Forums / Cotonti / General / Future Genoa

I want to know if any new development (other then bugfixing) is done on Genoa.

E.g. I see:  CoTemplate 2.7, Jquery 1.71
Those are not difficult things to put in 0.6.21....

I have looked at Siena (internal coding) and I have seen some changes that are, for me, NOT an improvement.. (well thats just me I guess, I like it simple).
I thought all the classes and the database layer class is not making the code more easier to read unless you really dive into it or know it.
(Dont get me wrong here, i think OO is a great .... BUT there is a big disadvantage as well), all this class stuff is making it harder to read, find, adjust the core code...
(Well the 3 or 4 core developers are going to kill me having said this... but i see what you are trying to do... and they try real hard.. so no flame intended..)
 

Back to Genoa >  Q: Is Genoa getting improved Yes or No.

 

Shortly, no new features are being implemented for Genoa. CoTemplate 2.6 was an exception because it was highly requested and has been tested on Siena well enough and it can be used with Genoa almost painlessly. Probably 2.7 will be available in Genoa as well if it is known to be reliable enough. As for jQuery 1.7.1 it is more questionable, we need to be sure it doesn't break any existing plugins if we update it.

In the Cotonti Team we don't have an explicit intention to halt the Genoa branch and force everyone to Siena migration. But we are a very few people in fact and we've made our choice in favor of Siena.

I don't mind if somebody continued the development of Genoa branch. But there is one major requirement: it must be absolutely backwards compatible with previous Genoa releases (i.e. it must not require any changes in plugins or templates and it must not break any existing sites).

@ ez

you already have one user :) up till today i find Sienna meh.. to much change... for the change? I dunno, but most implemented things are for the real developer, while userfriendlyness is kinda going out of the window.

before cotonti programmers jumping onto me and giving 1000s of reasons why sienna is better.. I am a user, and experience it the way I discrib above. Just an objective assessment, that's all.

Gert and other devs,

There are few things you might consider..

A lot of members just converted (1 yr ago) to Genoa, and now another big change is on the way. And you basically say that Genoa is stopping.
This is at least, for me, not very motivating... (I wonder how many ppl are still on SED because of plugins they cannot convert?)

Cotonti is a really small community, and core developers are less then 5 (3 ?).
If you make the core and also other parts harder to read with all sorts off classes and OO code, new unexperienced developers will soon get
lost in the classes, I am a developer with 25 yrs experience, and believe me i have seen it all and done it all in a lot off languages (yes.. also OO).
Dont get me wrong here, OO is really a good concept !, but it can be hard to get.
That brings me to the reason WHY i am using Cotonti Genoa: I can follow every bit in the code without any knowledge (That's AWESOME).

Back to the developers part: If you want new ppl to start working on the core or plugins, make it easier.
Second thing: If the lives of the core devs change (kids, mariage, demanding job whatever) then impact will be great on Cotonti, so thats why
Cotonti needs new devs !!!

One example in the new features:
The database layer... i think i know the reason why devs put that in... to be more independant from the database we use.
So we can put in MS-SQL or something.  (At least thats what i can make up from reading a little bit about it).
I think this is strange, considering 99,5 % will use MySql anyway. WHY o WHY waste time on 0.5%... ?? (just to have a cool abstract layer.. ?? )

I think that Cotonti is better of if we have much more nice plugins / features, and good tutorials for new users.
And I truly believe that Cotonti is helped much more if the core changes are done much more gradual... and not those BIG bangs you are doing now.

Well big story.. but I truly believe this.. 

Conclusion:
Keep it simple and no BIG bangs.. and more front end things (cool plugs.. like good FB stuff).
And improvement of the core, can be done slowly.

It seems I started a discussion here, and these are just my thoughts... but I am asking everybody to participate in this discussion.

Ez,

It's understandable to keep things simple, and gradually improve the core; however, the Siena branch has been around for quite some time now providing ample time to update plugins and themes. Most of the plugins people are requesting are included in the default package of the Siena branch (or achievable quite easily), and even then.. developing plugins for Siena is not that different from Geona.

To try and keep Cotonti from moving to object oriented code is silly; It is the future, and the next best thing. I don't seem to understand why it will shy away php programmers from Cotonti. Realistically any new comer to Cotonti is going to have to invest a good amount time learning how it is structured. The first time I looked at Wordpress.. I was completely lost, but any good programmer will pick it up in time.

The only underlying problem I am finding with the Siena branch is the lack of theme developers. Which it can be hard to find willing, and skilled, template designers/coders. I personally find both Siena themes extremely unattractive, and even more so the Geona themes.

AAAAAAHHH

I've typed a very large post.. press Reply

you are not allowed to do that.. redirecting

http://s3.amazonaws.com/kym-assets/photos/images/original/000/000/578/1234931504682.jpg

I sometimes forget to do that after a large post.. wasnt the first time, probably isnt the last one too.. going back didn't work. ;(

#32015 urlkiller: 

what i LIKE would be the possibility to customize the siena system as you wish. some kind of "system-modules" that would allow me to add or delete internal parts as i wish in no time. i want forum -> i add forums module in /system/, i dont need that -> i delete the folder in /system/

so for a small site i would just have a system with 100KB and insane speeds because it only loads exactly the stuff i need/want.

Well, this is exactly what Siena's "modules" stands for. You can completely remove "forums" or "polls" for example just like you do with plugins. Or write your own module. This was called "Modularization" and it is a good example of the price that we had to pay for a new major feature, because in order to achieve it we had to do the following:

• Change directory structure and unify it.
• Modify the internals of site configuration system and build a configuration API.
• Modify the internals of site authorization system and build an authorization API.
• Unify the way plugins and modules are handled by the system and build an extensions API.
• Develop universal methods for installation/uninstallation and updating of modules and plugins.
• Move a lot of stuff from the core to modules which actually use it.
• Introduce such thing as dependencies between modules and plugins.
• Add cot_incfile() function which simplifies inclusion of one extension within another.

I'm sure I have forgotten some other things we had to do to achieve what we have now: you can assemble your site of modules and plugins like of lego bricks. But as a result it is not backwards compatible, it's a big bang.

There are some unpopular decisions we've made without being enforced to. ez mentioned one of them, CotDB / PDO. A funny thing is that it was actually aimed at developers new to Cotonti, because in the modern world of PHP5 adoption, PDO library is very common and is used by lots of programmers around the world. It is also the basement of such popular libraries as Doctrine and phpActiveRecord. So it is likely to be more familiar for a newcoming PHP programmer rather than sed_sql_* functions.

I think the most unreasonable change technically speaking is the global change of prefixes from sed_ to cot_. We could easily live without that. But it's a kind of marketing thing, it's about project recognition. If you are new to Cotonti and start looking at the code, one of the first questions which appear is "This system is called Cotonti, why does everything start with sed_?".

Now back to Genoa and new features. Cache is another good example. In Siena we are developing a more advanced cache subsystem. It consists of several layers of caching, from caching particular variables to caching entire pages. All sounds very good in theory, why not apply it to Genoa fresh out of the oven? Well, it would break it then. It took a few releases in development branch to get it stable enough. Currently it is still several steps from being complete.

For about 1.5 years we actually used Siena as a laboratory for new features which then were backported to Genoa. It worked until these branches became so different that I got tired of backporting and also that topic of "stop making changes in Genoa" appeared. If there were people dedicated to Genoa branch, the process of getting new features in Genoa would be easier.