I agree, I prefer OOP to much of the spaghetti and 5 layer nested functions that is popular now. This feels like one of those cyclical things that will come back in vogue at some point. People will have trouble maintaining these systems over time, then "discover" OOP and encapsulation, write blog posts about how it is The New Way, etc.
We have some React class components in our system. People who are < 3 years into their career don't like them as they've been told they are complicated and generally considered harmful. Over time, these people tend to come around and realize that class components can be very useful for more complex components.
I feel that functions + data needs some really good linting to keep things clean. The one problem I've never been able to find lints for is "keep related things together" because the implementation of that varies so wildly.
Thanks for the heads up on sponsorships, we hadn't thought about accepting them. Currently the project is effectively funded by us at useanvil.com, but we'll talk about opening that up!
We added a number of things that came to mind while trying to get our docs dialed in ourselves. Glad to see that others were looking for these same enhancements! Let us know if you run into any issues
I appreciate the integrated examples. I wish I had your example generator before I rolled my own.
The documention generator we currently use isn't scaling well with 100+ types and 800+ mutations. It makes the sidebar too unwieldy. It looks like SpectaQL has the same issue. Not sure there's a great solution while keeping to static pages.
Thanks much for the open license. I hope to try it soon.
We might pay for a searchable hosted white label version.
Thanks for the feedback! Yeah I can see how it could get out of hand with that many types and mutations. The single page situation would probably break down in that scenario too. We'll think about how it might be able to scale. Feel free to open an issue on the repo and we can work it out together.
What I'd really like to see is a dynamic diagram of the schema, such that when you click on a type it becomes the center of a graph of associations. Then click to view the text docs. The diagram could be a menu / entry point for the text docs. A kind of map to the N-dimensional space.
Ah that would definitely be interesting, you could break out queries / mutations where it is an input, output, or related types where it is a resolver type
We quickly ran into the same issue at Shopify. We've since moved from a flat list into sections by "domains"; within each section is the breakdown by type. You can see it here: https://shopify.dev/docs/admin-api/graphql/reference
Of course this grouping has to be done a little manually. And it's not perfect, but definitely better than the flat alternative. For reference, we have thousands of types.
I could see us adding support for grouping via the "metadata" that would inform the nav bar on how to group and accordion / collapse things. There already is a pattern for grouping in the library that is exemplified in the Query and Mutation carats
Thanks for the kind words! I wrote most of the enhancements and thought the static and dynamic example support was crucial to have.
Our schema is not as big as yours, but I could see how things could get a bit unwieldy on the navbar. Come on over to the repo and we can work on a solution together - I'm sure there's something we can figure out if we put on our design + product + engineer hats.
Hey! Anvil founder here. Anvil’s goal is to help the world move beyond PDF’s and paper forms. We are building a simple online tool to help anyone convert their existing paperwork into online workflows that can be shared directly with customers / employees.
Imagine being able to create a TurboTax-like experience for any form or set of forms. Then instead of emailing around sensitive information, and manually typing it into other software systems, have the information automatically synced to where it is needed. This is what we are trying to do with Anvil.
As we build out our conversion tool, we have been dog-fooding it by converting a bunch of forms we need into Anvil Flows. Today we are launching a subset of these forms for anyone to experience the joy of filling out an Anvil form when compared to a musty old PDF or even worse, paper form.
If there are forms you hate filling out, post in the comments and we will prioritize getting those converted into Anvil Flows!
I would love to use this to let my clients fill out my contract forms. Right now I make a little form for them to fill out then I hand bomb the form. I Fill in additional company information then send it off for a digital signature.
Anyway I we can work on getting my forms digitized.
Other Anvil founder here. Thats great to hear! Shoot us a message at hello@useanvil.com mentioning the ShowHN and we would be happy to see what we can do for you.
This is sad. Looking forward to the post-mortems. A question in the interim: Did rethink ever consider building a SaaS business around the product? What were the reasons against it? Were there (a lot of) customers asking for a hosted version?
Are there certain types of continuous views where continuous triggers work better or worse? I can imagine things like alerts on moving averages could get tricky and possibly send repeated alerts when predefined thresholds are crossed multiple times.
They work better for non-sliding window queries. Triggers on sliding window queries are much more resource intensive, both for CPU and memory. Essentially, the trigger process has to keep track of tuples for each step in the window and combine them whenever any tuple is updated to get the new value.
You guys should check out OctaveWealth 401k (https://octavewealth.com/). They are also YC, have a flat fee 401k (no percent based fees!), and build their investments models in-house.
I was curious if YC had any issues with accepting/funding companies that are in the same space as other portfolio companies. I have heard that venture capitalists often don't. Zenefits and Zen Payroll/Gusto was what came to mind.
My startup is in the financial wellness space of the 401(k) industry (we are not advisors and just do participant education) and I'm going to be applying to to the next session so this bit of news is of interest to me. Thanks.
It's still much slower than, for example, Sublime Text. A search for a single letter [1] in a ~5,000 line file occurs instantly in ST, but takes about 1 second in Atom. It's not much, but it's very significant from a UI point of view.
[1] I know such a search is ridiculous, but both editors perform search-as-you-type, although atom does attempt to delay that if you type quickly enough. Anyway, it's just an example to demonstrate the speed difference.
There's still some major algorithmic optimization to be done regarding the time it takes to run the find-and-replace search. The optimizations discussed in this blog post were more focused on the performance of editing the buffer in the presence of large numbers of search results.
Right. Don't get me wrong, the article was interesting and I appreciate there's a lot of hard work going on. I really hope atom is a success because an open source equivalent of Sublime Text would be very welcome.
We have some React class components in our system. People who are < 3 years into their career don't like them as they've been told they are complicated and generally considered harmful. Over time, these people tend to come around and realize that class components can be very useful for more complex components.