HNHacker News
TopNewBestAskShowJobs

dvdhsu

3,806 karma · joined November 21, 2010

hn@davidh.su; https://davidh.su; https://twitter.com/dvdhsu
submissionscomments
dvdhsu··on Monetizing open-source is problematic
Hi all! I'm David, CEO @ Retool. I'm looking in to this right now and will update my comment here with a response in the next two hours.

Edit: just did a bit of digging.

It looks like we have a Retool template that has some fake data in it (https://retool.com/api-generator/). This is an app built in Retool, and has a few thousand rows of hard-coded data.

Most of this fake data is self-generated, but we used the faker.js library (https://github.com/marak/Faker.js/) to generate three datatypes: IP address, avatar, and industry. This MIT licensed library, when used, creates data that links directly to fakercloud (https://cdn.fakercloud.com/avatars/). Here is the code itself: https://github.com/Marak/faker.js/blob/master/lib/internet.j..., and here is a demo that shows the library generating those links: https://rawgit.com/Marak/faker.js/master/examples/browser/in....

I just spoke to the engineers who worked on this project, and we are sorry for including links to fakercloud. This wasn’t intentional, and we just pushed a commit removing all avatars from the template. This is already deployed. I hope you all can understand why we trusted the data generated by the MIT licensed project, and didn’t think it would link to anything proprietary.

I myself am an engineer (and avid HN reader, as evinced by how I found this while reading HN on a Sat night), understand Marak’s frustration, and agree that monetizing OSS is hard. While we’ve already contributed around $10k to various libraries we use (https://opencollective.com/retool), including faker.js, Babel, ESLint, and JSON Schema, I’m going to see if there is more we can do. We’ll be writing a blog post about it this week and I will follow up with more next steps. I wonder whether there is a better way of sponsoring OSS, other than just donating dollars every month? (Maybe we could commit one engineering-day per month for contributing back to OSS libraries we use heavily?) In the meantime, we'll certainly continue sponsoring all the libraries we’re sponsoring already, including faker.js.

(Also: I’m sorry to Marak for not responding to his email re. acquiring Faker. More in this child thread: https://news.ycombinator.com/item?id=27252420)

dvdhsu··on Uber, after buying Postmates, lays off more than 180 employees
This is one of my favorite comments on HN: https://news.ycombinator.com/item?id=25376346. It really changed my perspective on why Uber has so many software engineers.
dvdhsu··on Retool Raises $50M from Sequoia
Fixed. Thank you!
dvdhsu··on Retool Raises $50M from Sequoia
Hi! We've considered making Retool open-source (and we may yet). Is there any particular reason you'd want access to the source code? For example, you can host Retool on-prem (https://docs.tryretool.com/docs/setup-instructions), extend it with custom components (https://docs.retool.com/docs/custom-react-components), and bring in data from anywhere (https://docs.retool.com/docs/queries). And if you were concerned about bugs, we're comfortable providing very tight SLAs, along with 24/7 support if need be.
dvdhsu··on Retool Raises $50M from Sequoia
Thank you for the comment! Do you mind sharing what exactly you were trying to build and how Retool was too constraining? We'd love to learn which use cases we could be doing better in, so we can improve (even if we're able to convince you to try it again!). Thanks!
dvdhsu··on Retool Raises $50M from Sequoia
internal.io is pretty similar to https://www.forestadmin.com/, or even djangoadmin. If you're looking for a quick, non-customizable CRUD interface on top of your database, both of them are pretty good solutions.

Retool is really aimed at engineers who are building custom internal applications. For example, if you're Coinbase and looking to build a complex approval workflow for withdrawals above $100MM (with a custom UI depending on which Okta groups you're in, with custom approvals required depending on the KYC status of the customer, etc.), it's very very hard to do with either internal or forest (since the customizability and flexibility are much lower with both of them).

So: forest and internal are good if you want something quickly but have no intention of customizing it later. Retool takes longer to get started, but allows you to customize substantially more.

In the mid-term (i.e. 1 - 6 months' time) we want to make starting in Retool much easier. We will probably allow users to quickly generate admin panels (similar to internal or forest) with a few clicks on top of common datasources.

dvdhsu··on Retool Raises $50M from Sequoia
Hi there! Thanks for the comments, and thank you for using Retool!

We have a product in beta that (I think) does what you're looking for: the Query Library (https://docs.retool.com/docs/query-library). This lets you run queries, save them, and share them across your company. LMK if that's what you're looking for?

(In the future, we will be adding more permissions, etc. to it. We are actively developing this right now and would love to get some feedback from you on it if you'd be open to it? I will have our engineer email you!)

dvdhsu··on Retool Raises $50M from Sequoia
Thank you for using Retool! I'm really glad that we've delivered value to you all. Please let me know if there's anything we can ever help with!
dvdhsu··on Retool Raises $50M from Sequoia
Thank you for using Retool! If you ever have any feedback, feel free to email me directly. :)
dvdhsu··on Retool Raises $50M from Sequoia
Hi, thank you for the message! That is a very interesting use case. Do you mind emailing me at david AT retool? We've previously built something like this, but for Excel (you upload an Excel template, and we pull in data from various APIs [e.g. SFDC, JIRA, etc.] and fill it into the xlsx file). If you're interested in using Retool for it, I'd love to chat and learn more about the use case. (I bet there are many others who have a similar use case, hah.)
dvdhsu··on Retool Raises $50M from Sequoia
Yes, that should work well with Retool. You just need to build the app in Retool, and then enable "public applications", which'll let you expose the app publicly. We don't bill for public users.

(Theoretically, the end users could get all the data by inspecting the network requests. But there's really no way around that given that you're exposing the app and its data publicly. I suppose you could get around it by adding a password to the public app. LMK if you need help with that!)

dvdhsu··on Retool Raises $50M from Sequoia
Thank you — seriously — for trying Retool! You're correct on the difference — Retool supports writing back to databases and APIs, which is fairly different from most BI tools.

On the latency part — yes! We are currently working on scaling our systems and latency is a bit higher than usual right now. If this is something anybody is interested in working on, please ping me! :)

dvdhsu··on Retool Raises $50M from Sequoia
Thank you for trying Retool! I agree that we have a lot of room to improve on the writing of more complicated JS, as well as large-scale refactors. We are working on bringing git syncing to the cloud (currently it's only available on-prem), which'll let you edit Retool apps as code. I think that should help a good amount.

Anyhow — thank you for the feedback! I will send you an email in a few hours too. If you'd be willing to hop on the phone to dive into some of this in more detail, it'd be really appreciated. (I'll probably run a few potential solutions by you and see which ones you like.)

dvdhsu··on Retool Raises $50M from Sequoia
Yes, certainly! "VB in the cloud" is a pretty good description of Retool. (I think we're a quite a bit more developer friendly, but it's certainly 80% correct.)
dvdhsu··on Retool Raises $50M from Sequoia
Hi! Thanks for the comment. I'm an engineer as well, so the concerns you raise are things I've thought a lot about. I think Retool does a few things differently, which help:

1. On We allow you to sync all your applications to Git. All Retool apps just JSON, and we serialize that to YAML (that has pretty diffs). So when you make changes to your application, those changes can be synced directly to your Git repository, and you can use code reviews, PRs, etc. in order to manage everything. This means we also support code transforms (if you want to bulk-change Retool applications), support staging and dev environments, and more. https://docs.retool.com/docs/git-syncing

2. On the flexibility side — you can import your own React components. This lets you use the data-handling layers of Retool, but still customize the front-end as much as you want: https://docs.retool.com/docs/custom-react-components.

3. Most serious Retool users host Retool on-prem: https://docs.retool.com/docs/setup-instructions. By hosting Retool on-prem, you can be responsible for Retool's up-time. And because all updates are shipped via Docker, you can always downgrade / refuse to upgrade.

dvdhsu··on Retool Raises $50M from Sequoia
Great question! Against Airtable, primarily in two ways:

1. We are _not_ a system of record, and do not store any data. We'll connect to your data, no matter where it is (e.g. postgres, your own API, Stripe, etc.)

2. We are primarily for building complex UIs, not for simple spreadsheets. If you're looking for a spreadsheet for your database, there are lots of other options (e.g. postico). But if you're looking to build a UI (perhaps you only want some columns to be editable, or don't want to expose the whole database as a spreadsheet), you should use Retool.

To give a concrete example of a good Airtable use case: let's say you join a new company and have a list of friends that you admire and want to try and recruit. You'd probably use an Airtable for that. Retool is probably overkill here because you'd have to setup a database to store your friends, which probably doesn't make sense.

To give a concrete example of a Retool use case: let's say you're Doordash, and you want to manage all the orders in your database (e.g. you want cancel an order because the restaurant is closed). The only safe way for your support team to do that is you build an internal front-end for it. To do that, you're either writing React (or Angular, or Vue), or using Retool. Airtable probably wouldn't work here because Doordash is not storing their orders in an Airtable.

Against Metabase: we're primarily used for applications that both read and write data. If you're looking for some charts or tables that only read, I'd recommend Metabase (or Chartio, Looker, Redash, etc.). But if you want to write data back via Metabase (i.e. the CUD in CRUD), it's impossible without building a custom React app.

dvdhsu··on Retool Raises $50M from Sequoia
Fixed. Thank you!
dvdhsu··on Retool Raises $50M from Sequoia
Fixed. Thank you!
dvdhsu··on Retool Raises $50M from Sequoia
Hi all! David, founder of Retool here. We’ve come a long way since we our very first Show HN (https://news.ycombinator.com/item?id=14515494) and subsequent launch (https://news.ycombinator.com/item?id=17725966)!

Briefly — Retool is a visual programming language, built specifically for building internal front-ends. The idea is that we can let you use a visual programming interface to get you 70% of the way there very quickly, and then let you customize the rest with code. This lets you build apps much faster, but also retains the customizability and flexibility of code.

We support writing JS anywhere between {{ }} [1], importing custom React components [2] , hosting on-prem [3] (can be setup in 5 minutes: https://docs.retool.com/docs/running-retool-locally), etc. And then once your applications are built in Retool, we manage the authentication (via SSO if you want), authorization (syncing with your groups in Okta via SCIM), and audit logs (can be stored in your own database, so you can build apps on top of it.)

One interesting differentiator is that we don't store any data, and are happy to connect to any back-ends you already use (including both databases and rest / graphql / grpc endpoints).

Here’s a 4 minute demo video: https://d3399nw8s4ngfo.cloudfront.net/videos/intro-to-retool...

HN is what got us this far, so if anybody has any comments / suggestions, please feel free to let me know! We aim to be surprisingly responsive to HN feedback, since we’ve found it highly predictive of what developers in general want. Feel free to email me at david AT retool.com

edit: fixed video. Thanks!

1. https://docs.retool.com/docs/javascript-overview 2. https://docs.retool.com/docs/custom-react-components 3. https://docs.retool.com/docs/setup-instructions

dvdhsu··on Why build another website builder?
Hi there! That's what we're working on with Retool (https://retool.com) It's, as you say, a "DB-backed" (or API-backed) front-end builder that's designed specifically for engineers. We don't store any data, and you can customize the front-end by importing your own React components or by writing JS.

(I made it, so feel free to ping me if you have any feedback: david AT retool)

dvdhsu··on Amazon Honeycode – build web and mobile apps without writing code
Hi there! Founder of Retool (https://retool.com) here (and long-time HN reader). Agreed that Honeycode is adjacent to us. I think an interesting way of viewing the space [1] is who the product is for. There are lots of drag and drop app-builders (e.g. mendix, outsystems, bubble, msft powerapps, amazon honeycode, etc.), but they're all built for non-technical people. Their goal is to make programming more accessible to all. (Which I think is a very noble and very interesting goal!)

Our bet, however, is that code is here to stay, since code is actually a pretty efficient way of getting a computer to do something [2]. With that said, though, building specific types of apps (in our case, internal front-end apps today) is surprisingly difficult (have you tried recently!? you have to use react, redux, install 10+ npm modules, etc... just to make a front-end that has a table + POSTs back to your internal API!).

And so that's why we're building Retool for engineers, which has resulted in a few interesting decisions:

* we are not a system of record, since engineers don't like to move data around. We'll connect to your data, no matter where it is (e.g. postgres, a custom API, salesforce, etc.).

* we rely on the user knowing how to write code. I think this is interesting because low-code is good for getting to 50%, 60%, etc. But as you try to get closer and closer to 100%, it becomes harder and harder, since full customizability within a GUI (without code) is just hard (cf 2.). Retool lets you get to 60 - 70% of what you want very quickly, but then relies on you finishing the last ~30% with code, if you really want to. (We provide APIs, let you import custom React components, let you write JS anywhere [3] within {{ }}.

this also has led us to embracing git syncing (serializing retool apps as yaml files and syncing them to git repos: https://docs.retool.com/docs/git-syncing ), being on-prem by default (https://docs.retool.com/docs/setup-instructions), etc.

(If anybody is interested in working on this, please email me at david AT retool DOT com. I think there's really a small chance we can really change how business apps are built.)

1. Funnily enough, when we started Retool, there was no low-code / no-code space. If you look at our original Show HN (https://news.ycombinator.com/item?id=17725966), you'll see that nobody mentions "no code", "low code", or anything siilar. It's certainly been fun to watch a "trend" pick up around us, haha.

2. For example, imagine a switch statement. Very simple and concise in most languages, but incredibly hard to implement in a GUI. (A graph with nodes and edges, perhaps?)

3. We secure it by running all JS in a sandboxed iframe.

dvdhsu··on Ask HN: Is there any open source tool for creating simple CRUD Apps?
Hi, yes, this is what Retool is for! (https://retool.com)

It's like Metabase, but instead of for buliding dashboards, it's for building CRUD apps. LMK if you have any feedback? We've been working on it for a bit now.

Thanks!

dvdhsu··on Technology that changed air travel
I'm sorry. I made the change quickly at around 11pm my time (when I was a bit tired) because somebody in another thread here [1] noticed that a 767 wouldn't make it from SFO to LHR. I made two typos in this edit (typo-ing "286" to "287" and "388" to "399"), and have just corrected both. If there are any more issues you notice, please let me know (I'm david AT retool, or commenting publicly here works), and I'd be happy to fix them when I wake up again (in around 8 hours' time).

(The former error is indeed because I confused the outbound / inbound flights. I've taken BA286 quite a few times, but I forgot which direction was which. When I was initially fixing the article I thought it was "BA286" but then looked it up and found it was "287". I then fixed the error in the screenshot but forgot to fix the error in the text itself. Apologies the edit introduced annoying little errors — this is entirely on me trying to do fix something quickly.)

1. https://news.ycombinator.com/item?id=23231453

dvdhsu··on Technology that changed air travel
Ha, great question. We didn't have access a SABRE system so the lookups are fake. I was just about to go in and change it in to a 388 (my favorite plane!), which I think flies daily between LHR and SFO as BA286.

~Unfortunately, the output is in png format which was generated by carbon.now.sh, so I'll have to type everything out. I'll change it once I have a few minutes...~

Edit: just changed it into BA286, which is an 388. (For fun, I also added in first class fare buckets, which United no longer offers.) For additional accuracy, I also updated the UA equipment to 772s, which IIRC is what they actually fly (as opposed to 77Ws).

(If we did this more often, we'd probably actually design it in HTML instead of using Carbon, but an image was much faster, haha.)

dvdhsu··on Technology that changed air travel
Hi all! I'm the editor of this post. I love traveling, and have always wondered how reservations systems worked. This post explores that.

In the future, we're hoping to write a piece about how air fare pricing works too. There's a lot of interesting stuff re. fare buckets, forecasting supply + demand, etc. It's also very topical with COVID: https://www.flyertalk.com/forum/united-airlines-mileageplus/...

(If HN has any ideas about what else we should write about, let me know! I'm david AT retool.com)

dvdhsu··on Patio11's Law: The software economy is bigger than you think
Hey! I'm David, the editor of the post you linked to. If there are any other topics you're interested in learning about, please let me know! We wrote another one on Salesforce (https://retool.com/blog/salesforce-for-engineers/) that was well received on HN too (https://news.ycombinator.com/item?id=20277115). We were thinking about writing about Marketo / marketing automation solutions next. (I'm david AT retool.com)

(As an aside — when we started Retool, we ourselves didn't expect the "custom internal tools" to be such a big market. As it turns out, we now have customers that spend $400M+ / year on building custom, homegrown, internal tools! While we [or any single vendor] certainly can't capture all of it, our goal is to capture some percent of it.)

dvdhsu··on The Cost of Free Doughnuts: 70 Years of Regret
Hi, thanks for this message. I really appreciate it, since we generally don’t hear about “I didn’t buy because of the price” from the customers that already paid us, haha.

One thing we don’t advertise — we only bill for active users every month. So if 10 people login during a month, but 50 don’t, we only bill you for the 10. Do you think that’d help you guys?

(I’ll send you an email in the morning to get more feedback and see what kind of pricing would work for you. Thanks again for the feedback and for trying Retool!)

dvdhsu··on The Cost of Free Doughnuts: 70 Years of Regret
Hi, thank you for the kind words! (I'm the founder of Retool.)

Per-user pricing is indeed tricky for us, since people oftentimes think about the value as per-app, not per-user. But ultimately, we want to encourage, not discourage the building of apps on Retool, and our best customers view Retool as a platform for building all their internal apps. (The best customers typically build 100+ apps on Retool.) That's why we don't price per-app.

But it's possible per-user pricing also isn't optimal. If you're open to it, I'd love to bounce some ideas off of you. Do you mind sending me an email? (I can't see the one in your profile.) Thanks!

dvdhsu··on The Cost of Free Doughnuts: 70 Years of Regret
Hi, I'm the founder of Retool. (I do indeed read a bit too much HN, hah.)

Thank you for the feedback — that does sound infuriating. We were optimizing for sharing of applications, but it sounds like the workflow you've adopted is really quite annoying. I'll have a think about what we could do here that might help (e.g. perhaps disabling "auto-adding" to accounts). If you have any suggestions please let me know (my email is in profile)! (I will send you an email by end of day otherwise with some thoughts...)

dvdhsu··on Ask HN: What are your favorite low-coding apps / tools as a developer?
Retool (https://retool.com) might be what you're looking for on the front-end. It's built for engineers, so it abstracts away a lot of the boiler-plate stuff (e.g. fetching data from an API, showing errors if it fails, showing a loading indicator on the button when the REST API is in progress, etc.). But you still write code for the custom bits (e.g. if you want to hide a component for certain users).

Here's a 3 minute demo video: https://cdn.tryretool.com/videos/4_minute_demo_4827ae.mp4

It's something we started working on a few years ago before low-code was a thing, haha. It's funny to see what you work on become a buzzword, haha. If any of you have any thoughts / feedback, please let me know! (HN, honestly, has been the main source of feedback for us as we've been working on it.)

← PreviousPage 2 of 11Next →