Community Concerns

I would warn against strong personalities turning the project into a fiefdom. For example, I am agast at what M. Shuttlesworth has done to Ubuntu, they have turned 180 degrees away from user freedom in the last several years. For him, its all about his "vision", which means we are all just drones to be dictated to

Pretty much the same thing happened to Etomite, leading to the MODX fork and Alex Butter leaving Etomite altogether, so the foundation developers are well aware of the danger.

I think its an easy trap to fall into, not saying its a danger here. Its just that smart people often have strong viewpoints. Ego plus the lure of control...its somewhat intoxicating. Heh its a story that's been written many times in history.

I read through most of the posts here and still don't fully understand what is everyone expecting from the LLC, I mean besides Mark's post which is quite clear. Please note I do not work for the LLC, I am just curious about what others seek to get from this thread. I see Christian tweet about it every now and then hence why I am asking also.

I'm also a little unclear here about the problem we're trying to solve here - but perhaps I'm missing something obvious.

As I understand it, most of the commits to the MODX core are made by two people, Jason Coward (opengeek) and splittingred (Shaun McCormick), who are also two of the owners of the LLC. Others are free to participate, but it's a big complex animal and participation has not been huge.

Again as best I can understand, some people think there should be changes to allow faster progress on the core issues that the team want to address - user management, interface speed and usability, document and image handling. The concern, coupled with the lack of people flocking to help with the work, implies some systemic problem that makes it hard for new developers to get involved in moving the core forward.

Is this people's main concern? Or am I missing something? (Really - please tell me.)

Some additional notes:

• Having read Mark Hamstra's document it looks to me like something with a high risk of sapping all the enthusiasm of the blokes who are doing most of the heavy lifting. I am not a developer, so maybe I'm worrying unnecessarily and Mark's suggestion is a pattern which traditionally works well in small open-source projects. But I do wonder.)
• I like the idea of crowdfunding new developments. I have in different ways been involved in funding two small additions to the Extras repository, so it's not hard to do. Crowdfunding is not obviously that much harder. I don't have the time to do it, but someone could set up a crowdfunding system, then email Ryan and the team and simply ask whether they would like to stamp their imprimatur on that crowdfunding system.
• As I understand it, the documentation system is not hard to contribute to. I have several times solicited contributions for the slowly [url http://shorewalker.com/site-building/modx/modx-end-user-manual.html]developing end-user documentation[/url]; people should feel free to contribute.
• I would like to see another level of approval for Extras below the gold-standard "Solid" ranking.
• None of this stuff looks like an existential crisis for MODX.
• The existing product is extremely good, which is why I'm using it. It seems to me that while we all have wishlists ([url http://shorewalker.com/site-building/modx/improvements-to-propel-modx-to-leadership.html]mine is here[/url]), you can comfortably adopt MODX based on what it already does. Indeed, people who adopt any software based on what's going to be delivered Sometime Real Soon are making an unwise decision.

I have seen this thread (and the tweets about it) a few times now and I'm also a bit puzzled on what people expect. Sure, MODX isn't perfect. Neither are Drupal, WordPress, ProcessWire, Craft and any other tool I have used.

Even commercial systems like Craft are not just built by the company behind it. Even they receive recommendations and sometimes suggestions on what to change in their source code. So why should 'our' LLC do most of the heavy work? Like it or not; open source software is a community effort.

So if the LLC is not doing enough, why not take matters in your own hands and change stuff yourself? You really don't need to be a developer for that, although it might help. Anyone can learn how to add issues to Github. Anyone can learn how to translate MODX into their own language, etc. And if that's not an option, why not hire someone to change things? Plenty of options already!

The answer isn't being worried. The answer is starting to change things.

splittingred (Shaun McCormick) is no longer part of the LLC.
All the recent additions and/or bugfixes are done by the community (historic of commis is available on github) and it's a good thing.

Concerning the point mentionned :

• Mark hamstra's proposition is common in the OSS world (Drupal thrive like that). From a developper perspective it does not change a lot of things as most of recent intervention on the core are being done by them whether they are from the LLC or not.
  But being handled outside of the LLC may clear some issues, like having a clearer roadmap.
• I'm not too fond of the "gold" standard. All contribution should be welcomed.
  The LLC don't have time to work on modx and you expect them to review 3rd party components. It does not sound possible at the time.
• Modx as a product is fine as it is, but it looks stalled in a state of auto satisfaction.
  On the technical side notably it's lagging behind. Concerns have been raised on that subject here on the forum. From what i see here, most of the community show little to no concern about that.
  
• Ultimately people want Ryan back on the forum
  They want the LLC to participate like they used to. The problem is not the product (2.3 is coming along just fine), it's the communication.
  Proximity has been replaced by marketing.

The people who bring up "it's open source, fix it yourself", please review this post on page 2 as it is meant for you: https://forums.modx.com/thread/90793/community-concerns?page=2#dis-post-497564

I don't think the problem this thread is concerned about is development (the community has been picking up), but project leadership.

The fact that the team has remained quiet throughout this post that started three weeks ago shows the problem*. While in the eyes of the community they are accountable (they carry the name of the product, maintain trademarks and handle releases and all that other stuff), they are not living up to that accountability and people in the community (me included) are worried about that.

If the LLC is not interested in being accountable or running the project (which yes, is exactly what they have done in the past, and in the current situation are supposed to do - that has nothing to do with development), and that is something my proposal for MODX.org aims to solve. It's not about locking out the LLC from being involved, it's about putting the community at the wheels and considering LLC to be part of the community, instead of the answer to everything.

As an example of lack of project leadership, right now, there's no telling when 2.3 will be out. Sure, "ready when it is ready", and I fully support that. But 2.2 was released over 2 years ago (January 4th 2012) which is a long time between releases. If there's anything blocking a release, the project leadership should put out a call for contributors. The roadmap seems to have everything for 2.3 at beta, except Context Name Field (shown as "Delayed", which is weird as it was added over a year ago) and Package Dependencies (which I've seen Jason work on somewhere, but can't find anymore.. in other words if people need to contribute to that, it needs more info); so what can the community do to get 2.3 released? That is project leadership - helping the community find areas that need to be contributed to so we can keep on being awesome together.

Personally I am worried about the lack of project leadership because I built a business that 100% depends on the MODX community. If for some reason the MODX community starts running to alternatives, I don't know what will happen to my business. Many businesses in the community depend on MODX one way or another. These are all stakeholders that deserve to know what is going on. The LLC should either cater to that, or open source the project leadership. Personally, my vote goes to MODX.org.

* I believe Jay, the community liaison who would have typically responded to stuff like this, may be on vacation but Jay is not the only person in the LLC with a forum account. If anything this thread should have been an open conversation between stakeholders and not a big rant. It's a shame it is being given the silent treatment from the people with the power to fix the concerns.

I agree the biggest issue is probably communication. Its hard to shift from one task to another, especially when working under pressure, and especially if the new task is not your forte.

The question is also how to allocate resources, and David made a point that I think I disagree with, that a foundation would make the busy people more busy. I think that becomes another resource funnel, with different but intertwined goals. Yes, somebody needs to run it, but that means someone is actually dedicated to a goal which is not so fundamental to the LLC, instead of busy doing other stuff. That way, certain aspects don't fall through the cracks. One handles the forest, and the other the trees, so to speak.

I also think its important to recognize that as a platform, there maybe be a tipping point where the user base is big enough to allow for significantly faster development. As the platform matures and gaps get filled in, the ease of new users getting started improves, and the base can grow quickly. Modx is getting closer to this point, but hasn't reached it yet. Just for example, Discuss is a tool which is very common in the market (think Vanilla), and there is almost no community built around it (hint hint).

Otherwise, I agree with David on most of his other points. Crowd-sourced funding is a good idea, and this is not really an existential crisis. Just a good vent session

For me, the issue is not so much what is going to come out tomorrow, but what is going to be supported tomorrow, and how long the real lifecycle is of a product, so I can plan my project development. I think a lot of people have questions about this.

I've attempted to stay out of this thread, but since my name has been thrown around for a bit now, I feel I have to respond.

As lossendae has pointed out, I am no longer an employee of the LLC. I left MODX Systems in July 2013 to go work for Bigcommerce. (Someone needs to remove my "MODX Staff" badge, apparently.)

Furthermore, I never had equity stake in the LLC, so I could not be ever considered an "owner". At my time of employment, I believe only Ryan Thrash owned equity in the company (with Jason possibly having a tiny amount, that of which I am unsure) - the rest of equity was owned by the investor group behind MODX Systems. I have not rejoined because of differences with the LLC.

As for the thread specifically, I have the following concerns. This is my view on the project/company having been both a high-level employee of the LLC, and having now been distant for over a year:

The LLC as it stands is not equipped to properly manage a large OSS community and/or project.
This is because the LLC is of limited membership. It has to split its duties between Professional Services contracts, MODX Cloud maintenance, and infrastructural maintenance. It simply has very little to no time for managing an OSS project as the LLC is currently structured. This is self-evident by the lack of posts here, minimal technical roadmap direction, and general silence.

There is a lack of available, strong technical leadership that can contribute at a consistent enough level.
This was prior provided by Jason and I; however, Jason is now swamped (from what I know) with Professional Services work, and I have full-time employment elsewhere that prevents me (time-wise) from making sizable contributions anymore. However, even that was never enough - MODX needs a larger pool of technical talent to lead and direct the project.

Current contributions to MODX 2.3 are only bug-fixes, or small feature additions. There’s no overarching changes to the way the system works or improvements at an architectural level; the sole exception to this being moving the manager CSS to SCSS/SASS and a build system by Grunt, which was done mainly by JP, who was an LLC employee for a time.

For the project to reach critical growth, architectural decisions will need to be made by a group of people who are of a high enough technical skill to justify and advocate for them. A Foundation could provide this, but skill would have to be upped considerably in that group for this to be effective. Decisions would also likely move slowly because of the committee-based process, and that may be “too little too late” for Revo as a product.

There is a large amount of confusion of ownership and leadership of the project.
Technically, the project is GPL, and can easily be forked and the community could own it. However, the brand is owned by MODX Systems, and therefore any fork would lose that influence. Furthermore, there are significant roadblocks/questions to establishing a MODX.org Foundation (which I personally would be 100% behind), and they are:

- Infrastructural responsibilities: the SSO system (Rails app, last I saw), the MODX core Package Provider, the Amazon S3 instance hosting Extras and releases, the modx.com website, the forums, the github account, etc. All of these require both time, people and money to keep running and maintain. Who would own and manage this?
- The LLC would have to cede responsibilities, and at some level, ownership. This would likely be an issue.
- Ownership of technical leadership and direction; I doubt the LLC is willing to cede this at this time, but a lot has changed in the last year, so maybe this is feasible


To be honest, MODX Revolution is at least 2-4 years old in its codebase. Examples:
- The PHP codebase needs to abandon pre-PHP 5.3 requirements and modernize
- Needs to use PSR syntax for its PHP architecture to increase adoptability and usability
- Abandon a “not invented here” syndrome and utilize libraries for common functions
- Related, allow support of Composer within MODX, which has fast become the standard for PHP application development
- It uses a now-outdated ExtJS3 for the manager structure, instead of a more modern framework, like Angular
- It relies on xPDO for its ORM, which had its last master commit in Sep of 2013, and last develop commit Mar 12, over 2 1/2 months ago. It would be wiser to move for MODX3 to a more universally supported ORM, such as Symfony’s or Eloquent (Laravel)
- Very minimal test suite, which needs to be expanded upon significantly
- Heavy dependence on “god objects” and fat models, rather than a more service-class based approach
- The architecture as a whole needs to be separated into packages and service architectures, to allow more extensibility and separation of concerns (too much coupling of code right now)
- Still no “Content Elements”, or user-definable fields, for Resources, as well as Resource Types. This is pretty much a standard in modern CMS’s nowadays.

And on and on. These are the things that MODX should be striving for in its development; however, I have not see much movement in that direction outside of small bug fixes and UI adjustments (which are good, but are band-aids over larger concerns).

I believe that this is mainly caused by confusion over technical ownership. There needs to be a dedicated team of integrators, who are defining and over-communicating technical direction and ensuring that commits coming in are aligned with a constantly-evolving roadmap. This, IMO, needs to be owned by the community. The LLC needs to explicitly grant this ability to the community.

The roadmap needs to be far more defined, down to what specific issues are required for a release to happen - and those issues need tickets, whether in Github or Redmine or JIRA or whatever. The tracking system is not the problem, the lack of process is. Burndowns, sprints, and such need to be implemented and used. Feature freezes need to happen prior to releases. And there needs to be a community-driven point person who helps communicate and enforce that process. The LLC has shown itself unable to have resources for this role, and therefore the community must provide it to have the development for the project sustain.

I don’t know what all the LLC is willing to cede on this. But I do know - having been deeply in the community and now out of it for over a year - MODX is aging, fast. And it’s going to take extreme, well-defined, strategic measures to turn it around and move it to the next level. All major OSS projects have this moment. Whether or not MODX does this will depend on the community taking itself to the next level of self-organizing, and whether or not the LLC steps out of the gatekeeper role it’s currently in and into more of a cheerleader. That future is yet to be seen.

I hope the best for MODX, but things have to change, and soon.

Many points you just talked about have been talked superficially more than 1 year ago in this thread and (unfortunately), not much has changed since then.

Jason have started to port xPDO to Composer, but it's seems to be pretty much on hold.

If Modx could leverage third party packages, it would more than likely bring more talented developpers to the platform, allowing Modx to evolve along with PHP without relying on a single man army.