Co-founder of Reflect here: Thanks for the mention!
I think a lot of people think of "no-code" tools as something exclusively for non-developers; things like Webflow and Bubble for creating apps. I don't think that's ever really been the case, Zapier being a great example of that, but there's been a lot more tools coming to market in the past couple years that actually work well as a way to replace tedious code-based workflows. i.e. Use Retool instead of building UIs for internal apps, use Reflect to build automated tests instead of Selenium, etc.
Yeah I'm a dev but was a team of one embedded within a sales and marketing org at a a medium-large size startup with a mandate to automate and find growth hacks for sales + marketing, and I was able to use Zapier with its Salesforce integration, google sheets integration, Twilio integration, Marketo integration to do a lot of work that I wouldn't have had time to build, support, and do maintenance on if I'd built a full backend application.
All of what I did with Zapier could have been implemented with a database, a bunch of scripting, and cronjobs, but using Zapier saved oodles of time and meant I was outsourcing the maintenance to their API integrators rather than to myself.
I was able to do so much with triggers and webhooks I can only imagine what is possible now with their more recent code blocks feature and adding better environment variables.
I would add to this list no-code web automation and scrapers. I am founder of https://automatio.co, but there other great tools like SimpleScraper, Axiom.ai, Browsers.ai etc
I'd add Betty Blocks to the app builder section; and add a category for electronics laboratory automation and add 'Labview' for that category, there are many others as well.
I write a fair percentage of LookML as a ratio of my total Looker data dashboard creation time. (It's maybe low-code or "somewhat easy code", but it seems pretty far from no-code to me. I'm still a fan.)
It depends a lot on what you consider “Low Code” and what your target is. Corporate users tend to look at Mendix, OutSystems, etc. because they are typically looking at building end—to-end solutions, and those give you front-end, APIs and ORMs (or, rather, abstract all of those away).
Most OSS or IoT folk will immediately point to Node-Red or n8n.io, which look similar but are actually quite different in focus (n8n is geared at hooking up third-party services at a higher level).
And then you have all the Web Automation stuff that has been with us forever since Yahoo Pipes went down.
I tried a bit OutSystems and its learning curve is incredibly steep for what they claim to be. It all starts with a specialized Eclipse based environment to install and to learn and only goes downhill from there. I don't know any business user able to use it, and for a regular programmer it's just another language/environment to learn. Not sure why it's a thing.
[Disclaimer: OutSystems employee here] I'm sad you find the experience bad. Eclipse or VSStudio integrations is part of our focus on being able to extend the platform to the max, but the need for it is only required when you are at a very specific use case.
I interviewed as a developer for a company that used outsystems. They said they had a lot of technology challenges, but were going to stick with outsystems. Their flagship app is rated less than 2.5 stars on Google Play, and needless to say, I did not pursue that position.
I'm a founder of Appsmith, mentioned here a couple of times. It is a good open source project for building anything from simple CRUD apps to complex internal applications quickly. Drag and drop UI to avoid writing HTML/CSS, connect with any datasource, and use JS to write custom logic when you need to.
Agreed. We need to improve a lot on the mobile, we've had our hands full with desktop browsers. About the unfixed iOS issue, was it an issue about the tabs widget on Safari? https://github.com/appsmithorg/appsmith/issues/3835
If not, could you tell me more so that I can create an issue.
I have used Retool to create a (well-received) PoC of an internal-facing admin app. I was pretty impressed, even though it had some rough edges.
Setting up UIs was relatively painless thanks to many out-of-the-box integrations, and it was surprisingly easy to implement auth/error handling/component dependencies.
The thing I liked the most is that it's not really a "no code" - you need to be technical to build good apps in this tool. However, that's where the power comes from - it simplifies the mundane development tasks and lets you focus on something more high level.
I wish the team would make it easier to consume your own, custom API however.
We allow you to build polished PWAs and now full web apps using data in your Google Sheet (and soon Excel and Airtable).
You can build logic and complex actions in Glide, without code or formulas. You can use plugins for more advanced custom features or to run your own code: https://www.glideapps.com/plugins
Glide is special when compared to other NoCode tools because we handle a lot more of the design for you–your Glide app is built with higher-level components, rather than you deciding where buttons go or how they look. Glide apps also tend to look much better than what you might design yourself.
I've been a Drupal developer for 15 years and started using it for it's low code abilities. In my opinion it still is an amazing low code platform which is extremely versatile. If you really need to customize, it's all there under the hood to hook into, override and extend.
I was waiting for someone to mention this as I scrolled from the top to bottom... I have been implementing Drupal consistently since v4.0 and despite it's increasing footprint, I haven't seen anything it really can't do as a CMS and web app platform.
No code on Drupal isn't exactly pretty, it requires a lot of CSS work to look good, but out of the box it's reliable and quite secure and robust. I've also implemented it in some of the most secure settings without problems over years. The ke to stability is scanning code prior to deployment for outbound requests, not building/adding what isn't needed or used, applying updates on time, and designing and maintaining a secure architecture (zero trust if data value warrants it).
I have built a lot of cool and mission critical things and sites with Drupal, and I'm very thankful for finding it years ago. v7.X is still my fave, the modules were so much more thought out and available for it, and still are. One of the best aspects of building with it is that the community never really turned into a money factory for modules like Wordpress did. As far as downsides, the hardest thing is debating other developers on why it should be used over other frameworks or instead of a custom app without burning up valuable project time. As a testament to Drupal's reach, it's implemented on many key web sites to this day, all you have to do is google it.
If you need to handle translations, users, permissions, revisions... and want to provide a good editor experience for building landing pages, (web)forms, etc. Drupal can be a good CMS. It won't be easy to start though.
Is Drupal a generic "no-code" solution? I don't think so. Drupal is a CMS and makes sense to use it if you want a website and manage its content.
I can vouch for Odoo and its Builder module. Odoo itself is open source, but there's a paid support model. We implemented an open source version for one of my clients and for business/production/inventory/accounting I'd say it's second to none. The free modules cover maybe 90% of common business cases, and the 10% left can be handled by duplicating one of the free modules and customizing it in Builder. Odoo is actively updated by large team in Belgium if I remember correctly, and their paid support is quite good. Odoo is Python atop Postgres, and uses a simple HTML/CSS/js for the frontend. Quite customizable, and can be a no-code solution if one uses Builder, which covers most use cases.
Having shipped an Odoo project this year, how is the developer experience for you?
It seems like Odoo forces developers into time consuming and frustrating workflows (for example using xpath + upgrade modules + possibly reload the server) that make developing for it a pain.
Or I might be missing something?
In development mode with inotify you get python hot code reloading. In development mode the XML (with a handful of exceptions regarding menus, and ir.actions) will be read from disk on refresh.
I do wish there was hot code reloading for the views (xml), but realistically if you’re making an app of any complexity the views in the backend are the small part and likely not where you are spending the majority of your time.
Website stuff is absolutely 100% stuck in the past at the moment, imho.
I’ve built a significant number of relatively sophisticated applications as internal tools and one public application (which is used daily by ~300 people) using Airtable, Stacker, typeform, carrd, twilio and integromat. I also use MagicBell, Infinity and ApproveIt (a slack approvals tool)
By “relatively sophisticated” I mean “would have cost me a significant amount in dev time to do this” and “have useful functionality that I can’t get out of the box elsewhere”.
Stacker is about $3000 a year; Integromat is about $1500;
Airtable is $240 per user per year and I think I’ve got six users (but you’d only need one to do what I’m doing);
Twilio is pennies - maybe $100 a year;
Carrd is insanely cheap - $49 a year I think;
Typeform is around $600 a year
So I guess I’m laying out somewhere in the region of $5250 a year or $450 a month.
That’s split across several projects.
The beauty is I can prototype an idea in minutes (ok, there was a learning curve, but now it’s super super fast) and then test and push to “production” in hours.
I rebuilt my primary application in about three hours recently, massively improving the user experience.
Readit News
1) BPM tools: build workflows & business processes. Tools like Appian, Pega, kissflow, etc
2) Form builders: jotform, typeform, etc
3) automation: Zapier, integromat, etc
4) Frontend builder - Consumer apps -> bubble, adalo, etc - Internal apps -> Retool, DronaHQ, Appsmith
5) App builder - Build core business apps (including backend, microservices , etc) - Outsystems, Mendix
6) Field force apps -> prontoforms, fulcrum, etc
7) Extension builder -> extension.dev
8) Test automation -> reflect.run
Disclaimer: I am part of DronaHQ.
I think a lot of people think of "no-code" tools as something exclusively for non-developers; things like Webflow and Bubble for creating apps. I don't think that's ever really been the case, Zapier being a great example of that, but there's been a lot more tools coming to market in the past couple years that actually work well as a way to replace tedious code-based workflows. i.e. Use Retool instead of building UIs for internal apps, use Reflect to build automated tests instead of Selenium, etc.
All of what I did with Zapier could have been implemented with a database, a bunch of scripting, and cronjobs, but using Zapier saved oodles of time and meant I was outsourcing the maintenance to their API integrators rather than to myself.
I was able to do so much with triggers and webhooks I can only imagine what is possible now with their more recent code blocks feature and adding better environment variables.
WordPress, Squarespace and Wix... As their userbases dwarf all other systems mentioned
Not sure where it would fit in, but Looker adoption is crazy. It's deeper though: no-code data modeling
9) Online DB tools -> Quickbase, Airtable, Knack, etc 10) Website builder -> webflow, softr, etc
Surprised I haven't come across one yet...
Most OSS or IoT folk will immediately point to Node-Red or n8n.io, which look similar but are actually quite different in focus (n8n is geared at hooking up third-party services at a higher level).
And then you have all the Web Automation stuff that has been with us forever since Yahoo Pipes went down.
I'd advice you to take another look. For example take a look at this tutorial (https://www.outsystems.com/training/courses/124/reactive-ui-...) and see everything you can build with OutSystems.... without Eclipse :)
I opened an issue quite a while ago about a very basic component being completely broken on iOS and it still hasn't been fixed.
If not, could you tell me more so that I can create an issue.
Setting up UIs was relatively painless thanks to many out-of-the-box integrations, and it was surprisingly easy to implement auth/error handling/component dependencies.
The thing I liked the most is that it's not really a "no code" - you need to be technical to build good apps in this tool. However, that's where the power comes from - it simplifies the mundane development tasks and lets you focus on something more high level.
I wish the team would make it easier to consume your own, custom API however.
Deleted Comment
We allow you to build polished PWAs and now full web apps using data in your Google Sheet (and soon Excel and Airtable).
You can build logic and complex actions in Glide, without code or formulas. You can use plugins for more advanced custom features or to run your own code: https://www.glideapps.com/plugins
Glide is special when compared to other NoCode tools because we handle a lot more of the design for you–your Glide app is built with higher-level components, rather than you deciding where buttons go or how they look. Glide apps also tend to look much better than what you might design yourself.
I've been a Drupal developer for 15 years and started using it for it's low code abilities. In my opinion it still is an amazing low code platform which is extremely versatile. If you really need to customize, it's all there under the hood to hook into, override and extend.
No code on Drupal isn't exactly pretty, it requires a lot of CSS work to look good, but out of the box it's reliable and quite secure and robust. I've also implemented it in some of the most secure settings without problems over years. The ke to stability is scanning code prior to deployment for outbound requests, not building/adding what isn't needed or used, applying updates on time, and designing and maintaining a secure architecture (zero trust if data value warrants it).
I have built a lot of cool and mission critical things and sites with Drupal, and I'm very thankful for finding it years ago. v7.X is still my fave, the modules were so much more thought out and available for it, and still are. One of the best aspects of building with it is that the community never really turned into a money factory for modules like Wordpress did. As far as downsides, the hardest thing is debating other developers on why it should be used over other frameworks or instead of a custom app without burning up valuable project time. As a testament to Drupal's reach, it's implemented on many key web sites to this day, all you have to do is google it.
Is Drupal a generic "no-code" solution? I don't think so. Drupal is a CMS and makes sense to use it if you want a website and manage its content.
Having shipped an Odoo project this year, how is the developer experience for you?
It seems like Odoo forces developers into time consuming and frustrating workflows (for example using xpath + upgrade modules + possibly reload the server) that make developing for it a pain. Or I might be missing something?
In development mode with inotify you get python hot code reloading. In development mode the XML (with a handful of exceptions regarding menus, and ir.actions) will be read from disk on refresh.
At work we use Dockerized Odoo based on a [Doodba](https://github.com/Tecnativa/doodba). We use [click-odoo](https://github.com/acsone/click-odoo) to manage module upgrades (amongst other things). Dev workflow is largely managed through [invoke](https://www.pyinvoke.org/) tasks.
I do wish there was hot code reloading for the views (xml), but realistically if you’re making an app of any complexity the views in the backend are the small part and likely not where you are spending the majority of your time.
Website stuff is absolutely 100% stuck in the past at the moment, imho.
By “relatively sophisticated” I mean “would have cost me a significant amount in dev time to do this” and “have useful functionality that I can’t get out of the box elsewhere”.
So I guess I’m laying out somewhere in the region of $5250 a year or $450 a month.
That’s split across several projects.
The beauty is I can prototype an idea in minutes (ok, there was a learning curve, but now it’s super super fast) and then test and push to “production” in hours.
I rebuilt my primary application in about three hours recently, massively improving the user experience.
The costs against the business value are trivial.