What do YOU expect from the next major phorum release?

Fix outstanding bugs, like [github.com] :-)

I tried to come up with a vision of Phorum for the future but I just can't. IMHO it will never be more than that; but it won't be less because once you get into it, it's OK as it is.

Yes, there are issues when working in more organized development environments; like the generous usage of globals and stuff like that. Makes software testing and isolating things quite ... unexpected (but it's doable).

The consensus in this thread seems to be that we should just keep the current phorum and fix something here and add something there.

Just to make my personal opinion clear:
There will be something new, the question is with which content.
We are currently just 2 people left in the core team coding on Phorum when we got time and motivation to do so while in the same time we have moved on and got a family to care for.
Opening Phorum on Github brought just one pull request and no further commitment by other coders.
Also, in our opinion, the current codebase is tough to handle with future enhancements up to a state where its nearly unmanageable.

Therefore we'd like to get some fun again into coding phorum and make it manageable at the same time - fun with working on some better concepts like MVC and using (finally mature) stuff like exceptions and OOP in PHP, making it manageable by working with a bit of TDD (maybe not completely but with far higher test coverage) and pushing the module concept to its limit by having a really lean core.
This will also allow a lot of features requested over time.

---

Thomas Seifert

RTL Theme support full.
will be great!

Priority number one: Mobil phorum

Everything is growing around cellulars. So, the next obvious step is to have a phorum that is a native mobile platform. Probably the same functionality on his core, but easy to adapt to mobiles.

Cellulars and tables are replacing laptops and PCs. In order to have an exponential grow, we need to reach all that audience in an easy way. Phorum is the tool. What changes are needed?. Layouts, layouts, layouts.

In addition, a couple of tools to take advantage of cellulars: a map functionality, live chat, and more visual tools such a module integrated with the cullular camera.

Finally, on this days everything is around "the cloud". So, why not to include a couple of simple tools to offer phorum users the possibility to experience the cloud from a phorum platform?.

Thanks in advance for all your efforts. I´m sure there is no limits on the future of phorum.

In order of highest importance first:

1) To GET RID of those annoying confirmation screens, "your message has been posted", "you have been logged in", "thread has been deleted" etc.... Man that crap's from 1993. Back in the day when we had dial up and weren't sure if submitted commands went through. It's time to get rid of it.
2) DO NOT redirect a mod back to the list of threads after he/she delete a post inside a thread. I feel that's very improper.
3) Ability to allow users to post in a forum but not read or view anything in it. This would be perfect for all sorts of websites and communities, allowing everything from "contests" to an "anonymous tips" or "feedback" forums to be easily made.

=== Wild Wish List ===
3) to be able to sort a list of threads,
4) for an unregistered user to be able to report inappropriate posts

What I would like to see is a Meta Data Driven, N-Tier Design
* N-tier and MVC are not mutually exclusive, nor are they the same.


UI / Presentation

Access Centric
..Security / Business Rules

Data Centric / Access / Permissions / Data Integrity / Data Validation / Create, Modify Table(s)
..Data translation classes (for different storage medium)

Data Store (whatever flavor/medium)


--------------

What follows are my thoughts, not necessarily complete, or in the correct category/tier.



UI, User Controlled (things not in control of developers)
..JS not absolutely required
..Cookies not absolutely required
....Public
....Mobile access

..Impaired User Friendly
....Screen Readers
....Screen Magnifiers

..Translations (some browsers support directly, some don't)



UI, Presentation
..Provide options that enhance usability.
....Phorum
......Templates
..........Internationalization / Localization
........Social Media
........Translations
........Impaired User Friendly Considerations
........A way to track and flag edited messages and attachments (different than "New")
....Web Site Integration
......Other content management ;)



Business Rules / Security / Permissions
..Modular
....Connection Types Spported (HTTPS, HTTP, Console)

....Early hack / spam detect
......Hook at beginning of common.php (or equivalent). For running third party script, without a hack. < ?php $path/script ? >< ?php "common code" ...

....Force better and safer passwords
....bcrypt for passwords
....Anti Spam
....Event Logging

......Public User
........Registered User

........Hooks
..........User Controllable Translation For Any Template (Google, Bing, Others)


....Moderator Features
......User Approval (Needs to be combined with other modules listed to give moderator all of the usefull information for a decision.)
........Block User
........User List
........StopForumSpam
........Related IP Events from event logging.
..........Sorting + Filtering

......Moving messages, leave a placeholder for moved message that allows for removing message fron middle of tree, without requiring moving child replies (doesn't need to display).


....Admin Features
......Prevent deletion of messages by selecting an Admin general settings checkbox, message is hidden, not deleted.

......Debug Specific Problems
........Ability to single step through messages in a topic without loading the whole topic/thread into an array which triggers the error.
//(For example I had a message in a topic with an unserialize problem that took many hours to locate because the actual message with the problem was not individually identified. The error turned out to be a corrupted link to an attachment.)


....Developer Super Access
......Program Flow (order in which modules are used)
......Variables available for different access permissions.
......Test scripts
......Debug tools
......Hook list and where available
......Console (Brings up a list of programs that may be run "local", and allows running them from a console screen)



Data Centric / Data Access / Data Integrity / Data Validation / Create Table(s) / Modify Structure

..DB Search
....DB Full text search using the best available option for the current data store, and Regular Slow Search available from the search interface at all times.

....Choice on number of months to search. (Currently 30 days, 90 days, One Year, then All)

..Database Access
....mySql Class
....SqlServer Class
....ODBC Class
....PostgresSQl Class
....Other Storage Class

..Event Logging
....Improved Sorting + Filtering
......Ability to sort by count of ip decending among other things.
....Improved Logging Possibilities.


..Backup Database
....Full -exclude list for non critical tables
....Incremental (by date, which picks up edits, and changed attachments whether stored in tables or not)

..Orphan Message and Attachment Finder
....I still have orphan messages that don't get located, so there must be more than one way to have orphan messages.
....Moving messages may leave orphan attachments. (reported on phorum by at least one person, unproven?)

..Registration IP, Login IPs (so we can force a manual login when a new IP is used)
....Separate User Name screen from Password Screen (If UN doesn't exist, no password screen)


Data Store (Database) beyond what Phorum currently uses.
..Registration IP, Login IPs
..IPV6 storage

..Data Dictionary (Not the built in schema over which we have little control.)
....Data and meta-data (structure, indices, relations, variables, permissions, features)
....Program code

..Data Driven where it makes sense.
....I would put everything in tables (code, variables, permissions, docs, data dict.), even if it is not used for data driving. If performance may become an issue, the data dictionary may be used to create scripts files that don't require database access, and don't contain unused options.

..Internationalization / Localization



Documentation
..Program flow
....Variable list and where they originate.
..Document which tables and fields are critical, and which may be rebuilt from existing data, or ignored, even though they may cause minor inconvenience temporarily.
..User Docs


Testing
..Tests that should succeed .
..Tests that should fail.
..Debug code


Cache
..Structure gets too deep for client browser clearing.
....Create a cron job that runs on the server.

--------------
Modules (Other)

..Review all modules to determine if there is common code that may be reused. Move to class(s).



--------------

Further reading.

Two of the following sites belong to -- Tony Marston.
Besides a framework, he explains the "why" of many of his design decisions used in the framework. Well worth reading even if you don't like his way of doing things.

Data Dictionary
www .tonymarston.co.uk/php-mysql/data-dictionary.html

RADICORE -- Rapid Application Development Framework
www .radicore.org/


----


Jakob Nielsen's 457 essays on web site usability.

www .useit.com

Have you tried the Mobile template that is already available?
[www.phorum.org]

...
Steve H, currently on: (version 5.2.23)
contributions:
Birthdays mod, Top Users mod, Icon legend.tpl, (plus a handful of bugfixes and old 5.0 creations)

Login to phorum using Facebook's login system

I'd like an embeddable forum--so I can put the display pages and post editing displays inside a DIV inside my CMS. Admin pages can be their own thing. I don't want an Iframe. I don't understand how exactly but it seems this is sometimes done with complex Javascript.

Can't there be a simpler way? If all incoming post and get parameters (to phorum) had a well-recognized prefix (phorum_page=xxx) the enclosing CMS could filter all incoming posts and gets through a plugin filter. A plugin plugin. Posts and gets belonging to the enclosing CMS jump out of the plugin case statement and proceed as per the normal way. Writing a plugin filter for any given CMS would be a relatively minor project.

Posts and gets belonging to a plugin (phorum_page=xyz) would use a special *.tpl that includes no enclosing html head or body elements. And that would get stuffed into a buffer or variable first, before subsequent handling by the enclosing CMS.

Embeddable cloud-based forums are popping up like rabbits. This attests to a need. But those cloud based embeddable forums come with a monthly fee.

This is a follow-up to my last suggestion about an "embeddable" version of phorum. I have a buddy who works in E-commerce. He makes good money (big money) setting up interactive shopping carts, often for interactions between retailers and distributors, rather than for the retail public at large.

He reports to me that interactive "product reviews" on shopping cart sites are the number one selling tool. More important than any other single consideration. When people see others have tried product-X and like it. They are far far far more likely to click the "Add to Cart" button.

Phorum could not be used in this context because it insists on making its own page. Someday one of the open source forum projects will make their project "embed friendly," in a way that does not use an IFRAME. A few already make that claim. But that's where hot demand is right now.

Ability to embed Soundcloud and mixcloud must be worked into the new version. This is a feature of ALL paid bbcode forums and for a music forum such as mine those two now mainstream players must have the embed code function.

It doesn't function with the html embedding module enabled either so perhaps an Embed Souncloud and Mixcloud module should be built in complete with bbcode buttons.

Quote
Thomas Seifert
and pushing the module concept to its limit by having a really lean core.

I am playing with the Twig template engine for my own project and start liking it.
Other people may like other template engines:)

Maybe it is possible to use an own favorite template engine? I think it would be difficult,because i really do not know in what way to deliver data for more then 1 template engine.