Amazon's Honeycode No-Code App Builder Is No-More (honeycodecommunity.aws) 36
"Amazon launches cloud service to help non-coders build apps," read the 2020 headline at CNBC — both mobile and web applications.
But long-time Slashdot reader theodp has the rest of the story: Customers have told us that the need for custom applications far outstrips the capacity of developers to create them," Amazon Web Services explained as it jumped on the low-code and no-code bandwagon in 2020...
But just three years later, Amazon posted a "Dear Valued Customer" letter announcing it's pulling the plug on Honeycode at the end of February: "To our valued customers: Thank you for participating in the Amazon Honeycode beta program... After careful consideration, we have made the decision to end the beta service, effective February 29, 2024. Starting today, we are no longer accepting new customer sign-ups to the Honeycode beta. However, as an existing customer, you will be able to use Honeycode and your Honeycode apps as normal (and add team members to your existing account) until February 29, 2024, when the service will be discontinued. After this date, you will no longer be able to use Honeycode or any of the apps you created in Honeycode."
Amazon advises the "valued customers" it's leaving stranded to use Honeycode's "Export Data" option ("a handy way to get your info organized into a CSV file(s)", although "formulas will not export"). They also warn that "We will retain your data until April 29, 2024. If you do not take any action, your data will be deleted on April 30, 2024."
Amazon adds that the spirit of Honeycode (RIP, 2020-2024) will live on in its other products: "We are incorporating lessons from the Amazon Honeycode beta into current services, and remain committed to supporting no/low code services including Amazon SageMaker Canvas (2021-?), AWS Amplify Studio (2021-?), and AWS AppFabric (2023-?).
But long-time Slashdot reader theodp has the rest of the story: Customers have told us that the need for custom applications far outstrips the capacity of developers to create them," Amazon Web Services explained as it jumped on the low-code and no-code bandwagon in 2020...
But just three years later, Amazon posted a "Dear Valued Customer" letter announcing it's pulling the plug on Honeycode at the end of February: "To our valued customers: Thank you for participating in the Amazon Honeycode beta program... After careful consideration, we have made the decision to end the beta service, effective February 29, 2024. Starting today, we are no longer accepting new customer sign-ups to the Honeycode beta. However, as an existing customer, you will be able to use Honeycode and your Honeycode apps as normal (and add team members to your existing account) until February 29, 2024, when the service will be discontinued. After this date, you will no longer be able to use Honeycode or any of the apps you created in Honeycode."
Amazon advises the "valued customers" it's leaving stranded to use Honeycode's "Export Data" option ("a handy way to get your info organized into a CSV file(s)", although "formulas will not export"). They also warn that "We will retain your data until April 29, 2024. If you do not take any action, your data will be deleted on April 30, 2024."
Amazon adds that the spirit of Honeycode (RIP, 2020-2024) will live on in its other products: "We are incorporating lessons from the Amazon Honeycode beta into current services, and remain committed to supporting no/low code services including Amazon SageMaker Canvas (2021-?), AWS Amplify Studio (2021-?), and AWS AppFabric (2023-?).
NoCode is a scam (Score:5, Interesting)
At this point is should be evident to everyone that this was all wishful thinking by people that are clueless about computing
Re: (Score:2)
So, what's new? NoCode has been a scam since it was introduce with tools like Power Builder and Microsoft Access in the 1990ies. They didn't call it "NoCode" back then but it basically was the same.
"Low Code", the other buzz word in this context, is also nothing new, that used to be called Rapid Application Development (RAD) in the 1990ies.
But of course we keep on reinventing the wheel, because there is always some sucker who will shell out money for it.
Re: (Score:2)
This scam has being going on since 1981:
https://en.wikipedia.org/wiki/The_Last_One_(software)
Re:NoCode is a scam (Score:4, Interesting)
I don't understand why would people call LowCode/NoCode a scam. My workplace develops internal tools with one of them, it works fine. It's expensive (we pay per each user of our tools) but it's definitely not a scam. The alternative is to pay for that at all, hire more people instead and develop our management portal in php or something like that. The LowCode option was selected because the budget does not change but the products are delivered faster.
Re: (Score:3)
I don't understand why would people call LowCode/NoCode a scam.
LowCode/NoCode is not a scam. But it is very limited in scope. You can only do what the creators envisioned. It's not general purpose.
We use LC/NC to enable sales & marketing to generate their own reports. That works okay bcoz that's what the tool was designed for.
The scaminess is when LC/NC is pushed as a replacement for real development.
Re: (Score:1)
That's probably the issue. Most main stream programming languages are effectively LowCode in that a lot of development with these languages is essentially glorified wrappers of imported functionality developed over decades. (With Python even allowing you to import functionality from the future :S) Any new offering likely can't compete with that, and needs to be directed at a narrow market. Creating wrappers all the way down won't work because of too much abstraction, rather li
Re: (Score:2)
Most main stream programming languages are effectively LowCode in that a lot of development with these languages is essentially glorified wrappers of imported functionality developed over decades.
While that has made "traditional" programming much easier (and the dream of 'low/no code' less impactful), low/no code does not apply. Generally low/no code means those tools that purport to let you draw flow charts and *boom*, you have a "program". There may be some that stray from the flow chart paradigm, but if you are describing the logic using text, even if very simple text, that's just considered "coding".
I was party to a partner company that had a 'low code' product to demonstrate. To their credit
Re: NoCode is a scam (Score:3)
Your entire platform is limited to a single vendor, you'll always need to train new hires, if the vendor ever goes belly up or simply discontinued that product then you're really fucked. The vendor can eventually raise the fuck out of your license fees, and there's not much you can do about it because their salesmen know that your business is hopelessly dependent upon them.
On the other hand, you could go with Python (that language literally only took me a single day to learn, though admittedly I'd already b
Re: (Score:1)
It's a legal scam. Newbies don't understand long-term and maintenance cost, and these tool makers milk this gap. There are a lot of product gimmicks that take advantage of clueless newbies.
Insurance has a common one: people often buy too much insurance out of fear, when the probability numbers don't support it.
Capitalism is largely the art of milking the naive.
Re: NoCode is a scam (Score:2)
Was Power builder really billed as something non-developers could use? Making a UI was a drag and drop affair but I'm pretty sure I remember writing code to make it actually do anything.
Re: (Score:2)
To the uninitiated, C-Level work also looks like playing golf, so why haven't we tried the thing with the H1B monkeys there? Maybe it works better at that level?
Re:NoCode is a scam (Score:5, Interesting)
At this point is should be evident to everyone that this was all wishful thinking by people that are clueless about computing
Unfortunately some of those people are "in the biz". I remember we had a Windows admin who was adamant that then-new Sharepoint (or whatever it was called back then) was going to completely eliminate the need for web coders - non-technical people would be magically able to write complex, secure data-driven web apps.
I should note that our other (more senior) Windows admin at the time did not share his naïveté.
Re: NoCode is a scam (Score:3)
Re: NoCode is a scam (Score:5, Funny)
No it's not! They promised you no code and in a few months you'll literally have no code and also no data. Seriously people, learn to read!
Re:NoCode is a scam (Score:4, Insightful)
People who don't code see what to them is incomprehensible text and think the inscrutability of the text is the reason they can't code.
Then they come up with yet another "no-code" "solution" and are mystified that they STILL can't code. That would be because the language was never the problem.
Meanwhile, because they can't code, they don't understand that the verbose and imprecise "no-code" actually makes it harder for people who actually do know how to code to do anything useful.
Re: (Score:1)
Bullshit.
Between ServiceNow, Microsoft and SalesForce there are plenty of useful things built everyday with varied combinations of no-code, low-code and plenty-of-code.
Re: (Score:2)
It's mostly a scam, but *can* work in very specific domains. Problem is finding a domain that can be:
-Adequately modeled to be covered with 'no-code' simplicity, with hardly a chance of having to 'jump to code'.
-Yet still open ended enough to warrant a 'no-code' assemblage versus "just a purpose built app"
-Be a big enough problem set to warrant making a product.
Examples are slim, but if you've messed with many 3D modeling tools, but that is one area where I've *long* seen 'no-code' design applied and used.
It is a double speak (Score:1)
No-code is a misunderstanding (Score:5, Insightful)
Creating complex systems that work well across a wide range of conditions is hard
It doesn't matter if you express the system is code, specification, text prompt, graphic doodle or any other strategy
The programming language is not the hard part. Understanding how the system works is much harder, especially when dealing with unexpected inputs
Re: (Score:3)
I've worked with numerous groups - usually involving faculty) - developing different web-based tools. Invariably, getting them to even think about edge cases and non-conforming inputs is incredibly hard. Time after time, I've had faculty members tell me "we'll just manage things so those non-conforming situations don't come up" - despite my assurances that this isn't my first rodeo, I probably know how faculty think better than they do, and faculty members are masters of non-conformity.
Re: (Score:1)
... despite my assurances that this isn't my first rodeo, I probably know how faculty think better than they do, and faculty members are masters of non-conformity.
Note that I don't use anything like that sort of phrasing when explaining it to faculty - I can be diplomatic when it's necessary, and know how to stroke egos. :-)
Re: (Score:2)
Re: (Score:3)
I agree. However when I read responses to these various no-code projects, I cannot help but think back to the beginnings of cell phones. Waaay, back in the day the idea of what we have today, ubiquitous personal cell phones, was frequently 'down-modded' as an impossible fantasy.
"That's crazy, a car phone uses 30 watts, everyone is going to carry around a huge battery?"
"That's crazy, there are already pay phones and phone booths everywhere."
"That's crazy, who would pay $1 per minute to talk on the phone?"
Re: No-code is a misunderstanding (Score:2)
Of course they will, but the question is when. 2 years? 30? 100? Nobody knows.
Re: (Score:2)
So I think it's entirely plausible, that no-code applications will be part of our future.
They are part of our present. You might not like Salesforce (which speaks well of you) but tons and tons of business applications have been built in it with no- or low-code (you occasionally have to write a few lines for a matching criteria but drag and drop all the major business logic.)
But the children! (Score:2)
[sobs] They need to nocode. How will they nocode when they don't have the CS in every classroom? [wails] warrgh. Oh the price that programmers charge, whoops, I mean THE CHILDREN! The children are falling behind!
Anything not Turing Complete is a scam (Score:1)
So! (Score:2)
Hopeful (Score:3)
You see, I think its ability to create a (albeit basic) cross-platform app, arguably with the push of a button, is a great idea. To me it seems non-trivial to make such a front end, let alone having to update it all the time.
The problem I had with Honeycode was the way they grouped users together. My details are foggy, but I think it was more of a collaboration tool, meaning it was not possible to put all of your customers (IoT, in my case) in it because of the size limitations and the ability for strangers to access others' data. Too bad...
Re: (Score:2)
To me it seems non-trivial to make such a front end, let alone having to update it all the time.
You'd be surprised how easy front end work is if you prod it a little.
Re: (Score:2)
Do have any experience or recommendations for decent cross-platform mobile development for a tiny dev. team? There are so many
Number one NoCode tool (Score:5, Interesting)
#1 NoCode tool is and will continue to be Excel. That's where lawyers, brokers, mathematicians and business analysts who didn't learn to code write their domain data types, complex data flows and moderately complex queries, without the need for a caste of priests programmers gatekeeping the knowledge of the sacred code and building their tools for them.
The main problem with NoCode is that developers utterly misunderstand what end users need to build digital tools (call them 'apps' if you will), so they keep writing NoCode tools as glorified webforms that still depend on a full-code backend, instead of improving Excel-like environments with all the tools and utilities that developers use for writing robust code (version control, tests, refactoring, automated data persistence, advanced debugging with breakpoint and variable inspectors...), so the apps that end users create become fragile and unmaintainable.
I have hig hopes on reactive notepads, they may be the nocode tool that finally closes the gap as the perfect development environment for non-programmers.