input within the bookmarklet could help.
I've had a few issues with images (both URL and upload), but other than that this is by far the least buggy online recipe box I've used.
One suggestion - enable some sort of tagging/categorization system to help with planning meals.
Thanks for your work!