Do we really need an Evo migration path to MODX3?

There are users that decided to continue using Evolution when Revolution was released in 2010, and while I think that number is going down (don't have data to back that up, but just seeing less and less of Evo), there still are users building new sites with Evolution today. I do not want to go into the Evo vs Revo debate here or disrespect Evo users, but... despite recent-ish releases I think building new sites in Evo is a bad idea. The only time it really is updated is when a critical security issue finds its way to the MODX security team, at which point the pull requests are reviewed so it gets some bug fixes and small features too.

At some point last year there was this uproar of Evo enthusiasts which basically boiled down to people venting their frustration about Evo not being developed further. Difficult questions (with plenty of internal heated discussion too), several forks and bad PR. The answer at the time was a blog post "Clarity, Focus and Simplicity" which for the first time mentioned MODX3 and introduced some ideas for it. It also provided some clarity on Evo's (and Revo's) End Of Life (EOL): after the release of MODX3, Evo/Revo will be provided maintenance releases for broken functionality and security issues for at least a year, and security for another year. So, kinda what has been done for Evo since 2010.


There's one thing, though, that I don't like about what was promised for MODX3, which is a migration path from Evo to MODX3, which was repeated yesterday. I can understand the need from users, but there's not even a straightforward migration path from Evo to Revo despite promises throughout the years (Provisioner somewhat helps). How on earth is a migration path from Evo to MODX3 going to happen?

I'd like MODX3 to succeed and having to keep backwards compatibility or migration paths (from something that we can't even migrate easily now!) is going to hold MODX3 back. And then, when MODX3 launches, but there's no easy migration path, the community will be disappointed and a promise will have been broken. If Evo users did not want to move to Revo, I don't see why they would want to move to MODX3.

I want to suggest the project team to make a difficult decision and reconsider the Evo migration path promise. Announce Evo's end of life - providing links to the alternative forks - while releasing security releases for the next year as has been done for the past years, so the Evo chapter (which has been a good ride) can be closed in 2014 instead of 2016.

Is it necessary for the MODX team to worry about it? That is debatable, personally I have the skill set to write my own migration tools (https://www.revitalagency.com/blog/modx-evo-to-revo-site-content-migration-tutorial/), so I'll gladly do it again for evo to MODX3-vo. I'd rather the team focus on moving forward not backward.

Quote from: revitalagency at Sep 09, 2013, 04:27 PM

Is it necessary for the MODX team to worry about it? That is debatable, personally I have the skill set to write my own migration tools (https://www.revitalagency.com/blog/modx-evo-to-revo-site-content-migration-tutorial/), so I'll gladly do it again for evo to MODX3-vo. I'd rather the team focus on moving forward not backward.

Exactly - I think MODX has bigger fish to fry than Evo. If Evo is relieved of active duty by MODX, that's one less thing to worry about and more time to spend on building awesome.

I do not want to list the shortcomings REVO and answer the questions why I'm still actively using the EVO.

To close or not to close the support team to solve MODX but given that they have already given into the hands of community EVO is the only thing that changed is the name and the domain where the EVO will be available.

No need to kill a good product!
We need to do better then the community itself will opt for a new product.

I am willing to give up the first EVO in favor of the MODX3
When he met my expectations and would speed, requirements and at least at the level of EVO

I do not work in a large company, and basically all of my projects are small and fast, and then I do not need unnecessary technology that only slow down the development and debugging time but did not give any advantages

Well, all the great features that are in REVO and which was lacking in the EVO for a little bit implement
for instance repository: http://extras.evolution-cms.com/

Without starting the debate again over which is better, I would just like to say:

I have no problems moving to MODX3 when its ready - so long as it does what has been promised.

If its as fast and as easy to use as Evo and not cumbersome like Revo - then great, I will convert all of my sites and I will talk my clients into upgrading

Ive tested Revo, found it to not suit what I needed (both personally & commercially) and so reverted back to Evo

The other point to make is: why should users be forced to convert from "Evo" to "Revo" then to "MODX3" ?

Just because Revo came after Evo - Why should there be no tool for Evo ?

Just to clear matters:

Quote from: markh at Sep 09, 2013, 04:17 PM

while I think that number is going down (don't have data to back that up, but just seeing less and less of Evo), there still are users building new sites with Evolution today.

As of publishing this post: The latest Evo release has almost 14000 downloads

This indicates to me that Evo is still being used a lot, not to mention older sites that aren't updated on a regular basis.

Quote from: markh at Sep 09, 2013, 04:17 PM

The only time it really is updated is when a critical security issue finds its way to the MODX security team, at which point the pull requests are reviewed so it gets some bug fixes and small features too.

Ths is not correct: please take a look at the tracker for the next release

The developers that are working on Evo are doing a great job, so saying that Evo isnt being supported isn't correct.

No disrespect to the people working on Evo - I'm glad people are helping . I based my comments of inactivity on https://github.com/modxcms/evolution/graphs/commit-activity which showed some activity around the time 1.0.9 and 1.0.10 were released in April, and nothing since.

A 1.0.11 PR from Yama that was just merged today by Jay changed that graph with some more continued activity. What that tells me is that people are working on it more than I thought... but that development is not taking place in the MODX repo, which backs up that MODX LLC is not actively involved with Evo and that I think that should be acknowledged.

Isn't the community capable of running the Evo project? It could still be on the MODX website, discussed on the forums etc, but it would be clear it's a community project and not actively supported by MODX..

The other point to make is: why should users be forced to convert from "Evo" to "Revo" then to "MODX3" ?
Just because Revo came after Evo - Why should there be no tool for Evo ?

Revo was developed as what MODX was supposed to be.. there's a reason Evo was for the longest time 0.9 - Revo was meant to be the final 1.0. At some point that was changed to make Evo 1.0 and Revo 2.0, but IMO the version numbers are proof enough that Revo is considered the next version from Evo. They weren't meant to be two distinct products.

I don't mean to disrespect Evo, its users, or the community building it. But I do think that LLC needs to focus its attention on one, not two, supported versions, with Revo as most recent and most popular (68K downloads in the time Evo got 14K). Especially with MODX3 about to be shifted into gear, Cloud 2.0 etc.. they've got a lot to do and I think the community is more than suited to maintain Evo and, if there's a need for it, develop migration tools for Evo->MODX3.

@ Mark Hamstra

I think you should look at this from a different perspective.

If MODX3 turns out to be what Ryan has said, the EVO / REVO debate will vanish.

Simply saying no to a migration tool for Evo users will probably push them away from even looking at MODX3.

They will probably end up turning their backs completely on MODX. I can't imagine that the MODX team want to lose multiple thousands of users and community members.

My personal Interest

I will look at MODX3 when it arrives and hope that it's what Ryan has said it will be - the best of both Evo + Revo in one system.

With that said - if MODX3 turns out to be a great system and there is a migration tool I will glady move all my sites over and I will market it further to my CMS clients.

I have sites with over 4000 pages / ID's, chunks, templates etc. - so a move per hand is not realistic without intensive work, money and time.

I think a Migration Tool for anything MODX is normally quite hard to commit to, due to the nature of MODX as a CMS. There is no standard setup of a MODX website, everyone has their project setup differently, based on the default install we can have a Migration Tool, but with third party extras, I guess that will be down to the extra developer to provide a migration path.

I am more for some documentation to help developers easily migrate over to MODX3. I think whether it be from Evo or Revo, it shouldn't make a difference. I understand why some users still like using Evo over Revo and I think we should cater for them in any transition to MODX3.

I also think a service (paid) to help companies migrate over content, customization heavy websites to MODX3 would be a good way forward also.

It's interesting that you want to move all your sites over to whatever platform you prefer at one time - there's always the right tool for the right job, and unless you're on a generous maintenance plan with your clients, I wouldn't include migrating to different platforms in it. When I switched from Evo to Revo, I didn't migrate all sites I had build - I just started using Revo for new sites, and the odd site that needed a redesign or had new requirements would get Revo in the next iteration... it's not an all or nothing.

Maybe I am being too pessimistic (I'd like to think it's being realistic) about the chances of MODX3 living up to everything that has that been said... what exactly is the best of Evo and the best of Revo to make up MODX3? I'm willing to bet different people would answer that differently and not everyone will be pleased with the specific MODX3 direction.

Just look at the CMS space and how many different ones exist out there, each with their own specific approaches and ideas. I'm sure MODX3 will be a giant leap forward and there are real problems to solve in Revo that MODX3 will fix (possibly by looking back at Evo), but there's no one size fits all.

If it's possible for there to be an easy migration from Evo (and Revo for that matter), that would be great. I just don't want it dictating MODX3 development.

Quote from: markh at Sep 11, 2013, 10:14 PM

It's interesting that you want to move all your sites over to whatever platform you prefer at one time - there's always the right tool for the right job, and unless you're on a generous maintenance plan with your clients, I wouldn't include migrating to different platforms in it.

You are right in saying that theres always the right tool for the right job - well, Evo has been that tool for the last 7 years for my sites and those of my clients.

To date, I haven't failed on a site - all of my needs, those of my clients and their special requirements have all been met using Evo.

Ref maintenance plan: I am in the lucky position that my clients trust what I have to say and if I tell them that an upgrade is required to future proof their sites due to improved security, technology etc. they will do it and pay the bill at the end without any problem.

Quote from: markh at Sep 11, 2013, 10:14 PM

it's not an all or nothing.

It is if the system is no longer supported unless you can support it yourself.

Quote from: markh at Sep 11, 2013, 10:14 PM

what exactly is the best of Evo and the best of Revo to make up MODX3? I'm willing to bet different people would answer that differently

I fully agree.

I'm not going to write down what I would like to see, this is not the post for that, and to be honest, thats already been chewed through more than once