Scrap scrap scrap

Haha, yes, but Stuart I had to buy all of your stuff to work out how to do it! Seriously, though, what I realise is that the Stacks ‘mindset’ has conditioned us to think of RW as an environment to drop in ready-mades. What we now have, though, are toolkits to make all sorts of really interesting things, especially when we include some CSS and maybe some simple JS too. The change is really from ‘what is there a stack for?’ to ‘how can I make this?’

I don’t think there is much of a future for RW as code-less Lego — that market is under attack from the online sitebuilders like Wix and Squarespace, from the WYSIWYG apps like Blocs, and from Wordpress. Where RW can really shine, though, is as a construction kit. For instance RW + Source Grids is a really good environment for building and managing intricately responsive Grid based sites — it would be crazy difficult to do this in something like Blocs, and much less structured if we are coding by hand. In RW, though, I can have pages like accordions with colour-coded nested grids, which I can open up and drill into, and close back out again, and this is actually a really powerful way of structuring one’s build. And I can build my own gridded components (which can be partials or externals too) and drop them into the master grid. If they are set up to use flexible measurements (auto, %, fr), they will size themselves to fit any grid item they’re dropped into.


I could not agree more about how much ease and potential power that Source allows and when you understand what can be built with the Source grids, a whole creative world opens up. Stacks templates effectively allows the creation of your stacks and that is the biggest benefit of building with RW over other builders.

From my experience it cab be just as easy or easier to add code to Blocs or Wordpress. Writing code can be challenging for RW due to Stacks getting in the way, unless you are using Source and learn how to use Coder.

I consider RW to be more of a basic DUPLO set where the potential is there to progress to Lego but the bulk of RW users still build what’s pictured on the box. The RW market is fuelled by out of date DUPLO addons discounted over and over again. E.g. theDUPLO sets in all the old weird colours.

Actually, Blocs is based on Bootstrap5 which supports CSS Grid.

Why not just use Pinegrow if you want drag and drop but with the ability to code? The vast vast majority of RW users are drawn to exactly the things that are being decried here, namely super simple drop in solutions with direct support from developers. We always need to stop and remember that these discussions relate to the 1% of even the advanced users and not the main user base. RW would certainly have no commercial future if it catered for the hybrid use cases described here; the market would be tiny. Who one earth would buy RW as a construction kit (with the huge buy in cost including stacks) when there are already a myriad better and more powerful solutions. Users get to this point through a RW addiction and the reluctance to move on to more appropriate tools. If RW lost the drop in stacks ability and the consequential support from developers then it would have no appeal. Why on earth use a huge cumbersome app to then write your own HTML and CSS, I’m afraid I don’t get it.

1 Like

It is browsers that support CSS Grid and not frameworks. You can use CSS Grid anywhere. CSS Frameworks are just helpers.

Oh and container queries and dynamic layouts that are not dependent on browser width have just passed through the CSS spec and are awaiting browser implementation so we can focus on truly responsive elements and not multiple static predefined layouts - which is what the current media query defined approach gives us.

Andrew, I suspect you didn’t get it because you focused on every other point I made except the main one: Rapidweaver/Stacks, fortuitously, turns out to be a good environment for building sites which fully exploit the potential of CSS Grid. But I don’t get the impression you see the point of Grid.

As a beginners’ ‘plug and play’ application, RW is actually a total dog. Nothing works out of the box, problems can often take hours and hours to get to the bottom of, the behaviour of stacks is frequently counter-intuitive, they often don’t play nicely together and many are far too complex for people who don’t know much about the web to get their heads around. After more than a decade of using RW, I could not in good conscience recommend it to someone who knows nothing about how the web works. I’d steer them towards Squarespace if they want a stylish site which does everything a basic site needs to do. Or Wix, if they’ve got no sense of style.

One step up from this, Wordpress dominates the market. Even the friend who got me to build his first site in RW, because he thought it was great, abandoned if for WP. His point: “I can’t get anything to work properly any more!”

So the question becomes: “What is Rapidweaver actually good for?” Well, it’s a very convenient platform for managing a website. If you’re a Mac user and someone built you a RW site and showed you how to update and publish it, it would be as good as any other option. It feels like a Mac app. Likewise once you’ve gone through the hair-pulling of building a site in RW, it’s a great place to maintain it. Why would I want to manage my sites through pay-per month cloud based applications? But a design platform — well, it can feel like trying to design in Word 2.0. Everything is controlled through endless dialogs and boxes. You can’t drag things to where you want them or pull a handle to resize them. So for a designer, it’s easy to see the attraction of a the new generation of WYSIWYG editors. Except… WYSIWYG is not great if you are working with elements that change their size, position and behaviour in different circumstances. Then it becomes a liability.

I don’t hold out any hope for RW’s future — I sense, for what it was intended to be, it’s pretty much reached the end of its viability. I don’t get any confidence that Realmac knows where to take it. I see the community of stacks developers emptying out. Joe and Stuart and Jannis and a few others are left holding the can. RW needs to find a new positioning. And, actually, frequently the most successful technologies are those that have been repurposed for something quite different from what they were intended for. For me, the fit between RW and Source Grid gives me the confidence that this is the right platform to be investing in for my sites: RW turns out to be the best environment for working for complex grids, and — actually — this approach seems to involve far less hair-pulling than the ‘stacks salad’ way of working.

I absolutely do and I use it frequently. I just don’t get why it seems to be linked to a particular app, framework or whatever, its just CSS like everything else.

This is the part I don’t understand. What makes RW / Stacks and different to anything else from a text editor onwards. display: grid seems to have nothing in particular to do with RW/Stacks.

@habitualshaker has made a cracking stacks implementation of this part of CSS, that is not in question - I just don’t get the view that it is somehow different from the rest of the CSS spec.

In one sense CSS Grid is just CSS ‘like everything else’, for sure. But let me take an example from DTP. All the modern graphics apps are generating just PostScript, even 40 years after John Warnock and Chuck Geshke created it. PostScript hasn’t really changed all that much in that time, but what has changed is the sophistication with which we can use it. For instance Display PostScript made it possible to see on screen what was going to come out of your printer. That was revolutionary in terms of it made possible. The first most Mac users saw of it was in Adobe Type Manager, which made it possible to see fonts as they looked, rather than as a pixellated mess. Then PDF – which was basically just the Adobe Illustrator file with some extra headers — made it possible for PostScript to become a document format, as well as a printer instruction language. All the smarts, really, came from the applications. And quite soon nobody could have sat down and coded the kinds of documents they were able to produce with an application in pure PS.

The second thing is that CSS Grid didn’t just evolve as part of the CSS spec. It was a radical departure for the web, and a final victory over a view that had prevailed since Berners-Lee with his NeXT box at CERN, that the web should be an environment for rendering pure generic markup (HTML being the child of SGML, and the ‘separation of form from content’ philosophy which produced it). We can see this in the contrast between PostScript and HTML. In PostScript pretty much every statement begins with a position — ‘newpath 100 200 moveto…’. In HTML nothing originally had a position — every attempt to position elements with greater precision, starting with subverting tables for layout, has been relative. Even Flexbox is just relative. Webpages were jellyfish, which is why I liken the old web to designing in Word 2.0 (which was also entirely inline). CSS Grid overturns this. Now it is possible for every element of a page to have a defined position at every breakpoint. Suddenly the web has grown a spine.

The question then becomes: “what is the best way to manage this structure?” WYSIWYG isn’t it, because it can only show us one situation at a time, and what we need to do is create a set of rules which define behaviour across the whole range of situations. We could do this in code, but development environments are not great for managing content. Suppose we want to drop a stack — an encapsulated component — into a grid item? Or even just edit the text? It’s much easier to do so with a drop-zone, or an editor window. And this is where RapidWeaver’s hybrid environment becomes an advantage. With Stuart’s implementation of Grid, we can open and close nested Grid items like an accordion. Grid then is not just another element of the page — it becomes the basic structure of the page, the way that everything is organised. Of course RapidWeaver/Stacks/Source/Grid Plus Pro is not the only way to do this. But for the moment it‘s a pretty damn good one! (And a way in which RW might have a future, because otherwise it’s beginning to look like a bit of a dead duck…)

1 Like

I’ll start by saying that this is purely technical. I don’t really get the idea of passion around CSS beyond the organisational improvements and readability over inline styles of the early 90’s. I’ll also say that I don’t like or dislike CSS grid or any other part of the spec; its just a spec.

Of course it did, if it hadn’t it wouldn’t be implemented in browsers. It was first proposed by Microsoft and implemented in IE10. It then went to the W3C and after many years of evolution it was agreed as a spec and implemented uniformly across browsers.

From a technical standpoint I don’t understand what you mean and why gird would not be “relative” These are not terms that I understand. My view is that Flexbox is a two axis tool with the limitation of only controlling one axis at a time. The power of gird is that we can simultaneously control two axes. This of course allows a whole raft of layouts that were only previously possible using JS and lots of calculations. Those calculations have now been moved to the compiled C code of the browsers rendering engine and so are much much faster.

This is absolutely possible with every other element on the page. Nothing has changed with media queries yet - grid just controls 2 axes nothing more, nothing less. Let’s stay accurate. (Container/element queries will finally change this)

True WYSIWYG will absolutely show it - you would just see what you would see in the browser on the finished page at any given size. If you are cleaning that the finished rendered page and browser are not WYSIWG then where does this mythical state exist and what is its purpose. The whole thing seems to have crawled up its own grid-templete-column.

I said that RW/Stacks is not the best place to implement CSS Grid for sound technical reasons. The way that the stacks interface and controls work are not ideally suited to CSS Grid. It introduces a lot of complexity in settings due to so much being on the element and not the container. Developers have to work hard to keep settings and edit mode performance in check as it certainly does not lend itself to the way that stacks is structured.

I will finish by saying that I live in fear of the discussions that will happen on here once container and element queries are widespread. I suspect we may see a big star in the east when breakpoints become a thing of the past.

Andrew, clearly some clarification needed.

  1. My point about Grid not being an evolution of the CSS spec is that it was a revolution. Flexbox was an evolution — a kludge, half-hearted, an underpowered and overcomplicated nod to those who were against CSS developing into a full-blown page description language. The comments of people involved in its development, like Rachel Andrew, Elika Etemad and Jen Simmons make this clear (but maybe women just get this better than men?)

Hence Bert Bos, who has really driven the idea of Grid from the earliest days:

“It’s not just attaching your margins and properties to each individual element and placing them, You can now have a different model, a model where you start with your layout first and then pull in the different elements into that layout.”

  1. One huge difference between Flexbox and Grid is that it would make no sense for Flexbox to be used for the basic architecture of a page, whereas Grid is ideally suited for this role: designed for it, even. The axis point simply underlines this: the basic structure of a page has to be defined in two dimensions — page layouts have been defined in this way ever since Gutenberg. What is the point of a one-dimensional layout tool?

  2. Defined positions — Grid makes it possible, for the first time, for a page to be defined with the same precision as PostScript/PDF. If we ignore position:absolute, which nobody in their right mind was using to define all the elements of a page, the only way we could replicate a complex layout on the web is through a combination of kludges using floats, tables and flex. By contrast Jen Simmons’ recreation of a 1930 diagonal, assymetric Jan Tschichold layout here (CSS Grid like you are Jan Tschichold - YouTube) shows the incredible power of Grid, especially when you see how she has made this layout responsive. And that was four years ago, with Grid 1.0.

  3. The problem with a WYSIWYG editor is that you would be clicking and dragging and doing all that stuff on just one possible instance at a time — I struggle to see how that would be a good way of working, when what you need to do is set rules which define the behaviour of elements across a range of different situations. As far as ‘seeing what you’ve done’ RW’s preview mode(s) and options seem as good as any.

  4. “I will finish by saying that I live in fear of the discussions that will happen on here once container and element queries are widespread.”

You seem terrified that the children will loot the sweetshop! (So much so, you don’t even trust them with variable fonts!) I think we need to stop infantilizing people, though — we now have access to wonderful professional tools, and we should be helping others to get the most out of them. My fear is that Realmac will announce that they’re stopping development of RW. And unless we can find new purpose and life for it, that seems inevitable — its market share is diminishing to zero, while its competitors in the beginners and intermediary markets (Squarespace, Wix, and Wordpress) top the lists and grow and grow. Look at the two reviews for RW8 in the Appstore, and it’s inhabiting about as downbeat a place as a £80 app could possibly be.

I can’t respond to most of this as it is subjective and words like kludge have no technical meaning to me. Relative is still undefined from the previous question.

I will just say that Flexbox is an integral tool in alignment with grid items and is also often the best solution to making menus and other “two dimensional” structures.

I have over 120 unreleased stacks in addition to the variable font stacks that are currently in use by a private group of people. These are not all advanced users.

I have not released them because:

  1. I cannot adequately support even more public released stacks
  2. I have tried not to compete with people who are trying to make a living out of charging for stacks by giving away free stacks through respect for them.
  3. I have been busy with other projects.

Over the past 6 years I have given away for free over 150,000 downloads, answered nearly 15,000 support tickets and probably another 10,000 forum questions. I have never once refused to answer the most trivial basic question nor the most advanced request. I have always tried to push forward and to innovate. I do however hope that this has given me an understanding of the types of users that RW attracts and I attempt to cater for these groups. Not always going for the most technically advanced solution is not infantilising people, it is reacting to the users needs and requirements.

I have no vested interest in this other than discussion, I am not trying to sell anything to anyone.


Sorry, Andrew, I missed the bit in the rw4all FAQ which said that only ‘objective’ answers were acceptable. But, look, I don’t want to intrude. If this is your platform to play the ‘big man’, fine. I’m outta here. Bye.

1 Like

I am a scientist and I thrive on dispassionate debate. I am not trying to play anything and I certainly haven’t personalised anything, I’m truly sorry if you have taken it that way.

The most fascinating thread in a long time and both protagonists have given valid, coherent viewpoints.

Good stuff, gentlemen.

The northern scientist writes as he speaks and this concept takes a degree of familiarisation to normal folk :-)

1 Like

…are passionated 😃

I think we all want the same: RW being prosperous, it’s user base grown, and fun to use.

At the end it’s just a bit of CSS (ok, I never got to grip with CSS grid, but that’s a personal issue 😅).

Cheers 🍻


Not sure where you’re hosting but pages are taking 15-20 seconds to load for me - and I have fttp with download speeds of 700-900MB. I’m in the UK. Source projects re normally super fast.

I’m far to dumb to understand most of what’s been said on this thread. But I do know that comment is totally out of order, unfair and not true. Tav has done more to push the RW ecosystm forward than any one else, all for free, he doesn’t need to “play the big man”.

If you feel you are losing an argument, sharpen your wit, not your knives.


No need for this comment. Tav is a grown man he doesn’t need your help defending his points.

Is that aimed at me? I’m not defending his point, I’ve no idea what anyones point is, it’s all way above my pay grade. I will though make my own decisions on whether I choose (or not)t to call out someone out for what seems to be an unwarranted personal attack.

You wrote:

"I’m far to dumb to understand most of what’s been said on this thread. "

“I’ve no idea what anyones point is, it’s all way above my pay grade”

But yet you picked a side and went after the opposing view.
Your snide remark doesn’t add to the conversation it helps to end it.

TemplateRepo did not pick sides in the realm of merits of this discussion. He just opined on the behavior of one of the opponents. Personally, I agree with Steve.