Community Concerns

The upgrade process in MODX Cloud is basically 1-click. I know that doesn't help you for sites hosted elsewhere, but if MODX Cloud is in the discussion that would be a way to even up the WP vs. MODX tally.

Quote from: waizen at May 01, 2015, 12:54 PM

Quote from: BobRay at Apr 28, 2015, 03:50 AM

Good points. Personally, I'd like to see a pure HTML/CSS Manager for starters with sensible classes and very little styling. Themes could then be created with an overlay of CSS, JQuery, Angular, Ember, or whatever. Almost everything in the Manager, except maybe the file/image browser and datepickers, could be done without JS...

Bob, what makes me a little nervous about this, and please correct me if I misunderstand you, is that starting with very little styling for the control panel might add more development time to a site.

Where I work, my boss has been pushing us more and more toward using WordPress because he believes we can shave man hours off each project, despite all my arguments against this. Having to add control panel styling to the work flow can really work against many of us who depend on cranking sites out as quickly as possible. In the area my company operates, we cater to a lot of small mom and pops with small budgets.

Did I get your comment correctly?

Sorry If I wasn't clear. I was talking about early development for MODX 3. I never meant that MODX should ever be distributed with no styling, just that if there were a base that was pure HTML/CSS underneath the styling, people would be free to create new skins and user interfaces based on CSS and various flavors of JS at will, and that MODX would fall back gracefully for users with JS turned off.

Ah, okay, Bob. I can see that.

Quote from: lucy at May 01, 2015, 07:12 PM

The upgrade process in MODX Cloud is basically 1-click. I know that doesn't help you for sites hosted elsewhere, but if MODX Cloud is in the discussion that would be a way to even up the WP vs. MODX tally.

Yeah, our sites are hosted on our servers. Still, why not make those sites easily upgradeable, even by our non-technical clients?

I would love to see MODx, not only do all the WordPress-easy-for-the-end-user stuff, but also maintain all the blow-WordPress-away stuff.

Oh yeah, and I want to be 23 years old again and win the lottery.

For standard installations, you can use the install.php script for a quick upgrade as well as for a new installation. https://github.com/sottwell/installer

It can also be used to download an advanced installation, but the core would have to be moved manually before proceeding with the setup.

Thanks Susan. I'll check it out and test that script really soon.

I guess I'm thinking about something even simpler for the average non-technical person though. For instance, I just read the instructions in the link you provide and while you and I would have no trouble breezing through it, the people who own the sites I build wouldn't know the first thing about FTP access, or what to do with it, let alone understand the concept of permissions or why they would be necessary.

Nor should they have to. They're too busy running their plumbing business or dental practice to need to worry about such things.

But, guess what: WordPress does make the upgrade process that simple for these people. One or two clicks in the WordPress back end, or actually, an automatic upgrade by the CMS when a new version is available, along with a confirmation email to alert you when it has done so. This past week I noticed several WordPress sites under our care sent me confirmation emails that it had thoughtfully upgraded those sites for me. I didn't need to click anything even once; all automatic. Probably good seeing how many security issues that CMS experiences...but I digress.

Heck, even all our handheld devices...phones and tablets upgrade themselves. Not just the operating systems but all the apps, too. It's routine now.

MODx needs to become something beyond a codemonkee's domain if it's going to grow. It needs to be able to handhold the common mortals who get even too intimidated by simple HTML. It needs to speak to people like my boss who deal in the business of running a web development business, where too much development time cuts into profits. That's where the WordPress appeal comes from...it's ability to hold the mortal's hand through necessary stuff.

Gotta figure that if WordPress can do it, why can't MODx? [ed. note: waizen last edited this post 9 years ago.]

First of all I must say, I kind of like modx, as it has some good parts, but ...

IMO modx got the ExtJs implementation in the manager all wrong the first time and as far as I can see, you will get it all wrong the next time (whatever you are going to implement may it be raect.js or something else). Let me explain: When you click e.g. on a resource in modx manager the hole page is beeing reloaded. At that point you lose all the benefits from ExtJs as it has to load, bootstrap, fetch data and render everything again. This guaranties bad performance. Although just the right column of the manager changes everything has to be rerendered. That's a no go for javascript based webapps. Did you ever only look e.g. at the Typo3 backend? They are using Sencha ExtJS too, but they got it right. It's a onepage app wehre only the elements that need to change after a user action (event) are fetched and rerendered. Or look at e.g. the admin interface of a Synology NAS - they got it right too - they even implemented the Sencha Desktop thing.

And now you are talking about HTML first? You do realise, that e.g. react.js is not much more than a DOM i.e. HTML renderer (and event observer)?

I think, I know where this comes from: PHP developers fear javascript if it is more than just a little jQuery page tweaking. I admit that it's hard to imagine systems where almost everything is done on the clients side (data requests, data fetching, rendering, routing etc.) and the server side does nothing more than deliver some data over a restful or even a graphql API. But in the end you must decide wheter the modx manager is a modern web application or just a thing based on a simple document retrieval system called HTML.

Quote from: vik0xs at Dec 16, 2015, 03:13 PM

I think, I know where this comes from: PHP developers fear javascript if it is more than just a little jQuery page tweaking. ... But in the end you must decide wheter the modx manager is a modern web application or just a thing based on a simple document retrieval system called HTML.

Vik, I have to say that your post comes off as (more than) a little condescending. This is a 20 page thread that I don’t think you bothered to read through. We’ve been discussing the merits of why MODx has not been adopted like it really should have considering the power it brings to the table, and what can be done moving forward. That discussion lead to the realization that the manager is bloated, hard to maintain and build upon, and difficult to customize (not to mention the accessibility issues). The goal is to move away from being tied to a centralized technology that binds everything together, because MODx is just too big and too complicated to even consider a one page app. Your suggestion basically adds insanity to the original sin.

The truth is, we are all trying to group-think the correct way forward, and some of the brightest minds out there agree that the best way is to become less dependent on javascript, not completely dependent on it. There are more ways to look at it then just from the user’s point of view. Maintainability, scaleability, readability, extendability, usability and modularization. Tying everything into a one page app destroys five out of six of those tenets in my opinion. That doesn’t mean that sections of the manager can’t become a one-page app inside of the whole, but to think that the entire ecosystem of MODx could be contained within a single page load is not very realistic to put it nicely.

I do agree with you, though, that data should become more restful through API’s, allowing more avenues of consumption (not just HTML, not just PHP, not just Javascript, but any technology that can access the web).

Sorry, I did not want to offend anybody.

I know that global architecture and design decisions for large systems are really hard to make, not only because they will last at least for the next couple years.

So let’s look at some other really large systems:

• Facebook changed almost everything towards javascript, in fact react.js, relay and graphql are facebook technologies & products.
• Github as another large system is constantly moving towards this technologies.
• Instagram (belongs now to facebook) changed also to js and one-page-app.
• Netflix is also js & one page (with a very interesting tech called falcor)

So if very large systems from multibillion Dollar enterprises can do it, why can’t modx?

Because modx is bigger and more complex than Facebook? Come on …

And having a one page app does not necessarily mean you have to do everything within a single page load. The opposite is true, you can load dynamically just the parts of the page that you currently need…
[ed. note: vik0xs last edited this post 8 years, 4 months ago.]

Quote from: vik0xs at Dec 16, 2015, 06:43 PM

So if very large systems from multibillion Dollar enterprises can do it, why can’t modx?

No offense taken at all, and your argument is valid. But I think you answered your own question: all these applications you mentioned ARE multi-million and multi-billion dollar enterprises. MODx is an open-source system that relies on people that simply love the framework to put their hard work and valuable time in for FREE. Which means that having an easy to understand, open system that isn’t tied to a technology that’s guaranteed to change within 2-3 years becomes even more important. For an open-source CMS/CMF to live, it has to be adopted on a large scale (see Wordpress), and to be adopted, it has to not necessarily cater to the masses, but at least be friendly enough that normal developers can contribute, and normal content editors can access it without their heads blowing up.

Your argument is entirely valid for a system that has already been adopted by millions of people (see Wordpress and its new desktop apps). MODx is a system with a brilliant concept and a great core, but the interface is aging and bloated, and the community of developers is comparitively small. Sorry, it just is.

So how do we move forward within the constraints of an open-source system with little to no funding behind it? By simplifying, moving away from flavor-of-the-moment technology, and creating an API system that is easily adopted for whatever platform wants to use it.