I too am once again evaluating the upgrade of Zend Studio from 5.5.1 to 9.x.
I evaluated 6.1, 7.0.1, 7.2.0, 8.0.0, 9.0.0 and now am trying again with 9.0.2.
It truly is the "beast from the black lagoon."
My reasons to upgrade:
- have a tool with current coding features
- have a supported tool
- be able to run an OS X native version and thus remove one more mission-critical function from dependence on a Windows Virtual Machine.
My reasons not to upgrade:
- vast numbers of features I can't understand
- vast numbers of features I don't need
- bloatware and slowness
- it does not "just work", whereas 5.5 does "just work"
My list of mandatory features is this short and simple list:
- edit code
- integrate with SVN (not absolutely essential, as I am used to using a separate, standalone client TSVN).
- access mysql database (could probably live without by using some other tool as well)
So I am not demanding on the features list. But I am very demanding on the "it must work" list (My new tagline: "I will gladly trade you 250 new features for one 'it just works.'") The newer additions such as jQuery libraries, etc, would be very nice, but only if we get past the basic list first.
If I could get those items working smoothly, I would probably migrate to 9.x. But trying to do the simplest things in the current version of Studio very often results in all kinds of baffling issues using obscure/obtuse/Eclipse terminology and concepts. The very latest example of that is in trying to perform Search-in-files. It does not find anything, perhaps because my project is not "in a workspace", even though it seems to parse all the files.
The problems I have had with the versions between 5.5 and 8.0 include the following:
* having ZDE recognize the installed Zend Server with such fun scenarios as ZDE saying "no server found" but when going in to configure a Server, it was already there and would not let you set it because it is already set up
* issues with the ports used by Zend Server breaking things
* getting SVN to work
* getting debugging to work
* getting mysql client to work
* many, many other issues/bugs previously reported in connection with the earlier products.
I am glad to say that most of the above are no longer are a problem, but am including them as background.
Every developer expects the following from his tools: simple things should be easy and hard things possible.
My experience with each version of the Eclipse-based beast from the black lagoon has always been the opposite: that the simple things are difficult if not impossible. But 5.5 just keeps chugging along, doing everything with a minimum of fuss, and a much simpler mental model of the workflow.
There used to be in the documentation some "help" for those converting from 5.5. However, it was useless, because all it did was give the new menu paths to do typical tasks (which any developer could figure out) instead of explaining the new concepts and workflows that are needed.
Likely I will spend a bit more time on this evaluation in the hopes of surmounting this painful learning curve. But my past experience shows that it is more likely that I will get to a point of frustration as I have in the five previous attempts and just stay with the obsolete-but-working-well 5.5.