Upgrades from 5.6.x to 5.7
Permalink 1 user found helpful
I'm disappointed.
I had believed that by using Concrete5 for my customers and I, I would be providing a future-proofed platform, a leading platform, an awesome platform, but now I read that the upgrade path from 5.6 to 5.7 doesn't exist and probably never will.
I appreciate that this platform is free, and has cost me nothing to use, but it has cost me a lot of time and effort in investing in C5 and spreading its great reputation to fellow developers, and now I'm left feeling rather red-faced that they have to consider manually migrating websites of vast size/scale.
Am I wrong with the statement that an upgrade path will never exist? This decision has just baffled me... :(
I had believed that by using Concrete5 for my customers and I, I would be providing a future-proofed platform, a leading platform, an awesome platform, but now I read that the upgrade path from 5.6 to 5.7 doesn't exist and probably never will.
I appreciate that this platform is free, and has cost me nothing to use, but it has cost me a lot of time and effort in investing in C5 and spreading its great reputation to fellow developers, and now I'm left feeling rather red-faced that they have to consider manually migrating websites of vast size/scale.
Am I wrong with the statement that an upgrade path will never exist? This decision has just baffled me... :(
To be honest, having just had a read of some 5.7 related threads, it doesn't look like a release that I would want to move to any time soon due to many issues, primarily around bugs, stability, administration and building confusions - it looks like it really is a new application rather than just a progression from 5.6.x - it might be time to consider a new CMS if 5.7 is as painful as some are suggesting :(
You have reason! Many bugs are reading in the posts, and not automatic installation option.
First of all, I'm not speaking on behalf of the core team, just sharing something I know about this matter.
You're probably right that there will never be a complete 100% upgrade path from 5.6 to 5.7. It's just not technically very easy to do and it would require all of the add-on vendors to provide their own upgrade paths for each and every add-on you use. And the same for any custom add-ons developed for those sites.
However, there have been plans by the core team to provide some kind of utilities that help you in the migration process. No, they won't be tools that bring a 5.6 site 100% to 5.7 but rather tools that help you in the technical migration process.
Having said all that, you're perfectly safe and "future proofed" staying with 5.6. The official promise from the core team is that it will be supported AT LEAST a year from the date when 5.7 was released and probably the unofficial support will continue even after that. Just like you, there are many others as well who have many sites built on top of 5.6 so it is a mutual interest to keep it up to date and safe. You probably won't see much new features in 5.6 but you're guaranteed to get any necessary security updates and bugfixes along the lines of what is said here about the support period.
You're probably right that there will never be a complete 100% upgrade path from 5.6 to 5.7. It's just not technically very easy to do and it would require all of the add-on vendors to provide their own upgrade paths for each and every add-on you use. And the same for any custom add-ons developed for those sites.
However, there have been plans by the core team to provide some kind of utilities that help you in the migration process. No, they won't be tools that bring a 5.6 site 100% to 5.7 but rather tools that help you in the technical migration process.
Having said all that, you're perfectly safe and "future proofed" staying with 5.6. The official promise from the core team is that it will be supported AT LEAST a year from the date when 5.7 was released and probably the unofficial support will continue even after that. Just like you, there are many others as well who have many sites built on top of 5.6 so it is a mutual interest to keep it up to date and safe. You probably won't see much new features in 5.6 but you're guaranteed to get any necessary security updates and bugfixes along the lines of what is said here about the support period.
OK well a toolset that aids the transition will be very much appreciated - the current templates work seamlessly with 5.7 or do they also need adapting?
If by "templates" you mean 5.7 themes, the answer is kind of. 5.6 themes will probably work with 5.7 with minor modifications to them. I suggest taking a look at this screencast (and documentation page):
http://www.concrete5.org/documentation/developers/5.7/designing-for...
But there are probably some features you'd want to take advantage of in 5.7, like combination and minifying of the assets (CSS and JS) and using LESS for your stylesheets. These are somewhat covered by the screencast linked above.
If by "templates" you mean single page views and such, the answer is probably. They might work but if you're customizing e.g. a core based single page view, some variable names might have changed.
If your 5.6 pages use a lot of layouts, you're gonna have a bad time migrating. They are completely revamped in 5.7 to support the grid structures in different CSS frameworks, although you're still able to define freely modified "layouts" as in previous versions. But if you would do the migration manually anyways, there's not much difference in the time spent moving your content. This is only a problem for any automatic migration scripts.
http://www.concrete5.org/documentation/developers/5.7/designing-for...
But there are probably some features you'd want to take advantage of in 5.7, like combination and minifying of the assets (CSS and JS) and using LESS for your stylesheets. These are somewhat covered by the screencast linked above.
If by "templates" you mean single page views and such, the answer is probably. They might work but if you're customizing e.g. a core based single page view, some variable names might have changed.
If your 5.6 pages use a lot of layouts, you're gonna have a bad time migrating. They are completely revamped in 5.7 to support the grid structures in different CSS frameworks, although you're still able to define freely modified "layouts" as in previous versions. But if you would do the migration manually anyways, there's not much difference in the time spent moving your content. This is only a problem for any automatic migration scripts.
Franz has pointed this out before, and it's worth reiterating here. Look at Drupal. That platform NEVER offers an upgrade with Major releases. If you had a Drupal 5 site and wanted to upgrade to 6, you were out of luck. The same with going fromn Drupal 6 to Drupal 7, and when version 8 comes out, it will be the same again. The only option will be a migration. In fact, with Drupal I believe it's so difficult that many times a site will remain at it's version number until it's completely redone, rather than ever getting migrated!
Having said that, Concrete5 has not been "completely" redone in the sense of being a completely new CMS. However, at this point in time, it was just necessary to go this route in order to make the deep fundamental changes necessary to take advantage of all the advancements in the web world and the technologies used, such as PHP itself!
Put another way, had this not been done, then your clients would ultimately be left with a an obsolete solution anyway. And remember, Concrete5 had for a long time allowed seamless upgrades, and now that this major upgrade is done, that's the way it will be again, for another long time. So at least now everybody has the opportunity to do a migration, when the time is right, and as a result will be "future proof" if desired. And as it's been stated, there are plans for tools that will help.
Besides, all websites need to be redone at some point, or they become "dated" to say the least.
Having said that, Concrete5 has not been "completely" redone in the sense of being a completely new CMS. However, at this point in time, it was just necessary to go this route in order to make the deep fundamental changes necessary to take advantage of all the advancements in the web world and the technologies used, such as PHP itself!
Put another way, had this not been done, then your clients would ultimately be left with a an obsolete solution anyway. And remember, Concrete5 had for a long time allowed seamless upgrades, and now that this major upgrade is done, that's the way it will be again, for another long time. So at least now everybody has the opportunity to do a migration, when the time is right, and as a result will be "future proof" if desired. And as it's been stated, there are plans for tools that will help.
Besides, all websites need to be redone at some point, or they become "dated" to say the least.
Fair points, I think that I'll let 5.7 evolve significantly before considering a migration to it, I read of many issues/concerns with it so far, but then I guess that it has to be treated like a new fresh/application.
thanks for this post.
I thinks it's fair to say that there are a lot of unknows and that is a scary thing. I for one have invested a large amount of time in c5 and have benefited from the platform. Having enjoyed that success new version represents a unknown risk. We need clear and concise communication.
It is a shame there isn't more detail around upgrade / migration options. However until the new platform is stable I think we need to relax give the great core team some time and space. But maybe nudge them to give a couple more updates :)
All very promising and great work so far guys
I thinks it's fair to say that there are a lot of unknows and that is a scary thing. I for one have invested a large amount of time in c5 and have benefited from the platform. Having enjoyed that success new version represents a unknown risk. We need clear and concise communication.
It is a shame there isn't more detail around upgrade / migration options. However until the new platform is stable I think we need to relax give the great core team some time and space. But maybe nudge them to give a couple more updates :)
All very promising and great work so far guys
Well, at least you weren't using Symfony 1.4 because that upgrade was a disaster... but that happens as architectures evolve. The bigger popint is, it's opensource, so you can continue to develop in it no problem, make changes, etc, etc... it's not broken, just not bleeding edge
What really Do wish existed was at least a chearsheet that says...in 5.6 you did it this way (example)... now you do it that way (example).
There are SO many things that have changed that have no example, documentation or even user complaints (yet), so even a simple change takes hours of research and digging through code.
What really Do wish existed was at least a chearsheet that says...in 5.6 you did it this way (example)... now you do it that way (example).
There are SO many things that have changed that have no example, documentation or even user complaints (yet), so even a simple change takes hours of research and digging through code.
> What really Do wish existed was at least a chearsheet that says...in
> 5.6 you did it this way (example)... now you do it that way (example).
This might be close...
http://www.webli.us/cheatsheet/doku.php...
-Steve
> 5.6 you did it this way (example)... now you do it that way (example).
This might be close...
http://www.webli.us/cheatsheet/doku.php...
-Steve
nice find!
Thanks
(I wish they allowed comments, 'cause some of their examples don't work)
Thanks
(I wish they allowed comments, 'cause some of their examples don't work)
We have about 100 client sites and there's no way we will ever have time to upgrade them. My clients probably won't pay for time spent on upgrading to 5.7 either.