AWS App Studio Promises To Generate Enterprise Apps From a Written Prompt (techcrunch.com) 36
Amazon Web Services is the latest entrant to the generative AI game with the announcement of App Studio, a groundbreaking tool capable of building complex software applications from simple written prompts. TechCrunch's Ron Miller reports: "App Studio is for technical folks who have technical expertise but are not professional developers, and we're enabling them to build enterprise-grade apps," Sriram Devanathan, GM of Amazon Q Apps and AWS App Studio, told TechCrunch. Amazon defines enterprise apps as having multiple UI pages with the ability to pull from multiple data sources, perform complex operations like joins and filters, and embed business logic in them. It is aimed at IT professionals, data engineers and enterprise architects, even product managers who might lack coding skills but have the requisite company knowledge to understand what kinds of internal software applications they might need. The company is hoping to enable these employees to build applications by describing the application they need and the data sources they wish to use.
Examples of the types of applications include an inventory-tracking system or claims approval process. The user starts by entering the name of an application, calling the data sources and then describing the application they want to build. The system comes with some sample prompts to help, but users can enter an ad hoc description if they wish. It then builds a list of requirements for the application and what it will do, based on the description. The user can refine these requirements by interacting with the generative AI. In that way, it's not unlike a lot of no-code tools that preceded it, but Devanathan says it is different. [...] Once the application is complete, it goes through a mini DevOps pipeline where it can be tested before going into production. In terms of identity, security and governance, and other requirements any enterprise would have for applications being deployed, the administrator can link to existing systems when setting up the App Studio. When it gets deployed, AWS handles all of that on the back end for the customer, based on the information entered by the admin.
Examples of the types of applications include an inventory-tracking system or claims approval process. The user starts by entering the name of an application, calling the data sources and then describing the application they want to build. The system comes with some sample prompts to help, but users can enter an ad hoc description if they wish. It then builds a list of requirements for the application and what it will do, based on the description. The user can refine these requirements by interacting with the generative AI. In that way, it's not unlike a lot of no-code tools that preceded it, but Devanathan says it is different. [...] Once the application is complete, it goes through a mini DevOps pipeline where it can be tested before going into production. In terms of identity, security and governance, and other requirements any enterprise would have for applications being deployed, the administrator can link to existing systems when setting up the App Studio. When it gets deployed, AWS handles all of that on the back end for the customer, based on the information entered by the admin.
Oh man! (Score:4, Insightful)
Re:Oh man! (Score:4, Interesting)
Re:Oh man! (Score:5, Insightful)
Re: (Score:2)
This isn't anything new. How many consultants made their retirement plans from just this? Follow the chains of Oracle, DB2, or the sillyness of Visual Basic, and a hundred+ other big money makers on this continent, and others.
Nothing new here, just a fun way to make money!!
Re: (Score:2)
Unpopular opinion: Programming languages and tools should make things simpler and easier, not more difficult and error prone.
Say what you want about VB, but it was wildly successful for a reason. I could hire a kid in high school and have them productive in just a few hours. An experienced developer could do more in VB in an afternoon than they could do in VC in a week. It was RAD.
I don't know how many people here remember the transition from DOS to Windows, but it wasn't nearly as smooth as history rem
Re: (Score:2)
Don't see hate in what I wrote. See business plan, monolithic and eventually, entropy with long retirement plans. VB is not alone. It was good for the time, and like AI-generated goo, seemed like a good idea at the time. But it's really a business plan shot across the bow of a dozen silo-like development lines.
Re: (Score:2)
We're going to be inundated with shit apps that don't do anything they're supposed to do... so I guess that means nothing will change.
Sure it will! All the shitty apps will now have a shiny "Powered by AI" badge on them. YAY! PROGRESS!
Re: (Score:3)
it's not really that english is too vague.. it's that you need lots and lots of clearly defined words to have unambiguous specifications.
Not one sentence.
Yep, haven't laughed this hard, since the last article about one line of code to do
Re: (Score:2)
True, that's why we have Z
Re:Oh man! (Score:4, Insightful)
Don't forget that this is for Enterprise Quality [codinghorror.com] software. I can think of no other application better suited for the miracle of generative AI than that.
If you're not familiar with developing software at this level, you might want to check out Enterprise Quality FizzBuzz [github.com]. It's an absolute masterwork, distilling the essentials of Enterprise Quality development down to a tiny example, with all the elegance of an Enterprise Quality mathematical theorem.
anything even nominally complex.
The true power of Enterprise Quality Software is that domain complexity, which might believe to be unavoidable, completely vanishes behind ever-growing code complexity. Code complexity is no problem, of course, as Enterprise Quality Software leverages the many Enterprise Quality tools designed to manage the complexity introduced by those same tools.
Re: (Score:1)
Re: (Score:2)
It is even better: They want to specify something quite complex using a _small_ amount of _simple_ English!
My guess is they know they are blatantly lying and promising something they cannot deliver, bit who are the customers for this scam?
Credible... (Score:5, Insightful)
In my experience, "Enterprise Apps" are unfathomable garbage anyway, right up AI's alley.
Disappointed, Not Surprised (Score:2)
Re: (Score:2)
Every CEO will jump on that particular chatbox (Score:5, Funny)
- Alexa, make software that makes money ...
- Sorry Dave, I need more details than that
- Make software that makes a LOT of money
- Please, be more specific
- Make simple software that makes a lot of money for the shareholder
- I can't work with these specification. Please restate them.
- Make simple software that replace employees
- The software will require a supercomputer that costs $10m. Do you want to continue with the compilation?
- Make simple software that only requires cheap employees
Clippy could do better (Score:5, Funny)
Decimate that biosphere for less! (Score:5, Interesting)
Like a good nerd, I've been dreading the AI takeover since Terminator. But sci-fi has failed to predict the future once again. I never once heard a cautionary tale that predicted that humans wouldn't even recognize the advent of true AI because it would come buried under a stack of junk-mail brochures and pufferied-up press releases for crap like this.
The lesson I'm slowly learning from my life, I guess, is that any given future time tends to end up being just as bad as I expected, but not in the way I expected, and for sure it's always way, way stupider than expected. Possibly stupider than I can expect.
Re: (Score:3)
Cynicism is the only rational world view. Expect the worst, and all surprises will be good ones.
"A pessimist says the glass is half empty. An optimist ss the glass is half full. A cynic points that that either way, it's the wrong half."
will make bloated code that runs up your aws bill! (Score:3)
will make bloated code that runs up your aws bill! and has lots of AWS only hooks in it!
Re: (Score:2)
Sounds like an improvement
Re: (Score:3)
And Amazon will be declared blameless in the EULAs for any screwups by the code for which you asked.
Indemnification? (Score:2)
Read the fine print to discover that they say it has no warranty, no terms of service, and no responsibility for anything of any description now or in the future.
Don't be surprised when you discover that if you say try to sue them or say anything bad about them they can sue and automatically collect from you!
We'll call it.... (Score:2)
...a compiler!
Re: (Score:2)
Re: (Score:2)
"When it gets deployed, AWS handles all of that on the back end for the customer, based on the information entered by the admin."
That is 100% locked up with Amazon.
The Holy Grail (Score:5, Interesting)
Re: (Score:2)
Indeed. Or simplified, you cannot make something from nothing. If true, this would essentially be similar to buying mystery boxes.
Re: (Score:3)
This idea re-surfaces every decade, it was literally the marketing claim of COBOL ( COmmon Business Oriented Language ). Along with all the graphic programming languages like Visual Basic. There is a basic fact of information system entropy that you can't get something complex from something 'simple'. It might get you 95% of the way their, but will never competently handle all the edge cases which are 95% of the effort. Also, topology ( interactions ) increases exponentially with complexity. Human's sensory and cognitive interface for software is pretty much one or two dimensional, current distributed federated systems are n-dimensional, we don't even have a very nuanced human spoken language or graphical visualization techniques for time - see ' Allen's interval algebra' https://ics.uci.edu/~alspaugh/... [uci.edu] ( the closest we get is maybe orchestral scores and dance choreography ).
The only real difference this time around is in perceptions. C Suites are hyped up on AI everything, because they're being told by marketing dreamers that AI will replace most of these people's employees. And now, they'll replace programmers too! WITH SUGAR AND SPRINKLES AND FAIRY DUST AND UNICORNS! And not only that, it'll work so much faster, without complaint, and never ask for a raise, and never need healthcare or vacations or . . . .
This is the fantasy, and it's being pushed and bet on harder than a su
What they say is seldom what they want. (Score:3)
Re: (Score:2)
Sometimes the biggest challenge in a software project is getting to what the stakeholders need
I can see any attempt to get AI driven app building to end up like the refrain in the epic "Night before Implementation" poem (see below) - "it's just what I asked for but not what I want!"
The Night Before Implementation
Twas the night before implementation and all through the house,
Not a program was working, not even a browse.
The programmers hung by their screens in despair,
with hopes that a miracle would soon be there.
The users were nestled all snug in their beds,
while visions of inquires danced in their h
The hard part isn't the programming (Score:4, Insightful)
It's the requirements.
A few years ago, the executives at my company contracted with a third-party development firm to build them some dashboards. They explained that they wanted a series of dashboards to display information about the health of the business. "Can you build that by January?" they asked. "What exactly do you want on these dashboards?" the vendor replied. "We don't know, we just want some dashboards. So can you get them done by January?"
I have people skills! (Score:3)
Ports (Score:2)
Can you tell it to port existing software? Say, to Linux?
Obvious nonsense (Score:2)
The prompt has to specify the requirements. Hence the result cannot be fundamentally more complex than the prompt.