|
Thursday 22 January 2009 3:03:03 am
Hi there; eZ is such a great engine, yet it only two drawbacks which can be crucial for eZ future... I understand that learning curve is steep and it's an advanced engine, but these issues don't fall into that category. 1. Requirements - many freelancers will chose for their projects the open source software they can install on shared servers. The extensive requirements will make eZ less and less popular. 2. Speed, performance. - it's needed to have an accelerator, otherwise the site is extremely slow. The install engine itself can crush due to php execution limit, which most of people can not change. Also, to improve performance, I have to resign from the most beautiful eZ feature: fetching content. So what's the point with this feature?... I want to fetch content and have a fast site... People adore speed, they love surfing quickly. There's nothing more irritating like long-loading pages. Speed is at the moment the only thing that leaves me in doubts as if to change the engine of my long-living website to eZ or not... my current engine works lightning-fast. --- The question is: do you want to make eZ popular? If you want it to stay in it's niche, go ahead... but if you want it to become popular, you have to make it appealing... speed and accessibility are the keys. People don't need to know that they're editing some ini files... people don't need to know the basics of template language (you can make many basic processess automatical, for example: have a template edit section with a list of template operators ready to use, similarly as you did it with fetching and embedding content system) and should not e.g. manually assign the copy templates to just copied class. The override system should be very clear. People should work as little with source files as possible, and the development should be as fast as possible. And when I spend most of my time only on setuping eZ and bringing up the features that in other engine I already have... what's the point? My point is to not waste time... to make things fast. I might sound like a noob for you. Maybe I am... but you, working with eZ for your daily bread, don't neccessarily feel how much operations you have to make at once to change a little thing. Editing files and uploading them through ftp should be an option, not neccessity.
For example, adding an attribute to existing class should be two-step max:
1. Adding an attribute in admin panel (e.g. image)
2. Admin panel forwards you to opened assigned template files, ready for editing.
Copying a class for making a new one should be like:
1. Press copy
2. Receive a complete class copy with already assigned, copied templates The less thinking, the better! Now I have to check all the templates assigned to old class, manually copy them, change and upload, and then manually assign... maybe you know all the templates by heart, but I don't, especially as you have to surf through limitless numbers of folders in eZ content tree to get that one freaking template. Where's my time? It's lost. These are the things that pain me most. I think you already are heading for better class creation automation, the sooner the better! :) Custom class creation and content fecthing are for me key features of eZ, they should be as simple and intuitive as possible. Don't worry, it won't get too simple, there's still much to know to create extensions and datatypes. And once again: bring SPEEEED to eZ :D Thanks for reading...
Shiki soku ze ku...
|