Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: Github for recipes (forkingrecipes.com)
134 points by dpick on Jan 26, 2013 | hide | past | favorite | 99 comments



You need to hire a designer for this. Cooking/recipes/food markets are really visually driven. You also need to define what a fork is right on the landing page. A fork for a cook is not the same as a fork for a programmer. What you should have is a good header area that has the application work flow explained in graphics.

Example:

    Alice liked Juan's recide for thai rice. 
    She wanted to create her own version. Alice
    forked Juan's recipe and modified it to fit
    tastes.

    In forkingrecipes, the act of forking is when
    you give someone else's recipe your own touch.
    Thus creating a new recipe for others to use and fork.
I know a good designer who can help you with this, by the way.


I'd like to second this, and also say that you should really work hard to establish "fork" and "forking" in contexts that make sense to non-developers. They're second nature for us, but non-technical cooks will probably either read it as "fucking recipes", or conflate fork (to diverge) with fork (eating utensil).

That said, awesome idea.


Hi, I am part of ForkTheCookbook [0]. We actually actively face this problem, which is why in our FAQ, the 2nd question tries to answer that: http://forkthecookbook.com/faq

[0]http://forkthecookbook.com


Hi chewxy,

Do you have any insides how much those recipes get actually forked? I'd be afraid, that forking might be a nice idea, but actually isn't that common.

Thanks for your insight.


Hey, if you have played around with the website, you'd notice a lot of forks. We spent about 3 months convincing people forking was a good thing before finally launching it.

If you wish to know more, contact us at admin[at]forkthecookbook.com . Sorry for this, as half of us are terribly ill, and we take turns to answer questions


Tried going to the links and for some reason they are triggering my McAfee Site Advisor with "Warning: Suspicious Site".


Hmm, I wonder why. Is there any way to check why?

EDIT: According to Site Advisor (http://www.siteadvisor.com/sites/forkthecookbook.com), here are their criteria:

Country indicates where a domain is registered | US/AUS

Popularity indicates how many visitors a website gets | Not very popular

E-mail indicates our overall opinion of a website's e-mail practices | We don't spam

Downloads indicates our overall opinion of a website's downloadable files | No file downloads

Online Affiliations indicates our overall opinion of a website's linking practices | We vet every link. Every write up. It could be one of our authors are having a malware issue. Not sure about that.

Annoyances indicates our overall opinion of a website's pop-up practices. | No popup.

Exploits are rare but extremely dangerous security threats caused by a website "exploiting" a security vulnerability in the browser. | Don't think we have exploits. Something to check

Web reputation McAfee tests websites for web reputation using the TrustedSource™ system. | Totally not helpful

Reviewer and Site Owner Comments | I see no reviews


Possibly with a comparison like "like a fork in a road or tree".


I totally agree, unfortunately I don't have money to pay one and finding someone to work with me for free hasn't been easy :(.


Send me an email. Working on something similar and am a designer :)


I'd be interested in working on it as well, email me if you can't find someone.


Perhaps showing a fork as an icon, and then a simple recipe of a different variation along each prong might help visually?


We launched Fork the Cookbook about the same time as Forking Recipes too. You might want to check it out: http://forkthecookbook.com

We try not to call it Github for Recipes because that's not our target market (although other people have called us Github for Recipes)

We blog over at http://theforkingchef.com so you might also want to check that out.

OK, self promotion over


I like your site's presentation quite a bit more (ForkingRecipes just comes across as data carelessly tossed into Bootstrap), though I don't quite like how the food images are mostly obscured until you hover over them individually.


Understandable. We're working on a variation that doesn't do that. UX is hard yo.

Fork the Cookbook has existed longer than Forking Recipes, and we've done hundreds of user tests. It turns out users really like certain things, and so the current darkened background is a compromise on that solution.

We found out that users liked pictures, and also liked text (surprise?). In the next test we're doing, we're going to be swapping the order of appearance to see if the CTRs increase


>> OK, self promotion over

Good. Not cool to hog someone elses HN thread.


Hee, just being cheeky. Of course David's free to be cheeky if we ever post a ShowHN


\o/ I asked for this 16 days ago (http://news.ycombinator.com/item?id=5033978) and now it exists. God I love HackerNews :D

Probably not related, though still cool.

What I'd like to see though, is for all the pictures to be of food. There's a lot of happy faces, but the point of a cook book is to make me feel hungry when looking at the pictures. Not sure how to achieve that though.


Hey you may want to check out Fork the Cookbook [0]. It was there long before you asked the question.

http://forkthecookbook.com


Cucumbertown does this very well. They allow you to take a recipe and fork it – called “Write a variation” (on the left) [1]. IIRC quite a few recipes there are created this way.

[1] http://www.cucumbertown.com/recipes/55228/baked-mac-cheese


I like it, although I think there are other organizing principles to explore. Seconded on all the 'fork' comments its not at all intuitive to a chef what you mean here.

At present the top level abstraction is the recipe which you've sort of equated to 'source code' but when you look at something like Github or Sourceforge the high level abstraction is the 'project' and in the food world the closest I could come up with is 'cuisine.'

Consider one of my daughters favorite sites: http://smittenkitchen.com/ this site has lots of interesting recipes, and one might use them as a top level abstraction. So you've got the 'smittenkitchen' project equivlent or maybe 'smittenkitchen-breakfast' project which is breakfasts from smitten kitchen.

Then there are chefs which are much more important in the cooking world than individual developers are in the programming world. That is because the chef's 'taste' really defines the product in a way that is fundamental to the enjoyment. So you might want to build a '<Chef>-<category>' type model. Then you could do a pull of <Ramsay>-<appetizer> and get various appetizers that Gordon Ramsay makes.

Of course that will get you into a licensing battle. Because like fonts, recipes have this weird quasi relationship to name vs instructions (can't copyright instructions, can copyright the name, google Font Copyright for more info)

Conceptually I think this is a really winner idea (sadly easy to copy) but if you get the right mix going and critical mass it could easily be as successful as github or sourceforge.


Hey, at Fork the Cookbook, we're doing just that right now! It's one of the primary reasons why we said Search is Hard [0]. And we're working hard on search.

We've also put a lot of time and effort into researching the licensing issues, as well as design issues: how to present certain things like diffs, and stuff. So if you like, you should check out Fork the Cookbook: http://forkthecookbook.com

[0] http://theforkingchef.com/post/40890655560/search-is-hard-pa...


From a quick browse, I don't see any way to view diffs?


Hey there, it's not a feature we've pushed out live yet. There is code to see diffs, but no on the public facing UI yet.

The reason is this: We don't want to aim at the tech crowd (which is why you haven't seen a Show HN from Fork the Cookbook). The tech crowd knows what diffs are. We're used to reading things that look like this:

    + def function(var1, var2):
    - def oldfunction(var1, var2):
But the public aren't used to reading stuff like this. What we're working on is a comprehensive way to explain what diffs are, and what it means. So we're back at square one, designing a representation that is human worthy.

It'll take time. Sorry


I'm working on something in a domain not dissimilar to recipes, so I'm always interested to see different diff presentations.


Tell you what, I'll let you know first thing when we have diffs ready to be presented to the public


I like how easy recipe discovery is on forkingrecipes. I think the waterfall layout and search bar were good decisions.

Mappum and I made something pretty similar for the last Node Knockout.

http://foodhub.jit.su/

Like orangethirty said, foodies tend to be really visual, so we encourage the user to take a high quality photo by making it the focus of their recipe page.

Good job, I'm excited to see where you go with this. :)


[Forkinit](http://forkinit.com/) does a good job of this but hasn't been updated in a while. You might check them out for other ideas in this space. Looks like you have a great start.


Just like every other cool page for managing recipes: Please add automatic conversion from/to metric units.


And even between standard units. I have a re pie that calls for 6 tablespoons of water. I always have to translate that to fractional cups


I like it. It seems you are aiming at geeks. And I think geeks want a powerful search interface and this is why I haven't found a website i really like. I want to make queries like "Recipes with Tomatoes, Paprika but without meat of type soup" or "I only have Garlic, Paprika, bread, Cream, flour, and sugar, eggs and spices , what is the closest meaningful lunch? Not in natural language, but with a interface. Like a textarea for "what i have" "what i don't have/like", and "What kind of meal would it be" each.


"I only have Garlic, Paprika, bread, Cream, flour, and sugar, eggs and spices". Thats not a sane way to cook. You can season your eggs on toast (scrambled eggs with garlic makes a change), but there is nothing else sanely cookable form that list. You need to shop better, this is not a search problem. (I appreciate that you made up that list, but thats not the point).


Thanks for the idea, I'll give it a try.


Quoting http://www.badstartupidea.com/:

First, go and find out about Github. Figure out what it is they do, how they attract users, and how they make money. Github is already operating well in the Programming space, but perhaps the same basic ideas can be applied in the Cooking space. That's where you come in.

Edit: I think it's a great idea, it's just funny that that site mentioned this after only a couple refreshes.


I assume that's where it started http://news.ycombinator.com/item?id=5119787


That's really cool, actually.


That is really funny! I actually made this in the past few weekends so it's a total coincidence.


Having looked at some of the recipes, it might be nice to take the job of layout/formatting away from the recipe submitter and offer some differnt layout options to the person viewing it.

For example, some recipes include the accompanying photos inline, while others collect the photos at the end. If instead, the submitter merely tagged the photos according to what step they accompanied, then the website code can lay them out beatifully in a number of different ways to suit different viewers.

I think I'm thinking something along the lines of LaTeX philosophy: You provide the content and leave the layout to the experts/computer. So you don't write that "Zest of 2 lemons" should be part of a bulleted list, you write that it is an ingredient and leave it to the layout to put it in a suitably formatted ingredient list.

Another advantage is you reduce the ability of less savvy users to make parts of your site look ugly.


That's a great idea, thanks!


I started something similar[1] last Christmas (but lost interest as I never documented my own experiments). It's a great idea and there's need for a de facto standard source for recipes and cooking instructions.

My "next step" was to contact sous chefs of famous restaurants.

[1] http://palatelab.com/

Edit: rephrased.


Maybe Google's Recipe microdata?

http://support.google.com/webmasters/bin/answer.py?hl=en&...

Which is standardized at schema.org: http://www.schema.org/Recipe


I've thought a lot about this idea as something to add to my site Eat This Much (http://www.eatthismuch.com)

I have a pretty basic implementation of it right now - you can click a "personalize recipe" button on recipe pages to make modifications to it, and watch the nutrition change as you make changes. As a premium member, you can then have the meal plan generator suggest your modified version of the recipe in place of the original. When I have the time, I'm hoping to make all the existing branches and variations browsable.

There are some pretty great suggestions here I think. It would be really cool if you could build out an API similar to yummly's for serving up recipes. With different branches for each recipe, you could easily search through the variations to satisfy filters based on a user's tastes.


Hey Guys,

I'm the creator of Forking Recipes, I'll be happy to answer any questions you have and thanks for taking a look!


I agree with the other commenters: great idea, but needs a designer.

By the way, are you based in Chicago? I met somebody at the Chicago HN meetup last year who was talking about building this exact idea, and I loved the idea back then too. Are you that person by chance?


I am actually in Chicago, but I haven't been to a HN meetup so you're probably thinking of someone else.


To the GP, that was probably me. Unfortunately, we never got anything off the ground.

dpick, I'd be happy to share the ideas we were kicking around when thinking about this if you want to shoot me an email: john [at] nineteeneightd.com.


I really like this! Hopefully it gains momentum because it'd be really neat to have [a tool that can be used as] a crowd-sourced recipe-optimiser.

On a different note, I'm on a slow internet connection and it's very noticeable that the page is huge and not optimised for bandwidth (some thumbnailing and gzip-ing would be good to add at some point, to make the site as snappy as possible! :) ). https://developers.google.com/speed/pagespeed/insights#url=h...


Thanks for the feedback! I'm actually working on making the main page use thumbnails right now.


I just bookmarked a couple of recipes. It would be nice if the default name when saving as a favorite included the name of the recipe rather than the name of the site (particularly for those who want to bookmark more than one recipe). Think you probably just have to make the "<title>" include the recipe name.


I really should have done that a long time ago. I'll add it now, thanks for the tip.


I would use it if you built in conversions for us that think in deciliters and not spoons or lb.

Ie a dropdown on the profile settings page.


Great job. Maybe you could also use Foodspotting api to get awesome images of food.


You got it from badstartupideas, didn't you? :)


Haha actually I didn't, I started working on this a couple of weeks ago in my free time.


http://www.cucumbertown.com does this very well. Cleaner UI and seamless experience. Also has a lot of celebrity chefs onboard as users.


I know that twitter boostrap is a great tool to start a project and to prove the concept. I love the idea as a amateur cook and developer (not amateur :D), but if you want to reach a larger public (non developers) I think you should invest some time (and probably money) into a good designer (by the way, is a website where we could really "rent" a webdesigner? I know some where you coul rent a coder but not a designer).


99designs and designcrowd, to name some..


I love this. Every time I look up a recipe, I read the comments to see popular modifications and tweaks to the original recipe, and thought it would be awesome if those could branch off and become their own, with the most popular 'forks' floating to the top (even dethroning the original recipe, if they are indeed better). I'm glad to see this come to fruition!


Interesting application of software development paradigms to cooking. I think it would be useful to have categories, e.g. breakfast/dinner/dessert, hot/cold, and suchlike. And maybe a regular 'featured menu' on the front page to illustrate a meal combination, with a nice photograph shown and some encouraging body text.


this idea is awesome. The "forking" idea is great for recipes, but you need almost something like a difftool to be able to tell what's different and why? I think with a better UI and way's for users to interact with each other this could be extremely cool. Stick with it and good luck :)


+1 on a diff-tool. Its very hard to tell what's different between the forked recipes. Some sort of short commit-style message on the change would be helpful. (Something like "Added potatoes and carrots" or "Smoked the chicken instead of grilling it")

I really like the idea, but as others have mentioned, it does need some design help.


There actually is a really basic diff tool right now (Click on revisions on the recipe page). I'm working on creating a more advanced one now. Any ideas on what a good interface might be?


A diff visualization for ingredients should be pretty easy, you could show left and right and the delta between them as three distinct columns.

As for other distinctions, such as "mince" instead of "dice", that would be tricker, but you could probably do that with a human-friendly strikethrough coupled with a handwriting typeface to indicate what was changed.


http://www.chefday.com/ is Heroku for cooking :)


I thought Heroku for cooking would be that I, as a user, upload a recipe, and the food is cooked and delivered to me as a completed dish. If anything this is more like Arduino for cooking, maybe? An accessible jumping off point for an intimidating task?


Fair point. But with Heroku, you still need to build your app yourself. Coming up with a recipe is really easy compared to build an app from scratch. Maybe we're the Ruby on Rails of cooking. What do you think?


That seems like a fair assessment. I didn't give it much thought before, but now I'm really curious about the model; it this supposed to be exclusively for dinner parties, or a way for bachelors to learn to cook? I don't live in New York, and there isn't a ton of detail on the front page. How far in advance do I order? Do I pay a flat rate for n meals a month/week? I'm guessing the invite comes with more details, but looking at the landing page I have a lot of "How does this fit in my life?" sort of questions.


Thanks for the feedback. I'll update the landing this week. The most common use cases we have so far: dates / young couples that want spend quality time, friends dinners, supper clubs, cooking enthusiasts who want to learn new recipes from top Chefs... You just order the recipe you want: no subscription yet(people have requested it). Cut off time is 10pm the day before your delivery. If you have friends in NYC, feel free to share the joy of cooking :)


You should remove all of the pictures that do not show the food and replace all of the low quality pictures as well. When I went to your website the first thing i saw was a bunch of faces and non-appetizing looking onions. You want to make the food look amazing.


Very nice! I'd like to have the option (but not the requirement) to log in with oauth(2) from Facebook, Google, Twitter, etc.

Your site will get a lot more traction if you implement that, I think. Most people (even non-techie folks) are suffering from account proliferation.


Actually, its been shown that the confusion of which account type to use is worse than the bonus of having options at signup. I forget who did the testing, but non-techies have trouble remembering which they used as well as understanding what it means to "sign in with..".


I'd be interested in seeing the study. For me, at least (and I think for many others) the thought of creating yet another login is enough to make me back out.


Okay, I misremembered a bit. They showed that a tiny portion of people used social login (~3%) and that the gain they thought they'd made with them (fewer forgot password/username issues) was due to an unrelated change. The tiny portion shows that it can't be that important (plus, how many of the 3% that took advantage of it would have left without those buttons?).

The study can be found here: http://blog.mailchimp.com/social-login-buttons-arent-worth-i...

I apologize for the misinformation in my first comment, but I think my point remains. That said, I feel similarly to you - I just think that a much smaller portion of people feel that way than you realize.


Will do, thanks for the suggestion.


It's also a good idea to make it request only the minimum amount of information from the auth provider (e.g., name and email address). I (and I think others) have become wary of sites that ask for full access when there's no obvious reason for it.


This is awesome. I think it's a coincidence because the other day on HN someone said something about a "github for recipes". I'm not saying that to be coy or to be condescending, simply coincidental.

Keep up the good work! As an avid amateur chef, this could be fun.


I like it. You need to improve the editor (normal people won't understand markup, just have a text box for ingredients, another for directions and a list for images).

You also need tags so I can tag a recipe as 'Asian' or 'BBQ' to make it organizable.


This is pretty amazing. I had thought how cool it'd be to have something like this a bunch of times (specially when I want to find simplified curry recipes.)

I seem to recall someone put something similar on HN a few months ago. Any ideas who it was?


That's cool!! Can this be adapted for curriculums? I am a medical student and have been posting medical curriculum at https://github.com/aloo/hack_medicine


It totally could be, send me an email and we can discuss it.


Neat. When I saw the headline I thought that this would be too technical to get traction with non-geeks but it looks nice and I could see it becoming popular especially with particular diet communities like paleo etc.


I find it waaaay too geeky for a cooking website. The idea could be useful but as it is right now, it's very hard to understand how it's supposed to work. (Revision 2, what?)


Great idea! It seems obvious in hindsight, at least to people who understand forking and dvcs. Which is great.

Let's see how non-tech people are able to understand and use such concepts.


We made something very similar for Node Knockout (the 48-hour hackathon), http://foodhub.jit.su.


User experience still needs a lot of improvement, but I love it. congratulations.


Thanks!



Nice idea but the design looks like an poster for Trainspotting. Expecting to be emailed recipes for deep fried mars bars any time now.


u just wait...


Great idea. Needs a better app and UX but really good idea. Good job!


You missed the opportunity to use the name GitCipes.


And the average visitor would end up asking themselves "What does 'git sypes' mean?"


that doesn't sound too great; unless I'm missing something, it doesn't involve git.


Why not build this right on top of GitHub?


Because most cooks are not developers and don't have Github accounts, I guess.


I think trhaynes meant, Why not build this on the Github API?


I'm not so sure about this.

Why not use github Pages and scrape the github repository for necessary data? It's all harmless recipe stuff anyway. Non-github users could be asked to use e-mail to place recipes, and can be informed that they can use markdown. Or even present a web form that sends an encrypted e-mail for them, employing that wicked JS textarea enhancer we saw the other day, for markdown syntax or even allow haml.

    Re: {url_block}/new

    Recipe Name: {name}
    Date: {extracted_from_email}
    Subject: 
    
    {haml_recipe_body}
The github haml pages can have links station'd at the top or bottom to pre-fetch categories, previous, and next pages. The github Page Index can include a JS that builds the UI of the site via DOM manipulation, based on a lightweight Web app hosted elsewhere and called in via ajax.

Summary: Total MVR (minimum viable repository).


What an awesome idea! Love the site.


Man, I had this very idea a few months ago. :-)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: