Forums / Discussions / Waiting for mainstream

Tuesday 19 July 2011 11:33:55 pm - 6 replies

Introduction

For many years, eZ publish was good out of box solution. Since many months, I can see more often, that is easier and faster to write something similiar using alternative technologies. Rails, tornadoweb, chery.py, django are more on time.

» Read full blog post

Author Message

Łukasz Serwatka

Wednesday 20 July 2011 1:05:17 am

It would be great to read in details why do you think eZ Publish is a dead end. Looking on the amount of new projects built with eZ Publish I would say that it is completely different. So please elaborate why eZ Publish is dead end and provide some good comparison to other Rails based frameworks for instance.

First article is just an example how you can do things "traditional" way, there is another one coming where we will show how you can use mobile web frameworks such as jQuery Mobile or Sencha Touch to speed up your development efforts. I put everything I have on "table" as I'm fully in doubt if you can make RESTful application faster with Django then using eZ Publish built-in API with keeping same flexible content model.

Personal website -> http://serwatka.net
Blog (about eZ Publish) -> http://serwatka.net/blog

Gaetano Giunta

Wednesday 20 July 2011 3:37:59 am

Just a nitpick but moving to a different "technology" and "language" are not the same.

In my own very personal opinion python lost the battle for the web long ago to php.

Ruby is more trendy today, in a large part because of RoR and in part because of all the java devs who hated php and wanted to have something easy to develop with - ruby was a very good escape route to them. But js on the server is even trendier, and might in fact take away a lot of the coolness of ruby.

Php is not cool today because it's mainstream (not my words, Rasmus said it), but it is the mainstay of the web and it keeps evolving. With namespaces, late static binding, lambdas and closures, I think it is basically on feature parity with all other languages out there. And by its sheer mass, it will not wither fast: think about the number of php devs out there, hosting companies, php apps...

About "technology", I think you mean "framework". In that case, we might make a more meaningful comparison taking into account things such as django, symfony, cake, zend etc... True, a lot of those are lean and mean compared to eZ. But then again, software with age accrues in stability and features.

A few years ago I learned the hard way that Symfony did not create its cache files using atomic operations. eZP has been doing that since forever. It's not a cool feature, but it's one that big websites do appreciate when their traffic starts to scale. Same with varnish integration: the eZP plugin for doing ESI might not be the best out there, but it has been out long before symfony2 started to think about that.

Last word of advice: not every website needs a CMS, and a framework allows developers to build custom solutions. But I have a hard time seeing one developer/company taking over maintenance for an app written by another developer using a framework: there is still too much freedom to do things in many ways, and the cost of understanding the code you did not write is, imho, the same as the cost of understanding plain php code you did not write (case in point: different devs wanting to rewrite two times from scratch an xmlrpc adapter for a symfony app instead of focusing on business needs).

Principal Consultant International Business
Member of the Community Project Board

Michael Lee

Wednesday 20 July 2011 4:47:29 am

Hi Tomasz,

First of all, no matter what kind of solution you are looking for, I believe there is not an ultimate and perfect platform for it, unless you can give me a name of the perfect platform you've found.

So if there is anything that is so cool on a platforms, then it's quite possible that you will also find something that is so "bad" sooner or later and you will want to look for other platforms again.

Maybe in some aspects, eZ Publish loses to other platforms, but it still wins a lot of points in some other aspects.

Besides, given the fact that eZ Publish has been developed, improved and used for years, I don't think any younger platform can totally replace it at this moment, because I believe that time is fare to everyone.

So let's assume that all platform developers in the world are equally smart, if it took years for eZ Publish developers to make their today's achievement, I don't believe any other developers may deliver a better platform with less efforts.

Even though it's possible for the other developers to deliver a really good platform with less efforts, but it will still take a very long time before the code base becomes mature enough.

Without a doubt, the code base will eventually become mature, however, at that moment, you might find that some new technologies, best-practices or standards have been "invented" again and for you, the platform might change from a "perfect" one into a "not so good" one because you might find another younger platform that has better compatibilities to the new things. 

From my point of view, if we say a platform is "good", it means:

- It's “health”: it does not have critical design defects. For example, stupid database schema, very bad performance and etc.

- It's “clean”: well designed at the code base level including nice object, logic tiers modelling and etc.

- It's “mature”: it's a grownup project, not just a kid.

- it's "sexy": it's very user-friendly.

- it's "conversational": well-documented, nice API and etc.

- it's “popular”: it's widely used and that's a proof of the its value.

- it's "well-educated": compatible with popular industry standards and best-practices.

- it's "smart": it's able to "learn" new tricks. In other words, it's easy to be extended.

- it's "evolutionary": the official development team keeps it improved.

- it's "educable": new industry standards and best-practices can be integrated into the platform.

- It's "generous": variant resources including knowledge, technical support, plugins/extensions and etc are available through its community.

- it's "skillful": it may fulfill most of your requirements.

- it's "loyalty": it tries its best to keep back-compatibility and also changes itself at a reasonable pace, so that it gives you enough time to refresh your knowledge and experiences.

- it's your best "friend": it has to be your most familiar platform and you know almost everything about it. You know exactly what it's good at and what it's not.

From my point of view, of course, eZ Publish is not perfect for all of the items in the check list, but still it does a great job for most of them.

I still believe that other existing platforms may also have a high score for the check list above, however, don't forget the last check point: it has to be your best friend.

So If, however, you found two platforms that are as good as each other, but you are more familiar to one of them, then no doubt, you should go with the one that you are familiar with.

If you did find something that one platform can provide while the other one can't, you should confirm if that "thing" is a: smart idea, a best-practice or an industry standard.

From my point of view, smart idea generates best-practice, best-practice generates industry standard, while industry standard finally brings evolution.

So it can be risky if you are relying on a 3rd party great idea before it becomes a widely-accepted best-practice or industry standard because it may change anytime without notice.

On the other hand, again, it's just from my point of view, eZ Publish is a bit slower than other platforms in response to the "new things" because the development team of eZ Publish is waiting for the "new things" to grow up into "best-practice" or "industry standards".

So like I said, eZ Publish is "loyalty" to its customers and developers by keeping its back-compatibility and integrating high-quality best-practice and industry standards. As a friend, shall we be a bit "loyalty" to eZ Publish as well?

We don't give up girl friend just because we met the other girl who is more sexy, do we ;)

So why not talking to the eZ Publish development team about your concerns and requirements and be a bit patient to wait for them to "re-educate" or "evolve" eZ Publish in its next release?

Please feel free to get back to me with your comments.

Kind regards,

Michael

Michael Lee | Managing Director | ZerusTech Ltd | www.zerustech.com

Skype: zerustech

Nicolas Pastorino

Wednesday 20 July 2011 8:07:08 am

What a comprehensive reply Michael, excellent !

--
Nicolas Pastorino
Director Community - eZ
Member of the Community Project Board

eZ Publish Community on twitter: http://twitter.com/ezcommunity

t : http://twitter.com/jeanvoye
G+ : http://plus.tl/jeanvoye

Gaetano Giunta

Wednesday 20 July 2011 9:46:31 am

"We don't give up girl friend just because we met the other girl who is more sexy, do we"

Best quote ever in an eZP context, sir!

Principal Consultant International Business
Member of the Community Project Board

Nicolas Pastorino

Wednesday 20 July 2011 10:17:56 am

"

"We don't give up girl friend just because we met the other girl who is more sexy, do we"

Best quote ever in an eZP context, sir!

"

+1 !

--
Nicolas Pastorino
Director Community - eZ
Member of the Community Project Board

eZ Publish Community on twitter: http://twitter.com/ezcommunity

t : http://twitter.com/jeanvoye
G+ : http://plus.tl/jeanvoye

You must be logged in to post messages in this topic!