Drupal 7/8/Backdrop Upgrade for Warmshowers.org

[rfay]

Sadly, we'll eventually be pushed to a Drupal 7 (or possibly Backdrop CMS: http://backdropcms.org/). Upgrading to Drupal 8 is also possible, but it would require waiting until lots of modules mature in Drupal 8. (But there is supposed to be a direct migration path from D6 to D8)

The things that are pushing us now:

1. Drupal 6 support will end shortly after Drupal 8 comes out, probably in several months.
2. PHP 5.3 support has already ended, and Drupal 6 can't really run on a newer PHP.
3. Modules for Drupal 6 are rarely being maintained at this point, so we end up with possible security problems and with far fewer options for problemsolving.

This is a huge job. It's been a huge job when we did it in the past. It is in fact blocking us on lots of things though. I think we'll want to consider paying for development work on this.

In addition, our theme will need to be updated, and of course we might want to refresh it in the process.

[mihaeu]

With so many people behind Symfony and Drupal 8 shouldn't the modules mature quickly? Also going for the newest version might attract more devs. I myself have not worked with Drupal 6 before and somehow learning something old which is not reusable (I mean I wouldn't start a new project on 6) has always put me off. I think going for 7 would be the same problem for some people. The changes in 8 are exciting so why not go for it.

That being said I have no idea how much work you're actually talking about :/ any idea roughly how much more work it would be to go 6 -> 8 instead of 6 ->7?

[rfay]

@mihaeu I expect that any of the three options will be an enormous amount of work. As in the OP above, it's not a given that we'll do D7. But any of the three will probably take 200+ hours of work. It's unknown what the performance characteristics of D8 will be or how fast important contrib modules will appear. Basically, I'd be in favor of D8 for sure if we could wait a couple of years after its release (and perhaps we can).

[mihaeu]

Performance wise, as long as the site is upgraded to 5.6 or 5.5 it should have a very positive impact on all of the options you mentioned, right?

Did you consider going headless? Would allow for (slightly) easier upgrades in the future and would also help with the iOS and Android apps.

OT: I was on the bike for two years and am back since some time now. I wrote you, I think about two years ago, but I didn't have the time to help while on the road. I could barely keep up with the work for the customers that kept me going. Then after a couple of months I fell in love in Iran and traveled with my girlfriend (now wife) so there was little time. Bicycle adventures ...

Maybe now would be a good time, to give something back and get started. I'll try to get everything set-up and maybe I can tackle a few of the open issues. Any suggestions for issues that could get me started?

[rfay]

Congratulations on so many wonderful things in your life! Wow, what a great time of your life!

The PHP version will have very little impact on performance, IMO.

Headless would be a cool idea, with lots of potential, but a complete, overwhelming design redo.

The problem with big projects like this with volunteers is it's so hard to have enough attention span to get through them. Just a simple D7 or D8 upgrade will be months and hundreds of hours, and all of us as volunteers have a tendency to get distracted by a new project, etc. We've seen so many (much smaller) attempts left by the wayside, still incomplete.

As far as getting going on development, let's plan a skype or hangout or something. I can get you all the resources you need, and we can figure out what might be a good start. Just use the contact form to get me and we'll get started!

[mihaeu]

Ok, thanks @rfay, will do. I'm off to Italy for a week, after that I'll contact you.

[rfay]

Chris and I had a great conversation on this and the WS funding issues on Friday.

The key outcome: We should try to figure out the latest possible migration date and then plan to be done by then. That means we don't have to make a decision for some time and it will help us to have more info as time goes on.

The things that are driving migration:

• PHP5.3 is EOL
• D6 will be EOL

Mitigating factors:

• Ubuntu seems to be supporting PHP5.3 on our Ubuntu 12.04 through April 2017
• Drupal 6 will be supported until 3 months after Drupal 8 is released. That means we have at least a year before EOL, I would guess. There are more than 100 critical issues for D8 at this point.

I attempted to take D6->D8 migration for a quick spin with a trivial fresh install of D6 and was completely blocked by both my own inexperience with D8, D8 migrations, and nonexistent docs. Apparently nobody has actually done it and survived to tell the story at this point. However, one of the maintainers reached out to me about it and offered to help, so I'll try again when he has time for it after he gets back from vacation.

[rfay]

I've spent some time evaluating other options, including learning some python and trying out django and django-cms. Of course I'm overwhelmed by what a Drupal developer doesn't know about another platform, and also the multitude of things that we "just know how to do" in Drupal that we have no idea whether will work or have to be developed for another platform.

My bet is that we'll end up moving (probably next year) to mature Drupal 7, then waiting to decide between Backdrop and Drupal 8 a few years later depending on how both of those work out. In theory Backdrop would be by far the best for us, but it's not clear how much support there will be. But some time will tell.

[kalpaitch]

Closing as no longer relevant, can evaluate upgrade options in a few years.