Like how select cuts of beef slow roasted with veggies and served on a bed of glazed hand cut potatoes.
Is more appealing to a potential cook than cut beef in a slow cooker with some wedges.
I think the reason Google wanted microformats is for Google Assistant recipe answers.
Umami has Chrome/Firefox extensions that you can use while on desktop, which will sync your recipes to the app. Disclaimer: I built it :)
^ Colors in that space only just recently started getting supported by displays/browsers. And while p3 colors work in css on Safari, I had to use a single-pixel background png to make it work in Chrome.
But yes, you are right, as far as native apps go. FWIW, I just got an Android test phone so I can start on that version, should be ready in a few months.
1. Import from other recipe apps.
Recipe apps and sites have a finite lifetime. Its great that you already support export, but import from others would be great. Pepperplate for example is sqlite on android. Other apps might be harder, eg the New York Times recipe box. But without import, its hard to commit to a new system.
2. Cooking mode
This is a mode that avoids that situation where you are following a recipe and need to manipulate your phone but you cant because you have hands covered in flour/chicken juice or whatever. Voice control would be awesome too: "scroll down", "set a timer for 20"
3. Notes Some place where you can comment on the recipe, like "use peanut oil not canola"
Thank you
> 1. Import from other recipe apps.
Definitely agree. A couple days ago, someone asked for a Paprika recipe importer, which I've added here: https://www.umami.recipes/paprika-import I'll have to take a look at Pepperplate and NYT recipe box (though, you can already import any NYT recipe link!).
> Voice control
Love this idea.
> 3. Notes
There is indeed a notes section! Though I'm guessing you mean inline comments, like google docs style?
On 3, notes can either be for personal use (eg in Cozi's recipe handler) or a shared conversation in a recipe (like what the NYT does). I feel the former is more important, but your users might have different opinions.
This works great for things like StackOverflow posts where you can quickly judge if the answer is correct, and if it's not, you can quickly bounce. It's not great for recipes, which are almost always "correct" (in that they are a recipe of the type you're looking for), so spending longer probably just means you scrolled for longer.
Regarding context-switching: that's even more of an outlier! The vast majority never worries about that, and from my experience even among knowledge workers, that's rare.
I don't think most online recipes are "correct", especially chili recipes that call for Heinz baked beanz. Wut?
So, for me, a recipe can indeed be correct or not. My definition of "correct" is narrow, though -- if it produces a dish I enjoy, it's correct. Otherwise, it's not.
longer times on a page means longer enagement == more ads revenue.
is it possible that google is incentivizing longer engagements via adwords by using search priority that ranks by average time spent on page?
http://microformats.org/wiki/h-recipe
But it looks like there might be one or two others active in the wild.
Google presents what its customers (people doing searches) want to see, so shouldn't we ask the people who click the results that make it to the top (whether they leave the page dissatisfied or not is a different story, right?) over and over?
What incentive does Google have for prioritizing succinct quality posts versus blogspam posts? What incentive does Google have for prioritizing blogspam posts over succinct quality posts? Why is it the way it is if we're talking about the unsponsored (not paid) organic results?
As a blogspam author, why do you want people on your page? To get more AdSense views/impressions? You typically aren't selling anything on a website that gives away free long-winded recipes (so conversion rate/cost per click/paid advertising isn't really a concern?)
Google presents what it thinks people want to see, which it infers based on some number it can collect, like time-on-page, or scroll depth. The argument is that its proxy metrics for "the user liked this" are ill-considered, especially in this particular context.
Even when Google acts like a rational actor, it is not an oracle. It rationally optimizes an observable metric, which may or may not correlate to its actual goal.
Luckily it's also one of the only cases where a microformat is actually nigh-universally used. Outside a few sites where the "fluff" isn't fluff and is actually good content itself (e.g. Serious Eats) I just throw the URL into Paprika 3 (no affiliation, and there are tons of other programs and probably some browser plugins to do the same thing) and let it strip it down to the ingredients and instructions by reading the microformat markup. Works flawlessly nearly every time.
[EDIT] Credit where it's due, I'm also pretty sure the microformat is only popular because of Google. I think they downrank sites in recipe searches if Google can't "see" the recipe, using the microformat. They may be both the cause of, and solution to, the problem.