Show HN: Cinc – GitHub for recipes
cinc.kitchen
cinc.kitchen
One note I'd make, though, is that US, UK and Australian cups are all different sizes. (and possibly others?)
And from the one recipe I saw it looks like you're using UK cups? But, possibly not as well... Either way, it might be worth calling that out, or allowing for cup-type specification similar to how you do for weights?
Somewhat related, some sites call out egg weights too, and that might be something to try and do here too? (Because, if nothing else, egg weights are often written on the carton which acts as a guide; even if people don't actually weigh them.)
Another thought, again... Some kind of showcasing? A la Github's explore, or even "awesome" pages? I might be off, but to me discovery is really important for recipes and I love browsing "cookies" or "desserts in a hurry" or similar; and pages akin to that with community curation could be nice?
Hmm, another random idea... Some kind of more granular forking to facilitate things like "do you have a stand mixer?" and that kind of distinction? (Coupled with "equipment switches" maybe?) In the past I've done recipes that required something I didn't have, and I've had to tweak fairly aggressively to make it work, but when/if it did work, maybe adding that feedback into the recipe would be valuable? e.g. user ticks "don't have a dehydrator", and recipe tweaks to "use your oven and set it low", etc.
Anyway... It looks awesome! Keep doing what you're doing, and I'll use it :-) Just some random thoughts that came to mind :-)
Some technical recipe books I've used just specify the weight of total egg. Not that useful for a home cook though.
For that matter, I'm curious what egg size descriptions is like in other countries; in Australian shops they're like condom sizes - I haven't seen 'small' or 'medium' eggs for over a decade.
I second Bruce's comment about cup sizes being regional. So are teaspoons and tablespoons for that matter. The wikipedia pages do reasonable coverage on this.
So good to clearly indicate which region's measurements are being used.
Always been a frustration of mine when reading recipes as having to figure out what country the recipe is published in to figure out what the measurements actually are. (Unless they are in metric, which is unambiguous.)
[1] https://en.wikipedia.org/wiki/Teaspoon#United_States_customa...
[2] https://en.wikipedia.org/wiki/Tablespoon#Traditional_definit...
Definitely do value that as an option.
Cookware passed down too could help to promulgate Imperial measures, we have a heap of tins that are inch-based measures. They're small effects but they add up.
The harder and maybe more important one, is an issue that GitHub itself still hasn't figured out how to solve either. There should be a process for a fork of a recipe being able to explain why it is better than the original, and have the fork be able to be voted on so that the 'best' fork as voted by the most users becomes the canonical 'Chicken Noodle Soup' recipe, or at least display the forks on main recipe ranked by popularity.
Then you could pick the particular substitution map that you care about and avoid forking recipes unnecessarily and have most recipes support most diet types if the proper map exists.
That'd be a nice edition to GitHub's forks page. Over there star count might be the best proxy.
A combination of popularity and the ability to choose to merge changes back into the original seems sufficient.
And its an interesting idea. Recipes can't be copyrighted and and recipes are generally derivative. The data is generally well structured, and fairly standardized (ingredients, equipment, instructions, photo, with optional fields for prep & cook time, servings, notes, difficulty, etc). But there are two primary issues:
1. Recipes don't really have a single inheritance. For example, I cook a lot & really enjoying cooking, but when I'm trying to make something new, I won't follow a recipe. I will read a bunch recipes, and try to understand the underlying ideas & steal the ones I think are interesting, then implement my own. So say I look up Tufo Matar find 3 recipes the make my own, if I want to contribute which recipe do I fork?
2. This may have been dealt with by the use of stars. when I was using forkthecookbook, there was no way of "bookmarking" recipes and the so users forked them, which lead to the results having huge number of identical forks. That being said it seems like the star system may resolve that. But currently it looks as though unmodified copies of the recipes still appear in the Forks list, which makes it harder to find benifical changes. Also it would be nice to have a history section with a message summarizing what changed.
All that said this does look nice.
Additionally, a nice feature to have would be to "import" a recipe, though since phrasing of a recipe is protected this gets a little bit legally complicated.
Yeah, the version history, etc will have a UI at some point. You're also right that needs to be some sort of indication for the diff between forks. We'll get there. People haven't been going crazy with forks yet (turns out if you post to hacker news, that changes :), so that will be worked on.
There is an importer already! It's not perfect, but it works with most recipe sites.
Thanks for pointing out forkthecookbook.com. I hadn't seen that, and it's definitely relevant.
It is clear you put a lot of care into this, and I am sure there is a bucket full of features you want to get to, but I have a big request.
What I really want is one step beyond a place to store recipes. I want a meal planning site! I want to create a pool of recipes that we like, and plug in how often this week we will eat at home. Then outcomes a grocery list and a plan for each day. Maybe it even sees what we have liked, and suggested new recipes to add to the pool.
I am this close to pulling the trigger on PlateJoy, but my biggest hurdle is I can't add recipes I _know_ we like to their list of experimental (to us) meals.
Really, I don't mind/enjoy cooking, but never seem to carve the time out to plan an entire family of four's meals a week in advance.
Some aspects of the site are open sourced:
https://gitlab.com/keithasaurus/simple_html -- basically a port of Elm Html to Python to address what I perceive to be flaws of template rendering.
https://bitbucket.org/cinckitchen/kunits -- just a unit library I wrote. It's central to the site's scaling, etc.
That's a practice that will continue. But the full source will likely not be made public.
Also, cinc looks really nice - awesome work OP
Thanks for reading, and have a great day!
{{ TEAM_NAMES }}
The Eat This Much team
The ability to base your own work off of others' is one of the most important features of github. It is probably also a feature that probably wouldn't necessarily be understood by somebody not in technology or not familiar with how git works.
If this was my product, I would have named it it "Fork". Putting that feature front and center and making it really obvious that it is the stand out feature of the app, IMO, would help people to get right away why what you've built is so awesome. Even if they don't know what the heck github is.
Are you using mptt or treebeard under the hood to manage the forking?
If you need more data, check out this:
https://github.com/dpapathanasiou/recipebook
I've also enabled it to talk to this via TOTP, if you ever decide to switch to/want to enable mongo as your recipe doc store:
https://github.com/dpapathanasiou/ARMS
Good luck with it, and keep us posted!
Good luck!
I like that you have a scaling feature. => https://www.cinc.kitchen/info/features
I know this may be too much to ask, but if you are taking requests, some basic nutritional info (ex: Protein content, avg. calories) might make it even more awesome for those of us tracking daily calories, protein (for atheletes etc). I understand things like sodium, fat might vary depending on how much salt or oil the person cooking the recipe uses, so maybe this might not be feasible to implement...
I'm not sure how you would go about adding this though, its quite different from the structure you have already
Edit: And maybe a shopping basket in case someone wants to order ingredients from a few recipes.
Edit (again): And don't forget about letting people order the equipment too.
The first is "proper" the second is colloquial. At least that's what I thought.
It's written this way simply because it's a question I've heard a lot, and when people say it, they usually combine "what" and "does" and say "What's the name mean?"
In this case "What's the name mean" expands to "What does the name mean".
"Whats the name mean" is grammatically incorrect. :)
A weird piece of feedback: Something about the UI gives me a substantial feeling of loneliness.
But I prefer the pinboard-esque design to a flashier one.
It's utilitarian and snappy - just the way my grandma made 'em.
I think that's a refreshing change for this type of site; You may want to consider a means of allowing users to curate and discuss recipes on sub-domains, some kind of API or some other means of splitting it up? A kind of crisp, clean social platform for food prep ....
Thoughts on what the feeling could be rooted in:
I Think it would help with more than 3 recipes on the front page. That gives a feeling of emptiness that can be confused with loneliness.
I also think there could be a better choice of image cropping. One that does the trick of making them all one size, but that also lets me see the food and thus inspires me.
Some of the white areas can be filled in with some nice text on how to use the product or ideas of what to do next?
This is mainly on the home/list pages. The page for the recipes looks pretty balanced
I think the most obvious is the tiny sliver of a picture on each card. It feels like I'm looking though a key hole.
Besides that, the website is great!
The recipe entry experience was great though, some small details that I think would improve things even further:
- Esc should clear the 'text entry' modal, I kept clicking this by accident when wanting to add a new section heading and it was a nuisance having to click the 'close' link
- '+ Ingredient section' should replace last ingredient row if blank
- Would be nice to have section headings (similar to 'ingredient sections' for recipe methods) to break up recipes with multiple discrete sections - Hard to find errors in long recipe. Could be more prominant, or add a 'jump to next error' button?
I'll try to respond to the issues you raised one-at-a-time:
1. 3.3 grams should be valid. Unsure why that would be a problem unless some weird unicode stuff???
2. There's no ingredient limit.
3. Esc clearing: I agree :) Clicking on the black transparency layer should also clear it. I just need to build it in. Maybe tomorrow :)
4. Plus -- you're right. Hadn't thought of that
5. Headings: I've left that for users to choose to do on their own. Hard to get too strict about that stuff. (Same reason steps aren't numbered.) I'd suggest doing all caps and a line break at each section.
6. Errors: I agree. There are a number of things ahead of that on the list of priorities, but it's a good idea.
Thanks for the descriptive feedback! Heston Blumenthal's recipes are always welcome on the site! Let me know if you run into any other trouble.
One to try might be "Remix" and "Remixes"?
Is forking really good for recipes, do you want to automate inclusion of upstream changes, surely not?
- "1 slice" (of bread) is not a valid quantity
- "to taste" (of black pepper) is not a valid quantity
Oh well. I guess I won't be using this then, at least not yet.
1 | slice bread
(blank) | black pepper, to taste
Some aspects of the UI/UX still need work.
Thanks for the feedback.
As others have stated below, forkthecookbook, forkingrecipes, and also recipelabs. All do basically the same thing. I just don't think there is enough market for it. Certainly was a fun little project though!
Is this project open source or does it plan to be at some point? I've looked at some of the other projects out there that are similar, but many seem to die based on the founder running out of steam or getting busy with other stuff. I would really be interested in contributing to this or a a similar project if anyone has a suggestion for one that is open.
Please allow me to G+ connect or facebook connect, b/c I don't want to have yet another password to remember, and I'd like to (maybe?) share some activity on Cinc with my FB peeps.
Just a thought.
For example the receipe of some cake where nearly all commenters advise to put 1/4 of the sugar advised. If you don't read the comment, chances are that even with a recipe approved by many people it tastes like shit.
I really hope this will grow and success!
Have you thought about simplifying the UI by changing "Forks" to something like "Twists" or Takes (as in 'my twist on' or 'my take on').
We all understand what I fork is but the average user will likely get confused, and not just because a fork is a kitchen utensil!
In all seriousness though, definitely looking forward to a public API. Great work!
Just asking
Glad to see somebody actually brought a similar idea to fruition :)
A couple weeks ago I was looking for a way to find similar recipes (forked recipes, in a way) and I ended up making my homebaked solution. [1]
I should be able to go to a user and their recipes with the same ease as on github. Having unique IDs for recipes exposed in the URL isn't really necessary.
But before I do that, what is the future for this site?
* Do you monetize by getting a cut from the shopping cart? * Will I be able to easily download my data in the future, like I do with git repositories?
My one criticism right off the bat is the name -- the pronunciation isn't immediately obvious. ("Sink?", "Kink?", "Kins?", "Since?")
Then again, I don't know if that's just me...
edit: Also should have known. "Cinq" is five in French as well.
- it's an SPA, so it's mostly just doing API calls.
- the database is designed to be fast at a few orders of magnitude of size beyond its current state
- everything is gzipped
Are you storing the recipes in Git repos or in the Postgres DB?
If in Postgres, what is the format? Do you put each receipe in a single JSON field? If not, what does the data structure look like?
Would be most useful to me with calorie support. I see some users are hacking it by adding it to the title or in the notes. That should be a strong hint.
I had started on a similar idea a while back, but never got around to it. Even have a cute domain for it (pifork.com) in case you're interested :)
An analog to Docker images would be something like Blue Apron or Hellofresh I guess.
Surprised it is not a startup yet..
E.g. I'm seeing 1 cup of steamed rice - being British I have no idea what that means. I turn metric on and it tells me 237 milliliters of rice, whereas I would be expecting grams. I'm pretty sure measuring solids by volume is a US-only thing.
The mix of Sans-Serif and Serif is a little weird.
You probably mean 480 kilo calories.
Otherwise very nice idea :)