Which Stacks / What do they do for your site?? It might be possible, depending on the stack, to recreate its function in Elements easily. Others, not so much.
It’s alright. I’ll remain with Classic for now and won’t switch to Elements. I’ve invested too much to start afresh. I’m really concerned about the future, especially with so many Stacks developers closing their stores. If Realmac eliminate Classic and leave all the users in the dust, I will learn my lesson and never buy a website builder tool ever again. If I need to learn something new I will go straight to AI, lovable, Cursor, or whatever points to a real future of technology. I’ve been a loyal Realmac costumer since day one, so I hope they don’t disrespect us like that.
Sure I get it… but if you told us what features/stacks you think are missing from Elements we could build them into the product. At least then you’d know you could move to Elements should those third-parties fall by the wayside.
Ok, I will make a list of my addons and send them to you. No problem.
Great, looking forward to seeing it!
I recall in the early beta period a user had similar concerns - they had a stack which allowed the display of data from Google Sheets. Within an hour - and using ChatGPT to write some custom code - the stack function had been created as a component. I posted the original code and I have ZERO experience writing web code. Within days the same user was posting modified versions will all sorts of additional functionality.
I believe Elements is sufficiently powerful that pretty much any Stack will be able to be replicated. RealMac are throwing their arms open to assist people, you might find the ‘transition from classic’ impediment disappears.
I’m really concerned about the future, especially with so many Stacks developers closing their stores. If Realmac eliminate Classic and leave all the users in the dust, I will learn my lesson and never buy a website builder tool ever again.
The World Wide Web on top of the Internet is a living organism. It has evolved over time and will continue to. Even the most basic HTML components have changed some over the years, and once we start talking about Javascript and more, then the changes have been constant and often disruptive. Security concerns forced us all to go to certificate-based HTTPS: instead of HTTP. The list of changes a Web site owner has to deal with is ginormous and growing.
As a developer (both programs and Web), I’ve had to give up old tools and move to new ones with regularity over the past five decades. It’s easy to blame the tool (“they should have just updated it”), but behind the scenes conceptual and baseline things also trigger the need to start over.
Most Classic sites are built using tools that sit on top of tools that sit on top of a tool (Stacks that sit on a platform/foundation that sit on the application). If any one of those things starts to become an issue, then the site built that way ends up with problems. Stacks developers have been slowly dying off, and the foundations used by many have turned out to be some of the least used on the Internet, so themselves likely have a short future ahead.
I’d say your implicit anger at Realmac is misdirected. Consider Realmac’s problem if the Stacks developers slowly went away and the foundations on which many of those Stacks also became unmaintained?
Elements is a much better approach for the foreseeable future. It doesn’t depend upon Stacks creation, and the foundation it primarily uses is arguably the most popular and growing one. Realmac made a really good choice here that’s likely going to make those that pick Elements as their tool a long-term one that can weather the next changes in Webdom.
Meanwhile, as Dan keeps saying, Classic isn’t going away and it’s being well maintained. If it currently does what you want, then there’s no reason to move from it. But for someone who’s using Stacks and Foundations in Classic, the real problem is whether those are being maintained, and that’s out of Realmac’s control.
If I need to learn something new I will go straight to AI, lovable, Cursor, or whatever points to a real future of technology.
This is a complete misunderstanding of AI. The LLMs that power AI are backward looking not forward looking. If you run a business solely from AI, you are stuck with what has been done and not looking at what’s going to be done next. Again, the World Wide Web is a living organism, so it’s going to continue to change.
Absolutely. That’s precisely my point. Lovable, Cursor, and similar AI tool alternatives represent the future, while options like Elements, Blocs, and Pinegrow are falling behind. This is the reality we face at present.
I respect your perspective, but I firmly believe that AI is the cornerstone of the future. Currently, it has transformed my company in remarkable ways. With innovative tools like Cursor and Windsurf, the web is poised to become more vibrant and dynamic than ever before.
I didn’t say that AI is not useful, I said it was backward looking. The way current AI works is by building models on existing things, not inventing new things. If I were to invent a new Web function, it would take time for that to promulgate into AI answers.
This leads to another thing that is still not commonly understood: AI will follow the same path as search, eventually leading to something like a monopoly, duopoly, or most likely, an oligopoly centered on existing successful US tech companies and a path to business model payback (e.g. Amazon, Apple, Google, Meta, Microsoft). The necessity of keeping your AI engine up-to-date leads to constant and massive indexing, which requires huge server farms. We’re talking billions a year in investments, with, at present, no clear business model to make those investments profitable for a new, purely AI company. While I hope I’m wrong about that, if you follow the money, it all leads to the same entities having a clear advantage in keeping their AI models up-to-date.
It’s sort of the Library of Alexandria problem: in order to be the repository of all knowledge, you must keep up with the knowledge.
I think you might be missing an important distinction in how modern AI (specifically LLMs) actually works.
You’re right that training starts with existing data — but that doesn’t mean the model is just “looking backward.” The training creates a probabilistic system of relationships between concepts, language, logic, and problem-solving. Once the model is trained, it isn’t simply reciting from a library — it’s generating new combinations of ideas, reasoning paths, and outputs that never existed in the training data.
Think of it like this:
-
The Web at first was also “backward looking,” since it was just publishing digitized versions of newspapers, catalogs, and books. But it quickly evolved into an ecosystem where new functions (search engines, social networks, streaming, cloud computing, etc.) emerged on top of the foundation.
-
LLMs are at a similar inflection point. They’ll be the substrate on which new tools, applications, and even entirely new types of computing interfaces are built.
On the monopoly/oligopoly point — you’re right that the infrastructure costs are massive, and that favors existing tech giants. But just like with the Web, the underlying platform being centralized doesn’t prevent massive waves of innovation on top of it. Google owns the search index, but it didn’t stop Shopify, Uber, Airbnb, or Figma from existing. AI will follow the same path: a few companies may control the models, but the real explosion will be in what gets built with them.
The “Library of Alexandria” analogy isn’t quite right. LLMs aren’t repositories of facts — they’re engines for pattern recognition and generation. They don’t need to contain all of human knowledge to be useful, they just need a large enough base to generalize, reason, and combine things in novel ways.
That’s why AI isn’t just search 2.0 — it’s a new computational layer. Like the Web, it started by imitating what already existed, but it will increasingly invent new categories of functionality that weren’t possible before.
This is - I think - the reason for AI to hallucinate
I like Willison’s latest analogy for LLM: a lossy encyclopedia. I believe that you see that here with the Elementsbot, for example. Willison’s exact example of why he suggests that analogy applies 100% to programming (Web or app) with AI: unless you provide a correct foundation, you can’t expect it to act correctly. Most of the so-called “vibe” programmers I’ve talked to have already discovered this: if you’re looking for AI to create new code you’d better give it a strong foundation to work from, not just let it use all code its seen and come up with something.
Again, I’m not saying AI won’t be successful or won’t have uses. Just like I didn’t say that the Internet wouldn’t be successful or have uses back as it was being developed (I was in Silicon Valley and involved at the time). However, be careful of thinking that new tech that comes along like AI will do what you think it will do for you: money controls what AI will do. Google and Meta, in particular, are going to try to force it into something that benefits them, not you.
I will change just the company names: “However, be careful of thinking that new tech that comes along like AI will do what you think it will do for you: money controls what developers will do. Realmac and Blocs, in particular, are going to try to force it into something that benefits them, not you.” see? Arguments are logical solely in the perception of the observer.
You seem overly angry about, well, not much.
Classic is still available, still supported. It is a mature product dependent upon third-party developers (who charge you additional money). What exactly is it you want Realmac to do for you that they aren’t doing?
Meanwhile, the Web has moved on from the old style development tools. Far away. For Realmac to stay in the Web Page Tool business they’re not just competing against other desktop products (some old, some new), but they’re competing against the AI-will-design-you-something tools that have started to emerge.
You are a consumer of something, in this case Web Site Tools. You pretty much have a full array of options in front of you, one of which is Elements.
However, I’ll point out one reason why some of us are defending Realmac and baffled by your repeated assertions: Elements has been one of the most customer-driven design-outs I’ve ever seen for a piece of software. Those of us that opted in early have been suggesting things and getting them back over time. You could have paid US$150 and joined those of us who are prodding Dan and team to design a remarkable tool that does what we need done, but you’ve essentially come to their site and attack them for not doing something you haven’t really specified they should do that they aren’t. Again, Realmac seems committed to supporting Classic until that well runs dry.
You seem to think AI is going to do everything you need it to for the future of your Web work. I invite you to try that. I have. First, it costs more—you’re not going to get “free AI designing” for very long or for very sophisticated projects. The current minimum to stay current and get enough computation seems to be US$240/year. While I can get pretty sites from AI, what I can’t get is real control over the underlying code design, and worse still since I run informational sites, my content is sucked into these big players who will then create competition for me.
I believe you also have an overblown concept of “invent” as it comes to the AI engines. I actually don’t see them inventing. Your “new categories of functionality that weren’t available before” doesn’t resonate with me, as I’ve not seen them. Give me an example of a Web design that’s new functionality that was invented by AI.
Again, I’ll ask you: what specifically do you want Dan and team to do for you?
This appears to be a difference in interpretation. I believe my previous explanation was comprehensive, so I’ll leave my point there. I don’t see further discussion on this specific point as productive.
Really? I will Ignore this speculation as well.
Just to keep Rapidweaver Classic under development. That’s all. A very simple suggestion.
This is a fascinating discussion. My intention was never to defend any company fanatically, but rather to offer a perspective on their marketing strategy and future developments. I’m simply a user sharing an opinion, not an advocate. Thank you to those who engaged thoughtfully. I’ll leave the conversation here.
It’s amazing how this discussion has evolved into other issues, such as AI and other topics. I posted initially to get advice. What I’ve decided as if no one is in agreement on web design softwares and different options. What I like about Elements so far is that there is no time limit on trying it out. If there was a limit (seven to thirty days), I would have given up. One of the best comments was “It’s a different way of working.” Boy, is that accurate. I’m still working with it, but just simply do not find it to be very intuitive - at all. Even with the video tutorials, how it all works together, and more importantly the tools to modify the (forgive my wording here) elements on a page are difficult to find. I’ve looked at probably 12-15 different softwares, including Elements, and the intuitive interface does not exist, in my opinion. I can start a web page in some of them, but it falls apart quickly when I try to either move things, change a font size or style, etc. I’m going to keep trying, though, hoping that it will come together and make sense, at some point.
Absolutely true, and has been persistently true since the beginning. The Internet and the things that sit somewhere on top of it like HTML is open, extendable, and at a design level ultimately non-restrictive. That has bred a lot of creativity and differences in approaches.
This is a result of what I just wrote. Any Web design tool these days is like Photoshop: lots of possibilities, somewhat organized, but fairly non-restrictive in order and use.
I don’t disagree here, but some of that is just in step with what the tool wants you to do. For instance, in Elements, you generally should be moving things in the Page Layout panel. But that really only works well if you name the things as you’re putting them into the layout. If you end up with 40 “Container” bits, you’re not always sure which Container is doing what. On the other hand, if you named that Container “ProductX Features,” you’d now what it is you’re moving. The outline approach used in Page Layout takes a bit of getting used to, but once you’ve learned it, it is remarkably helpful in helping you keep things organized as to where they should be.
In terms of changing a font size or style, that’s a little different, because there are two levels of that. Generic and site-wide changes in Elements should be done in the Theme Settings. Specific words or phrases you want to be different need to be in the Inspector.