Jump to content

Prestahop fork


yodapt

Recommended Posts

I've been wondering this as well. Why did you want to fork PrestaShop? What features are you hoping to add that you don't think PrestaShop will add themselves? I'd like to see a PrestaShop version with support for HTML5 and the latest web technologies.

I can't see why you couldn't fork PrestaShop, but of course it would have to remain free and open source to comply with the license.

Link to comment
Share on other sites

Prestashop works, generally speaking, but it has many problems.

Regarding clients, it's admin panel isn't very KISS-style.
Too much information badly displayed, bad module system interaction and some basic secondary features like CMS are hard to find. Lets face it, the client isn't supposed to buy a GPS just to know were to change something on his e-commerce site.

About the code, I really hate it. The coding practices are old, bad and don't provide a quick way to interact and understand the system, and that's why people ( devs ) are leaving to other platforms like OpenCart ( I don't like this one either, but at least the code is much better ).
Nowadays, it's expected that a web application have some conventions and style that can be as intuitive and organized as possible, as well as a good filesystem and use of an architectural pattern like the widely used MVC.

There's many improvements that can be made to the actual software. To give you an example, there's plenty of native functions in Prestashop that are rarelly used, or optional and never used, that could become addons / modules that would be only activated if there would be a planned use for them. This change alone would cause major impact on the learning curve of a new client.
Like this one, there are many other issues that could be solved if the whole software were rebuild.

I don't know if you guys are happy what Prestashop currently offers, and the way it works, but I'm not .. But I understand the efforts that Prestashop Dev Team putted in this software from 2005 until now, and that's why I feel something for the project, the idea.

Link to comment
Share on other sites

I have been think along the same lines myself, but a very basic fork for the UK market, so that all date, times and address format is displayed correctly for the UK market. and maybe an update on the jquery side as 1.26 is really getting quite old now

Link to comment
Share on other sites

@bartman,

I thought of that as well, but for Portugal. The system should be easy to change, and not customize. It's nice for the users with no knowledge of any kind related to programming that Prestashop can be customized, but the "price" of that customization is a system with too much information that is rarelly used, intrusive and bad for the learning curve.

Link to comment
Share on other sites

yodapt

I tend to agree, but their are a lot of people out they who have no wish to learn a system, all they want is a plug and play, as soon as they have to start editing files to acheive the necessary result they will lose interest and decide the system/platform is not for them, I suspose this where a freelance developer comes in to the equation,

Even looking at the way the date system is embedded into prestashop their are numerous places where changes will have to be made. In an ideal world it should be a configuration option in the BO

Link to comment
Share on other sites

@bartman,

I agree. Unfortunelly, imo, anyone nowadays tries to set their websites on their own, wich I don't see any problem, except that it would require a real good system, user-proof if you know what I mean. Prestashop doesn't seems that kind of system.

I'm fan of opensource, and consider it not a cheap and easy way of getting what you want, but somehing that must be taken seriously. Times have changed, and I like the slohan "best things in life are free".

Note that I'm talking about a fork, not a major change in Prestashop. I believe people deserve different sollutions for their problems, and a Prestashop fork directed to those clients who need something customized for them and have no interest in plug-and-play at their risk, leaving all the work for the "experts", would be a bless.

From my point of view, a system with developers as their primary target should only provide basic functionality, a good documentation and lots of modules only available if the final client really needed them, not embed like Prestahop actually does.

There's a couple of things that Prestashop missed from the start untill now :

- A real OOP system (OOP increases the learning curve of any developer and helps keeping the code organized);
- A real template engine (Smarty isn't a template engine, imo, because it involves logic in the templates and therefore those templates are useless outside Smarty environment);
- Code abstraction (mixing markup with server-side languages is a bad idea, and hard to mantain);
- Simplicity (for those plug-and-play clients, the current system is awfull ... KISS is the way to go, always);


There's a lot more than this, but those I refer are the major ones I can think of now.

I'd like to fork Prestashop into a opensource product oriented to developers, and that could also be abstracted enought to be merged into other working systems. E-commerce should be thought as a module itself, not a system like Prestashop and all the other "rival" systems do.

Kohana Framework seems pretty neat for this, as it currently is the only one capable of making real use of OOP and real templating.

If anyone's interested on doing this with me, PM or answer here :)

Link to comment
Share on other sites

  • 3 months later...

I'd be very interested. I don't know how much time I'd be able to put into it but certainly interested.

I just ran into a problem that I posted about consistency in handling of objects in PS. I have run into that so many times as a developer that I cannot count.

I, too, would like to see it done MVC. I've seen one open source Code Igniter shopping cart and it looked pretty good. A lot of people claim that PS is OOP but I beg to differ. Any OOP enthusiast will laugh to think that PS is anywhere near OOP. Just because they have a few classes doesn't make this OOP, see my post on consistency and you'll see why.

A good templating engine such as the one that Drupal uses is an absolute must. Trying to build a template for this thing takes a genius. Whoever thought of this "templating" engine needs to be taken out back and shot, or, since they are French, to the guillotine. The CSS is rather stupid, I had to create a new selector because I wanted the tree lines on the sitemap tree but not on the category block tree. Shouldn't have had to do that.

Oh, and don't even get me started on Smarty. It is rather stupid to have to learn a totally different "language" to use it I'd rather just make my templates PHP and use one language.

The DB architecture looks like something a "script kiddie" built in the basement of his mother's house. Any professional DB developer would laugh till he peed his pants at the "relational" design of this database. Where's the referential integrity? Where the transactions? Where are the foreign keys?

Need I go on?

It is sad that the developers are
1. Not willing to communicate with people like you and me to determine how to remedy some of the shortcomings of PS and to make it more competitive in the market.
2. Either to stupid or too arrogant to fix the problems.
3. Don't take the time to write documentation for the developers, this is area where Magneto and OsCommerce beat the hell out of PS.
4. Shy away from helping in the forums. Who better to answer questions about the product than the developers, themselves? I don't know, maybe there are a lot of them in the French forums, I just know that there are none answering questions in the English forum, oh, except Patric, at least he makes a small dent, albeit very small.

Link to comment
Share on other sites

  • 9 months later...

I've been thinking about a fork of PS recently mainly because of code style, developer reluctance to communicate and general PS development trends (i.e. "yes, we've got ugly code, ancient modules (fPDF, etc..) and unsolved issues, but instead of fixing it all we'll just go on and add more features and release new versions).

I believe that a fork could get popular enough soon and attract mature developers who aren't satisfied with current PS development practices. The only problem here could be the OSL 3.0 license which is used in PS currently. I heard it's kinda restrictive when it comes to forking, altering and distributing source code. We'd need someone keen on licenses to clear things up.

Meanwhile I'm wondering how are things going here. Looks like nothing has changed much since the last reply.

Link to comment
Share on other sites

  • 3 months later...

let me say something:

I've tryed several open source ecommerce platforms, in my own opinion the basic feature of a ecommerce platform is only one: PERFORMANCE. Google also changed it's page ranking rules to give more weight to performance, and for a ecommerce platform there are no compromissions on this.

it seems that only prestashop's team understood importance of performace in ecommerce, I've tryed Virtuemart, Magento, Opencart,and no one of them has eye on performance. the only one that has is Oscommerce, but it's too dated and it's developement is really slow now.

 

What I want to say?Instead to fork Prestashop work on it helping developement team to improve it and reach a better code, the way of prestashop, focused on PERFORMANCE, I think is the right way.

 

 

 

 

Prestashop works, generally speaking, but it has many problems.

Regarding clients, it's admin panel isn't very KISS-style.

Too much information badly displayed, bad module system interaction and some basic secondary features like CMS are hard to find. Lets face it, the client isn't supposed to buy a GPS just to know were to change something on his e-commerce site.

About the code, I really hate it. The coding practices are old, bad and don't provide a quick way to interact and understand the system, and that's why people ( devs ) are leaving to other platforms like OpenCart ( I don't like this one either, but at least the code is much better ).

Nowadays, it's expected that a web application have some conventions and style that can be as intuitive and organized as possible, as well as a good filesystem and use of an architectural pattern like the widely used MVC.

There's many improvements that can be made to the actual software. To give you an example, there's plenty of native functions in Prestashop that are rarelly used, or optional and never used, that could become addons / modules that would be only activated if there would be a planned use for them. This change alone would cause major impact on the learning curve of a new client.

Like this one, there are many other issues that could be solved if the whole software were rebuild.

I don't know if you guys are happy what Prestashop currently offers, and the way it works, but I'm not .. But I understand the efforts that Prestashop Dev Team putted in this software from 2005 until now, and that's why I feel something for the project, the idea.

Link to comment
Share on other sites

Instead to fork Prestashop work on it helping developement team to improve it and reach a better code, the way of prestashop, focused on PERFORMANCE, I think is the right way.

 

You are right: duplicating the code is a bad idea.

 

I think there is people willing to work on the project, but remember PrestaShop is not so opened: to submit a change, you need to submit it to PrestaShop's team, explaining what you did. This is a huge problem, and it may discourage the most motivated developers.

 

And I think PrestaShop's performance could be improved a lot! If you look at the code, there are LOTS of calls to file_exists and reflection is used to instantiate even the most basic class. Remove this and it will be even faster :rolleyes:

 

You are right about performance, but remember that a clean code can always be optimized, while an optimized code is hard to change.

Link to comment
Share on other sites

What I want to say?Instead to fork Prestashop work on it helping developement team to improve it and reach a better code, the way of prestashop, focused on PERFORMANCE, I think is the right way.

 

I disagree, the current dev team does a awfull job, and is very strict about "helpers". I don't have time to fix what should have been fixed years ago and risk my work to be denied for reasons beyond my comprehension.

 

Achieving performance in a system like this is perfectly possible, as long as the product is made with an extra, let's say, "commercial target" : developers.

 

They are the real ones capable of doing wonders to the clients, bringing respect and publicity to the platform, as long as the platform is coherent enough to allow them to spend their time doing their job instead of trying to understand messy code and kill the meaning of the word "WTF".

 

I have so much more to point ... and so little help on this :(

 

One advice for those who feel that there should be improvements to Prestashop : Do it yourself, focus on what you're trying to accomplish and don't take any related subject lightly (otherwise PS would be a good piece of software).

Link to comment
Share on other sites

Yodapt, the reason for this is very trivial : Prestashop is deployed on hundreds of servers, and there is people making money with it, and often they are not developers at all.

 

You can't ask those people to make changes that may be dangerous, and tell them it's to make developers life easier : they don't care, all they want is a software that works.

 

I think Prestashop's developers would like to start the project from scratch, but it's kinda impossible.

 

Software is used by people, and also by developers. Software that is only used by developers is just useless.

Link to comment
Share on other sites

As xmurph said, PrestaShop is more than willing to integrate the community's ideas and improvements.

 

You have many ways to help us.

 

For example, you can use the forum and create a topic about your ideas/suggestions.

 

You can also post in the bug tracker and create a new request.

 

And if you have any other idea on ways to help us do not hesitate to tell us.

 

As I said, every help is welcome!

Link to comment
Share on other sites

Salesman talk. I'm a web developer, don't give me that. You know there's a lot to do about the system and you just don't want to compromise your financial situation. People complain for years about many things rather than the system itself and at how cost did they had what they needed?

Link to comment
Share on other sites

Hi yodapt,

 

Of course we now there is a lot of room for improvement and we are working on it.

 

For the next version we have taken into account a lot of the complaints the users may have had in the past. We also integrated codes that developers sent us.

 

Have you checked the latest milestone for 1.5?

 

Could you give us examples of things that people complained about?

 

Also, I do not understand the “you just don’t want to compromise your financial situation”.

 

Could you explain in details this one?

Link to comment
Share on other sites

Yodapt, the reason for this is very trivial : Prestashop is deployed on hundreds of servers, and there is people making money with it, and often they are not developers at all.

 

You can't ask those people to make changes that may be dangerous, and tell them it's to make developers life easier : they don't care, all they want is a software that works.

 

I think Prestashop's developers would like to start the project from scratch, but it's kinda impossible.

 

Software is used by people, and also by developers. Software that is only used by developers is just useless.

 

 

You missunderstood my point. I'm not talking about a dev oriented system, I'm talking about creating a base that can be easilly extended, giving developers a solid and easy way to build up their client's / personal systems. This alone will give room to way better modules / plugins, without the (sorry, but it is) "stupid" limitations that PS currently offers as a "plug-n-play", wich sounds more like a "plug-n-deal-with-it".

 

My point is simple : don't let the "plug-n-play" utopy overcome the need for a stable and extendable product. People need more than things that work.

Link to comment
Share on other sites

Hi Yodapt,

 

My point is simple : don't let the "plug-n-play" utopy overcome the need for a stable and extendable product. People need more than things that work.

 

Could you give me example of what you meant ?

What are your propositions to improve this ?

Did you ever submit an improvement or an update to the dev team ?

 

You said that we don't want to accept community help but that's not true.

For example, the USPS module was updated after someone of the community submit major improvments (http://forge.prestas...owse/PSCFI-3189), the OneClickUpgrade was made partially by a member of the team and a member of the community (http://www.prestasho..._click_upgrade/), ...

 

Yes, you have to submit your change to the dev team if you want the modifications to be integrated. You want us to accept every code even if there are bug or security flaw ? Please, submit ideas of improvment like you made in http://www.prestasho...-remove-smarty/ (even if it was a bit late because someone else alreay made the proposition a few posts above) and if the idea looks ok, if several members of community wants to have it (not only you), we will see how we can make it. But once again, if you want to participate and code something, I think it will be more constructive to submit your proposition of modification to the team instead of trying to make a fork.

 

Ps: I'm also interesting to know what you meant here :

you just don't want to compromise your financial situation
Link to comment
Share on other sites

Fabien, Carl,

 

I think yodapt is talking about the fact that you are a commercial open source company (like there are many today), you earn money only with support and paid services, and you can't do what you want, because you have to satisfy both developers and businessmen.

 

Personnally, I have no problem with that. This allows bringing better software.

 

The only suggestion that I could make is to open a little more PrestaShop.

 

 

You want us to accept every code even if there are bug or security flaw ?

 

Certainly not. This requires a good organization, but for example if you host your project under Git, contributors could branch the project, submit pushes, and then it's up to you to take what you want, modify it, and intergrate it on the stable branch.

 

I have always wondered why this is not like that :(

 

Why ? Is it because you fear that it will be unmanageable ?

Link to comment
Share on other sites

Hi Mexique1,

 

I can't agree more with what you said.

I think it's most because we had not enough time and we already have a lot of work. :/

 

Still, it is a good idea, we are already thinking about opening unit test to community, but that kind of project takes time. I don't know how this idea will evolve but I promise we will talk about it.

Do not hesitate to send me a private message if you do not get an answer from me by the end of next week :)

 

Edit: But again, we are already open to comunity submit as you see for USPS or One Click Upgrade, so if you already have something to submit to use, do not hesitate. For the moment, we didn't get a lot of community submition about the core (I'm not sure, I may be wrong), beside the themes, modules and bug reports of course :)

Link to comment
Share on other sites

Thanks again Fabien :)

 

Actually I submitted only 2 fixes :

- http://forge.prestashop.com/browse/PSCFI-3439

- http://forge.prestashop.com/browse/PSCFI-3425

 

But I'm relatively new to PrestaShop. But believe me, I will submit as many as I can.

 

The thing is if the change is huge and involves several files, it's a nightmare for us to send the files, and for you to understant. With a good SCM it should be fairly easy.

 

But we have said that before ;)

Link to comment
Share on other sites

  • 1 month later...

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...