If you want heavy load, I can mention some PosgreSQL clusters, the decision where an application is going to connect to, is definitely not the job of an application.
In some cases I've done multi master replication, that would by funny with an application outside the database process and interpreted language.
Currently I'm programming on an Oracle cluster, and it is not my job as a programmer, to make decisions about which node my application should use.
On MySQL it is normally done by one database as replication master for reading data, and slaves as much as needed for reading. The decision where a connection will go to is the job of the load balancer, also not the application.
I really don't like ExtJS, why should I start working on things, that would not come in mind for using them at all?
The reason for me to use Revo might have been about two or three years ago with a translation work-flow for documents implemented, but at that time Revo wasn't ready and it is without this until now. Translations are something, that I've needed all the time working with Evo. I've done something with snippets and TVs, but that was only workaround, but better than nothing.
What was the reason to change such a lot of things, that the old modules and snippets are not compatible with Revo?
I don't want to rewrite mine every some years. This is a behaviour, that I do know from a well known company in Redmond.
Currently I'm using Evo to support the old installations, and to have some fun. I really can't see any of that fun for me in Revo.
On the database side I will use database functionality, if available. On MySQL with InnoDB constraints are working, without full text indexes, of course. And I know that crashed InnoDBs end up in buying support from Oracle.
With PostgreSQL I have both, constraints and full text indexes and I haven't seen a crashed one in sixteen years.
The same is on MS SQL Server or Sybase Adaptive Server Enterprise, once there was a time when they where one, the better one is still Sybase.
But I want the user to have the choice to decide, which one he/she likes, MySQL with MyISAM or InnoDB, PostgreSQL, SQL Server, ASE.
For example I'm thinking about a GIS extension for Evo, of course only with PostgreSQL and PostGIS. It is fun, what one can get out of databases. You will not be able to do this in the same speed with an interpreted language.
Gone away and found a better place to stay
-
☆ A M B ☆
- 24,524 Posts
We're dealing with two different desires here. Personally, I want a light, quick CMS. The MODx development team has gone in a different direction, and I have no quarrel with that, I just have made a personal decision to continue with what has made my life a pleasure for the last 6 years or more. You like driving a bus and I like driving a sports convertible. If I ever need to develop a heavy web application that will run on its own dedicated server, Revo would most certainly be my choice. As long as I'm building relatively simple sites, it's just massive overkill.
I see no reason why there can't be room for both. Why try to kill Evo and force everybody onto the bus?
-
☆ A M B ☆
- 24,524 Posts
Has anybody else looked into Auditor for Evo? It seems extremely promising.
http://modx.com/extras/package/auditor
http://modx.com/extras/package/modxauditoranalysis
I would like to see the analysis application re-worked as a module rather than being a separate, customized MODx installation.
I haven't seen that thing before, thanks for pointing it, Susan.
This goes exactly in the direction, where I want to go to and what I had in mind.
I think there should be a configuration page, to set this active. The triggers are the correct way to implement those things, for me.Nor
As this thread goes on and on, I'm thinking about a list with all those ideas, not to miss any of them. What about wiki pages on github?
Gone away and found a better place to stay
-
☆ A M B ☆
- 24,524 Posts
There are several "must-have" add-ons for Evo, or that could certainly be developed into killer add-ons.
YAMS - not being developed any more, latest beta needs work
WebLoginPE - needs work
SkinGraft - a template management module that could easily be morphed into a general add-on manager
Auditor - version control, needs updating
MXCalender - events and calenders
Easy2Gallery - my favorite gallery app, could use front-end management features
I add to the list:
PHx
getField
TemplateRules
Quill
CodeMirror
-
☆ A M B ☆
- 24,524 Posts
There's something wrong with the regular expression used for PHx's garbage collection. It eats everybody's placeholders, not just its own.
-
☆ A M B ☆
- 24,524 Posts
I've never looked at Quill before; so far it looks like Custom TVs does much the same thing?
I am not sure, I don't understand Custom TVs, there is no documentation. I use Quill to create a dropdown list of resources in the manager. I can choose the parent, depth, identation, fieldname (in my version of Quill, which is different from the one in the extras section, I can also exclude resources by ID) and much more.
Anyway, I think that it is similar to the @SELECT binding.
☆ A M B ☆