What I want to see in MODX3? MODX & The small things...

I'm a MODX proud MODX user since the first the I tested it. I was also partner of the old partner program.
First time I used MODX I was personally impressed. Following the MODX Beginners Tutorial from the MODX Evo Wiki, in less than 30 minutes I had a website running! MODX Evo was easy to use, stable, flexible... It had some lacks, but everything it had was well done. I missed more extras or features, but we were a small community.
MODX Revolution promised to be great, and the first impressions were really good. It was really different but the base was the same. The team was so proud to be the first ExtJS based CMS. The community started growing and the new extras came fast. In each new version, new promising features were coming. But, IMO, everything wasn't so good.
Recently I used sometimes to compare MODX with a car. MODX is like a car that wants to be a great car, the best car with the best engine, the best wheels, the best look, the best hi&fi equipement, the faster car... MODX implemented everything to be the best. But this things were not well finished, so finally we had a promising car with a promising engine that use to fail, the radio is not so good, the wheels are not the best, and the car is not so fast... So finally, we don't have a good car, we have a car that fails a lot because we didn't take care of the details. The manufacturer thought about the entire car but each piece of the car was not well finished.

Some examples:
- ExtJS sometimes can be frustrating. Tree is slow, windows get frozen, buttons does not always save states, page refreshing times use to be slow... Also, development of new components is a little bit complex.
- Media Sources. One of the best features I've ever seen, but it has some bugs for advanced features that I've discussed with support stuff and are unsolved. ( problems with contexts, template variables and editor ).
- Permissions. Can't say more than said before.
- Form Customization: Great, but has some important bugs unresolved.
- Components. There are some important components wich development is stopped and has some issues pending. The team focused in launch a lot of components at MODX start but can't give support to every one of them. The team, then, was involved in some "complementary" tasks like development of an entire forum... Hey guys, won't be better to solve the mentioned issues before start a big development thing like a forum? Couldn't be better to develop bridges for other forums systems instead of developing an entire forum?
- Minnor bugs: There are a lot of small things that need to be solved and some of them are frustrating in the daily work.
- Documentation: Really improved at this moment, very bad when Revo was released.
- New features: We see how new services are appearing ( like MODX Cloud ) but no MODX new features or improvements are coming, only security fixes. The competitors are there, there are a lot of new CMS and MODX can't stop improving! There are a lot of important things that MODX needs and aren't implemented yet.

I now I'm being so critical, and that some of you won't think the same. I've been thinking about writing this post or not a lot of time, and now finally decided to do it.
I know that it's easy to talk from the outside, and the team is doing a GREAT work. But I really think that what some of the things I've commented are so important for users and for MODX success.

I can't believe this thread hasn't kicked off yet. Maybe it is because everyone is so focused on 2.3. Once 2.3 is released I know the core team will be shifting more focus towards 3.0.

Hi JP,

I guess people haven't seen it. I myself wouldn't have seen is if it weren't for your post to bring it up the chain.

I think that given how much focus has been put most recently on discussions of the Manager of the future, I think that I see a lot of community input being sought.

I think the ExtJS has been discussed a lot recently in several posts, so I won't comment on that.

On Media Sources, Form Customization, Minor Bugs, I can't comment without knowing more about the bugs that exist? Have they been logged?

On the topic of Permissions, I think I finally understood why they designed them as they did, but until I understood I thought it was overly complicated. I think that with some minor changes to the way they are presented and set, those could easily be recycled into something more intuitive.

On the topic of components, I think that this was addressed by the recent post by JP on the Adoption of Add-ons http://modx.com/blog/2013/10/29/announcing-adopted-add-ons/ but I admit the OP's post is from over 8 months ago.

Hey guys, won't be better to solve the mentioned issues before start a big development thing like a forum? Couldn't be better to develop bridges for other forums systems instead of developing an entire forum?

My understanding was that the development of the forum was to make it more user friendly for the community. In my opinion, I think it was a good move.

On documentation, there's nothing wrong with us patching up documentation as we see issues with it, or raising a "this is documented" flag.

On the point of new features, 8 months ago, this would have been completely valid. Knowing what we know now, we as a community are discussing a lot with new features and how to use existing ones better for our daily work. MODXpo, in my opinion, was a great forum for that.

I see MODX 3 as a chance to simplify a lot of things. The Jasons of the world should keep finding the right way to architect something, and others should help find ways of presenting it so that by default, the person using MODX sees the simple tip of the iceberg, but within close reach, are all the complexity you could ever need. I'm a big fan of user experience (I really enjoyed the book "Designing the Obvious" http://www.rhjr.net/dto/. In my world, I see people struggling with things that can be made really simple for them, and I try to do that to the degree possible in my job.

Regards,
Tom

How has this thread not taken flight yet?

"MODX 3" was how we referred to breaking changes for a long time. For a front end and UX perspective, I think it will be more like MODX 4 when we get those breaking changes. But don't be sad, the core team isn't opposed to dropping a 4.x soon after a 3.x.

The way things are heading, I think the MODX 4.x core will be incredibly minimal. For example, I've written a draft to pull the entire management interface out of it, so the core is just the core, and everything else is managed as packages. To the end user, there wouldn't be much of a difference, MODX could be distributed with some "core packages" like the default Manager interface. But the point is to make things more maintainable. If the management interface is decoupled from the core itself front end contributors won't have to wait for breaking changes in the backend to make breaking changes in the front end.

I want to vanish the concept that "the MODX Manager is powered by _________ JavaScript framework". First of all, I don't think there should a "the Manager". There should be a default one, and people should be able to make new ones. Yes this is technically possible now, but the system isn't completely architected for that level of abstraction. If the MODX core is just the core, and it doesn't ship with (but can be bundled with) a Manager interface, then we no longer have to incur the technical debt of the two being tied together by ExtJS, or any other framework. React is hot now, but I assure you if we went ahead and built the whole Manager in React, or Angular, or whatever we'd incur the same technical debt as we have with ExtJS years down the line. The core should be the core, and it should be distributed with whatever "Manager package" the core team feels is the most relevant. If the front end becomes irrelevant overnight, we can swap out the Manager packages without needing breaking changes to MODX itself.

Quote from: stalemate at Nov 19, 2013, 04:34 AM

My understanding was that the development of the forum was to make it more user friendly for the community. In my opinion, I think it was a good move.
Tom

I've been told forums are being moved to something better.

Quote from: stalemate at Nov 19, 2013, 04:34 AM

On documentation, there's nothing wrong with us patching up documentation as we see issues with it, or raising a "this is documented" flag.

There's a MAB DRAFT to move the docs to Markdown, and GitHub
https://github.com/modxcms/mab-recommendations/pull/19

mIDO's original statement including the analogy of the car is really great. How it's all the best bits, but just not completely optimised for the combination.

I appreciate exactly how much the decoupling will provide us in terms of value. The points JP makes about that allowing us a fair amount of agnosticism when it comes to technologies is really key.

And this is one point I've seen some people begin freaking out. "What do you mean there's not going to be a manager!?" Obviously that's a pretty genuine concern if you've been drawn into MODX by how easy it's made things in terms of sticking in code, adding field placeholders and being ready to deploy.

I see the decoupling is exciting - I also understand people worrying about the fragmentation that might result.

I think along with the discussion about decoupling, it will be key to the stability of the community to say that there will be some defacto offerings for key functionality. While it will be able to run 'headless' that won't be of interest to the vast majority of the people using MODX - possibly talking in terms of overrides instead of missing parts might be helpful.

Anyway - great original write up. Right on with some very key points. Can't believe it didn't come to my attention until today.

Quote from: jcdm at Jun 06, 2017, 02:36 PM

And this is one point I've seen some people begin freaking out.

Yes. I suspect more rotten tomatoes would be thrown my way than the movements towards Grunt and Sass combined. It's any easy topic to rant about, and people like to rant.

I think we just need to be clear the idea is we would be pulling out "the Manager" just to put it back in again as a decoupled and default core package which can be swapped out, customized, and overridden far more easily than before. The new Setup concept aims to make MODX configurable during the unboxing process, so people would have all the packages they need installed even if they don't know what a package is:
https://forums.modx.com/thread/102211/new-setup-installer-w-a11y-preferences#dis-post-550718

I feel whatever "Manager" ships with MODX by default should be architected with web standards, accessible, and inclusive to end users. That would be a huge improvement over what we have in 2.x which is not built with web standards, literally impossible to make accessible, and not very inclusive to end users.

Quote from: jcdm at Jun 06, 2017, 02:36 PM

I see the decoupling is exciting - I also understand people worrying about the fragmentation that might result.

I think it could be the responsibility of the MAB to make sure any of those issues are addressed, or at least considered. But I actually would like to see some fragmentation so when people refer to "the MODX Manager" you ask "which one?" because there are several options to chose from.

I think I may be on the other side of your Manager argument.

I've railed in the past about having both a traditional and an advanced distribution. I'm confident that MODX has lost a lot of potential new users over the years because when they went to download MODX, they didn't know which one to get, or why, so they went elsewhere.

I have a friend who uses WordPress. When I asked what browser she used, she said "I just use my computer." She didn't know what a browser was, but was able to use WP with no trouble. MODX has to be usable for people like her, imo, if it's going to go beyond being a niche platform for savvy developers.

I like the idea of a decoupled the Manager, but I'd much rather see one installed by default and let users replace it with an extra rather than face a choice on their first install (though maybe that's what you're suggesting).

I'm curious to know what you consider to be the architectural roadblocks preventing us from decoupling the Manager in a 2.x version. I've toyed with the idea of creating a simplified Manager, but never got far enough along to run into them.

Let me add that what I think has hurt MODX the most, by far, is the absence of any significant MODX-specific marketplace for add-ons.

Quote from: BobRay at Jun 06, 2017, 09:52 PM

I have a friend who uses WordPress. When I asked what browser she used, she said "I just use my computer." She didn't know what a browser was, but was able to use WP with no trouble.

WOW :O :O \

MODX has to be usable for people like her, imo, if it's going to go beyond being a niche platform for savvy developers.

Useable on the back-end, aka the manager, sure. As useable as WP - definitely not. Otherwise it will become like WP, a big pile of..... things.
Not that being useable is bad, but like WP has a plugin for EVERYTHING. Hell, they have a dozen plugins for making a "about the team" page. It has become like one of those damn auto-builders. MODX is awesome the way it is.

https://www.youtube.com/watch?v=oZu2JfM2Aq8

Quote from: BobRay at Jun 06, 2017, 09:52 PM

I like the idea of a decoupled the Manager, but I'd much rather see one installed by default

Well, I'm not talking about not installing a Manager by default. I agree that would be ridiculous. As far as end users are concerned MODX would be the PHP magic sauce, the default (or another) Manager, and any packages they may or may not know about because the Setup installer allows them to just say what they want and it installs the right packages (and a Manager).

I'll rejoice the day I hear someone say "I installed MODX right from my hosting account, told the installer what I want, and was making my own website within an hour. Wanna see it?" and someone asks "What packages did you use?" and they reply "What's a package?". In other words, to us "coders" we'd say they are using the Core, and all these Extras, but to end users its just MODX.

This is mostly an "under the hood" thing I'm talking about so that front end contributors can work on their own semver and own roadmap and not have to wait for an "eclipse" to make breaking changes.

Quote from: BobRay at Jun 06, 2017, 09:52 PM

I'm curious to know what you consider to be the architectural roadblocks preventing us from decoupling the Manager in a 2.x version.

Honestly, I'm not entirely sure. Mark could probably speak to that, since he basically did it with HandyMan.

Quote from: BobRay at Jun 06, 2017, 09:52 PM

Let me add that what I think has hurt MODX the most, by far, is the absence of any significant MODX-specific marketplace for add-ons.

modmore is MODX-specifc :p

The modmore site is an excellent source, and I'm always impressed by the quality of the extras, but it has only 6 extras and I think little interest in extras not developed in-house. There's nothing wrong with that, but I think MODX needs a much bigger, more open marketplace.

Here's a post from Ryan promising such a thing (note the date): https://forums.modx.com/thread/50841/modx-marketplace-for-template#dis-post-297609