This Gutenberg thing really isn’t working properly

I have. Sent plenty of feedback to the Gutenberg team, like many others, only to have it fall on deaf ears.

Some of my criticisms:

____

The whole Classic Editor plugin approach is very clumsy and unnecessary considering that Gutenberg is in effect just a react override of the current editor, much like enabling a builder plugin might work. The only difference is that Gutenberg takes over the whole backend interface.

Fom my understanding all the workings of the current editor will be extant in WordPress 5 with Gutenberg activated as the de-facto default editor. It is feasible therefore that only new installs would activate Gutenberg with a switch in writing settings to trigger a Gutenberg -> true/false filter.

There is in fact a filter that can be dropped into the functions.php file of a theme that prevents Gutenberg from being active on all posts. At the moment the folks behind Gutenberg have stated that that filter will get a name change once Gutenberg is folded into core. To date that name change has not occurred and no guarantee that there will even be a name change. This suggests an element of control in the Gutenberg camp. Having a concrete filter would make it feasible for theme vendor to use it saving their users being dropped into Gutenberg.

I have been using and testing Gutenberg for a while now and some plugin developers have gone to great efforts to make blocks that potentially could be very useful. What lets them down is Gutenberg itself. The bugginess of the configurations, especially if you use it on a theme that wasn’t built for Gutenberg suggests that a complete separation of only using Gutenberg on new builds with Gutenberg specific themes and plugins is the only route.

I use Divi for work and this make Gutenberg redundant for me. But I do see a better built Gutenberg being a good starting point for those new to WordPress, web design and development. My understanding is that Gutenberg was supposed to be unifying in that it would provide a common foundation for content and make changing theme less painful. In many respect it does none of this because it has gone in the opposite direction of layout towards micromanaging more granular elements.

And there are many other issues that need to be resolved like where do all the custom fields go if metaboxes are an after thought. There is a all in blocks solution which potentially could work down the road. Providing a mechanism to only have one classic block for basic content creation and/or a de-block mechanism to take everything back to a singe classic block are two other important pr-requisits to making Gutenberg more usable.

The biggie though is something that Gutenberg share with the current editor and that is the lack of a proper code view. its a big bug bear for me and always has been. I like the fact that in Gutenberg you can target the html in each block. The big let down is that it is often not human readable. It needs to be more like an IDE as a work environment with colour syntax highlighting, proper formatting with nested indentations and perhaps code completions and code folding. We see how this achievable with Froala editor and Code Pen. Recently WordPress introduced Code MIrror and what we need is further development of this and for it to be used in far more places other than the HTML block. This would go some way to making working with blocks and how it handles delineation with comment tags more useable and palatable.

_____

On the bugginess and block disintegration, to be honest I have been hammering Gutenberg switching between themes etc… I wouldn’t be too worried at this stage as a lot of these issues should be ironed out with time. I am sure that Gutenberg will become more robust. In the meantime…they shouldn’t make it default and I don’t mean this in a bad way. There is always a place to push out new software to make it grow, take feedback and improve. We had this when OS X replaced the older mac operating system. it took up to five years for Apple to get the new operating system mature enough to be adopted by most users.

A few quick things on the interface in you block settings. You have the three device (desktop, Tablet, Mobile) selectors a the top. I would expect that when clicking these would animate the content down to reflect previews for those breakpoints. Maybe that’s something that is coming in phase two for the customiser in Gutenberg? I am not sure if you are familiar with Divi. In the visual builder on the front end in the settings panels, each section operates like an accordion, only the settings you are working on are open, all others are closed. It affords a less busy workspace and makes it easier to find what you are looking for. Again, probably something for Gutenberg to get right.

I upped the star rating to 4. I may have been skewed by my impressions of Gutenberg, which will remain at zero(1 star) till they get things right.

None of the builders can have IDE code editing as far as I know because they depend on TinyMCE/core implementing. There are exceptions. In Gutenberg it looks like the HTML block uses code mirror. Likewise Divi uses it in the code module.

Behind the scenes Gutenberg relies a lot on TinyMCE which is a variant of the full blown version of what Tiny provides as a subscription. Generally TinyMCE does not have a code view at all. For this additional feature you need to subscribe to anther pricey plugin. It is not available to WordPress user either and from what I can see is possibly not as good as code mirror. The text tab for code view in the current backend editor is a WordPress implementation. It is a bit blurred in terms of what TinyMCE is and how much of it we get in WordPress which is obviously not the totality, seeing as we as the end user pay nothing for it. The TinyMCE Advanced plugin give us extra functionality all right.

There is in depth co-operation between Tiny and the Gutenberg team, the vested interest reflecting the 30% + share of the web market that WordPress has. If you look at the Froala editor it has a fairly decent code view option and with full screen is useful. Again though if you want full functionality, you need to pay big bucks. The free version is not much use if you want to code. I am not sure what hurdles are involved in attaching a proper code editor to the Gutenberg editor. It may be a bridge to far but I would like to think that at least somebody would try.

On the overall philosophy of Gutenberg. I think the idea to change and what it could do is very interesting and much needed. I just think that it has gone down a dead end now in how it focused on the granular level of elements of paragraphs, images and headers to the detriment of the layout elements. The under developed column module speaks volumes.

I would have thought that providing the fundamentals of layout in the form of sections/rows/columns and/or flex etc. should have been a priority affording a good foundation for switching themes without unexpected results. On this foundation would be the basics of adjustable paddings and margins for spacing and then, on top of this a very good API for page builders to add all their own bells and whistles and custom interfaces. This would also cater for single layout block instances.

This basic Gutenberg/WordPress install would cover enough for most people to comfortably build a site without too much stress but instead we have issues targeting all the block elements and unexpected bugs. What I find ironic is that we are encouraged to avoid the mystery meat of shortcodes while, in its current guise, Gutenberg is the biggest piece of mystery meat I have seen in a while in terms of software for general consumption.

We also have talk of a fork, but this is as bonkers as the notion of Brexit, though at this stage I am not sure which is more bonkers. We really need to hope that the Gutenberg design can be resolved to work properly.

On testing. The errors I observed were not tied to any one block and I did experience them prior to testing Kadence. I will test a bit more over the next few weeks and let you know if I spot anything in particular.