It is not fair only to point out what is not good with upcoming WordPress, as I did yesterday. There are many things WordPress need, and for some reason all that is completely overlooked by core developers, so I will list all those things that will actually make WordPress better.
Before we go on, here is the reminder for the article published yesterday about upcoming WordPress:
What will we get with WordPress 3.4?
This list is based on the features requested by WordPress community, and things I consider important. This features are in circulation for over a year now, and still most of them are not even mentioned in the future development plans. Problem with most of these features is that they need much more time to do, and current quick development cycles core team insist on are simply not long enough to make these things happen.
- Replacing outdated and no longer maintained ThickBox with jQueryUI Dialog. This is actually something that was in works, and from what I have seen, it should be ready for the next WP 3.5. ThickBox popups are very complicated to control and they lack many features that jQueryUI Dialog popups can do. Replacing them would be great. But, that will also cause all sorts of problems for plugins developers, because most of the code using ThickBox will not be compatible with new popups. Colorbox is also a good way to go to make popups, but I would prefer jQueryUI approach, considering how much WP relies on jQueryUI already.
- Custom post types posts relationships. This is very, very important feature that will move WordPress in the field of full-fledged CMS systems. Right now, you can’t connect posts directly and with more and more popularity custom post types gain, this is a very important addition. There are several plugins for this, all implementing this differently, but this is something that must be in the WordPress core. This would need additional database table for bridging posts.
- Meta data for taxonomies terms. There was some talk about this, but it was always dismissed when plans for next WordPress were made. Right now all data types in WordPress have meta fields (users, posts, comments) with the exception of terms. It would be very beneficial to have those. This also would require another database table.
- Improved media library. This can be found on most WordPress wish lists on the Internet. I don’t have too many complaints to Media Library, but there are many things that can be improved, including attaching one image/file to more than one post. Other important improvements would include: user controlled folder structure for storing files (not only dates based as it is now), with improved URL rewriting rules that can hide wp-content from the URL. Also, real galleries implementation would be of great benefit for media library.
- Shared resources between sites in multisite. It would be of great benefit to have central media library with assets and images that all multisites can use (logos for instance). Only super admins get to control that, but any site admin can use those resources. Beyond files, sharing can be done for taxonomy terms, maybe even pages and posts to some extent. This can be useful if you have same Terms And Conditions page that any site in the network can use as its own.
- Improved posts management. Right now, posts lists on the admin side are real pain to use when you have large website with thousands of posts. Changing category for hundreds of posts at the time is impossible. We need better system that would allow simpler and more powerful filtering on that page and operations that go beyond current, flawed bulk operations (if you want to bulk edit 10 or 15 posts, you can wait good minute or two before you can do anything, and some browsers can freeze with bulk edit).
- New Taxonomies Terms management. This current is very bad, and if we get meta data for terms, this current interface must be replaced with something that is actually usable for website with large number of taxonomies and terms.
- Real search. Maybe the worst thing in WordPress is absolutely useless search feature. There is no need to explain how bad it is, and how much we need something useful to search posts. We need full text indexing, some search operators, easy set filters for post type, custom fields or taxonomies terms. There are some plugins for this, but this is a feature that should be in the core.
- Remove Akismet from core distribution. Akismet is a commercial plugin, and for most websites, it must be purchased to use. Bundling it with WordPress core is not fair to all the other commercial plugins developers. Considering that many users prefer solutions that work on site only with no remote servers involvement, there are better solutions that Akismet, and also free solutions.
And what we should not see in future WordPress version (for a while at least):
- We don’t really need new core themes. TwentyTen is great, and TwentyEleven is awful. I would like to see TwentyTen as HTML5 theme and be done with it. From what I have seen of TwentyTwelve, I am not impressed.
- No more cosmetic admin side changes. We need real changes to posts lists panel, taxonomies terms pages, leave the rest of WordPress admin as it is. It looks great as it is now.
With few teams working only on these features, I think (based on earlier development cycles) that these would take 9 to 10 months (maybe a year) to implement and test properly. Also, having longer beta testing period for it would be beneficial to convince larger group of users to switch to new version when it gets released. And I am sure that powerful set of features would be the best incentive for adopting new WordPress.
What do you think about these features? Would you prefer to wait longer for a WordPress that comes with most of these things, than to waste time on small updates as we have now? Let me know with a comment.
55 thoughts on “…And what WordPress actually needs?”
I agree about removing Akismet, even if you just have a few affiliate links to pay for your domain name and hosting that makes you a commercial customer.
What spam tool would you recommend? I’m currently trialing AVH First Defense Against Spam.
For the media library, I’d like to see some kind of categorisation or tagging of images, that would then allow people to drive gallery like functionality via plugins without the need for adding lots of extra tables.
I’m also not keen on short update cycles, every 9 months should be plenty apart from security fixes.
Thanks for the comment Andy. For spam, I use AntiSpam Bee and my GD Press Tools Pro that has a lot of tools for security and fighting spam. So far I am really satisfied with these two. This is something that should remain outside of the core and with plugins. And yes, all current Gallery plugins are too complicated and we need some simple categorization in the core.
Lets hope that someone in the Automattic and WP Core development team decides to listen to users on the things WP needs in the future, and we might see some positive changes down the road.
I mostly agree with your points! My list goes like this:
– Real Search!!!
– Media Library: especially gallery management so everyone could say goodbye to the gallery plugins…
– Removing Akismet and Hello Dolly crap! (the first two things I delete everywhere!)
New theme “TwentyTwelve”: it’s a shame that it’s not ready yet! I’ll recommend to do just child themes for TwentyTen or TwentyEleven – why does WP core team promote the child theme concept not more? I guess I’ll never use TwentyTwelve… (Btw, I really like TwentyEleven over TwentyTen but that’s just my taste 🙂 )
3.4 is a “strange” release as the theme customizer is useless for me – don’t need such stuff… But the new theme API might be the best thing in this release I’ll guess it brings good things in the future.
Performance-wise I couldn’t see any more speed yet, also translations loadings seems not to have any effect one can see and feel…
I’ve had no problem to wait until christmas for a BIG release addressing most of the stuff mentioned above. I am sure, millions of users would update in no time with such a release!!!
Thanks for the comment Dave! It is a great idea to promote more child themes from the core, it would be better than new theme.
As for performance in 3.4, both Beta 1 and Beta 2 are exactly the same, so my preliminary report in WP 3.4 post stands, new WP 3.4 is right now slower than 3.3.
Just FYI, the latest Types versions have some Post Relationship mechanism – http://wordpress.org/extend/plugins/types/ – it’s 1:M for now, but looks much better than having no options for this at all. However the lack of any metadata for taxonomies (like image and other custom fields) is something that I can’t understand at all.
Search (front end and back end) and the media library are the two that are on the top of my list, closely followed by post type relationships.
I don’t think that I entirely agree with you about the default theme. I love that there is a new default theme released every couple of versions (even if I really don’t like Twenty Eleven), but it would be great, as Dave said, if the core team focused more on building new child themes for the default theme, rather than building a whole new one each time.
What is to be added in a media library, except for the multiple attachments? I can’t think of many options that could be extra added, not to mention some basic image manipulation functions that are already in the media edit section.
For the search part, the best I’ve found so far is http://wordpress.org/extend/plugins/relevanssi/ . It needs some manual indexing, but at least it’s heard of the term ‘custom field’
Thanks for the comments Mario. Media Library structure must be flexible, allowing user set folder for images. And anything dealing with data must be in the core. Plugins for relationships and indexing exist, but that needs to be in the core because other plugins need to rely on that. I understand that WP core developers find it easier to wok on some visual elements that take less time, but there are many big features WP core needs and that plugin authors and users want to see in the core.
One very logical feature missing in core is auto delete of attachments. Since at the moment you can’t attach an image to 2 posts at once, they are still kept in the Media Library even if you delete the post. I just came up with a small plugin on WPORG last week that does it by default but I see no reasonable explanation on this.
Yeah, that too. I have made this for my GD bbPress Attachments plugin, so that when reply or topic are deleted, attachments are deleted too. I added options to control it, but this is something that can be part of the core.
Also it needs a real developing environment support!
I’m kinda new to WordPress and I am in the process of implementing it for my company. (I’m hired to work with SharePoint) and I think it’s hard to have the normal web development process supported in WP.
Ex: I want to have three environments, a development, a test and a production. All development is done in dev (of course) and then tested in test and if it passes the criteria it will be passed on to production sites.
So far so good, but in all projects there is first a launch where I’d like to ‘copy’ the site with theme, settings, contents —- but to a different URL and database!
I tried to manually do all configurations to my test.mydomain from looking at dev.mydomain since the ‘export’ only grabbed posts and stuff. It took forever and I didn’t get it right, at all. (Using Genesis too).
I honestly don’t know what to do, since a backup points to the test database.
Thanks for the comment Pia. This is a real problem, but not necessarily problem with WP core. We can use some solid solution that can be used to easily deploy code between dev, test and live environment, and that could use some help from core developers on building such system.
NetBeans is a great as environment to develop for WordPress and a plugin for it to do all that would be great addition.
Download this zip, extract it to the root of your WP folder: http://www.interconnectit.com/wp-content/uploads/2011/05/searchreplacedb21.zip
it’s from these guys: http://interconnectit.com/124/search-and-replace-for-wordpress-databases/
How to migrate:
Upload all your files to the production environment;
Upload the searchreplacedb.php to the root dir;
Export your database and import in to production;
Access http : // yoursite.com/searchreplacedb.php -> fill the inputs
Then access your http : // yoursite.com/wp-admin (log in)
Go to Options -> Permalinks and click save;
This is the way I do as well, but, honestly is not soo “automatic”.
And there is no easy way to sync content dev-stag-prod on a heavily used production server.
But this may really not be a “core” issue / problem / requisite .
That you both very very VERY much!
Totally agree with the post. In addition to Media Library, I’d like a better API for image manipulation, especially more options for resizing.
I don’t know why WP comes with both TwentyTen & TwentyEleven. They confuse me. I think, WP team should make ONE great default theme, instead of making SEVERAL good theme. More than that, people always look at the default theme as the exemplary theme, it should be solid, include new features, have good code practices, etc.
Right on the money. One theme with all the latest things in WP (even some example templates for custom post types and taxonomies, many users have trouble with those), and if needed, give few more child themes from it.
Bundling two themes is not good solution, bundle TwentyEleven (or latest one), and older WP default themes can be downloaded from WP.org through themes installer.
My #1 request would be a user profile system. We’ve had to customize so much of our WordPress install in order to make it even semi-intuitive as a user-based CMS. It would be nice to easily add additional profile fields, themed login and profile edit, etc.
Just my $0.02.
I agree with many of the things on this list, although I’m more inclined to give the core devs a pass in discussions like this. The reason we want all this stuff, after all, is because WordPress is so good already.
Still, my big thing as a developer is post-to-post relationships. After custom post types and custom taxonomies allowed us to really treat WP like a CMS, direct relationships is the next logical step.
Honestly I think the “Posts 2 Posts” plugin is at a point where it could be incorporated into core with minimal changes. It’s developed by a core contributor (scribu) and has been thoroughly tested in the wild.
Second on the wish list: revamped media uploader. From a UX perspective, the thickbox is cumbersome and stuff like setting featured images have way more steps than they should. But they seem to be addressing this incrementally (recently removing the unnecessary media buttons, adding the drag-and-drop uploader, etc).
Thanks for the comment. Yeah, if there is a plugin good enough and mature enough to put into the core to solve post relationship, it can be easily adjusted and added without too much time spent on it.
We definitely need custom post relationships and term meta data. Those two would be a definite plus. I don’t understand why the WP developers are dragging their feet for these features; they are the logical progression for custom post types and would help to create more advanced, data-driven websites and blogs.
Apart from the obvious Search deficiencies, this is the one feature that I feel is truly missing for just your average bloggers.
I hate having to manually link up posts together, and I never manage to get the formatting the same across all the related posts.
The WP CMS runs in a database for crying out loud, so the ability to create related posts only seems natural.
While these changes may be needed, they are not nearly as important as a change I think is needed. WordPress needs to change their entire architecture to support being able to easily handle content delivery to multiple devices ranging from desktop computers to small smart phones.
I’d agree with New Taxonomies Terms management.
1) Ability to sort categories (menu order, etc)
2) Featured Image for Categories
I’ve had to create my own but it’s still not as good as it could be.
What about a real shopping cart that works and does what a real cart does. I like all of the other ideas and do agree. But a shopping cart would be a great addition and really make WP great.
The jigoshop and Shopp plugins have this covered for me.
There’s no way the core devs would tackle a shopping cart. That’s plugin territory all the way. Remember, a core WP install doesn’t even have a contact form.
Yeah, definitely no way WP would ever, ever include a shopping cart system, or anything of the sort.
You have to be careful when thinking about what WP core should have, vs. what you would love to have a plugin for. Shopping carts and contact forms are perfect for plugins because *some* people want / need them, but the vast majority don’t.
I’m a little late to the WP game, only having been using it for a little more than 2 years. But where is the WP Bible that says that certain features should be left to 3rd party plugin developers?
A contact form or shopping cart seems like a pretty fundamental feature. If they want to add it, I don’t think many people would be upset – except the 3rd party developers who already built plugins.
Matt explained that many times already. There are a gazillion APIs, hooks etc. for this, they want to keep core as lean and minimal as possible .
It’s like a smartphone with only the essential apps, rest is in Market ..
It’s the most sensible approach.
Don’t get me started on smartphones, there’s a whole bunch of stuff on mine I’d like to uninstall but can’t.
Here’s a link to the philisophy page that explains that 80% principle
I agree that core should be slim, but there are features that must be in the core, features that are essential to the CMS: posts relationships, terms meta data and many other things. And Matt’s explanation fails on the new WP 3.4 that is adding bunch of useless stuff to handle themes settings when 99% of themes do that already. WP core doesn’t need that, but it does need things that deal with data and to allow better data management.
My reaction was mostly geared towards ppl asking for built-in contact forms, gallerie thingies, e-commerce shops.
I agree, basic functionality like posts2posts linking should be core indeed (and won’t take up that much space too imho)
Theme settings etc.. I hate practically all “custom” solutions with their own fancy UI’s and own logic.. A good base standard I’d surely welcome, but it’s too little now to make real good use of it.
I am not for adding complex features like forms or shops and things like that into the core: most people will not use that, and having multiple options for these to choose is a good things. As for the theme customizations, it is too late to change things, and spending months of development on that was not good use of the core developers time.
Great list! Pretty much have every single one of these on my own wishlist. I would also echo Pia’s earlier comment about a solid content staging environment being built into core. WPEngine have already recognised the latest demand for this and rolled their own solution. If I recall Alex King also built a deployment platform for this last year aswell – but for me this should be in core.
So true… agree to sign under every item…
Actually, I really don’t understand what’s the benefit of this so-called “short development cycle” if everything we get is only some cosmetic changes… all these adminbars and screen options… when every item from your list is REAL improvement…
What’s wrong with the only one new version in the year but with real improvements?
When site on WP should go a bit further than personal blog every developer, I’m sure, needs to find one or another solution to deal with nearly each of mentioned items… of course there are number of plugins for that… but incorporating such things into core provides a standard, the solid foundation for others developers to do things
Thanks for the comment Anna. Some features must be in the core, so that developers like myself can make better plugins and don’t worry about the WP core. And with this short development cycles only cosmetic changes can be added, and that is something on WordPress.com can use for its own purposes. WP 3.4 will bring a single important change that in all fairness is not something most developers will need (theme API).
Great post! I don’t want to sound like “me too” kind of guy, but I do agree with every point you made.
Thickbox had to be taken care of long time ago – it has been abandoned by the author in favor of better libraries for quite some time now. The plugin became conflicted since WP 2.8 if I’m not mistaken… instead of declaring it obsolete we were patching it up… I don’t understand why the core team is dragging their feet on this item.
I totally don’t understand the purpose of creating a brand new default theme with every release – it’s confusing to everyone and a waste of time. While new users will look for help from the community on how to use the new theme – the mature users have moved on to something more familiar/advanced and stable, and simply unfamiliar with the new “default” theme. New information and instructions will be needed and for how long? Till the next release? I’d rather see a single default theme with exceptionally polished code, well commented and grown into a showcase of all features that WP supports. New iterations of updates will make it stable. It will be utilized as base theme by many developers because of the code samples, community involvement and abundance of information available for it. It’s very shortsighted for WP core team to limit their default theme’s lifespan to a lifespan of a single release.
Thanks for stopping by Max. This whole comments thread proves that WP users want substantial features in WordPress even if they take longer to make. It was awhile since anyone in WP Core Development Team asked plugin developers and normal every day users what they actually need in WordPress. It is about time they listen to us all.
I agree with all the suggestions. And sometimes it seems to me, that the WordPress software development goes too much on the direction of what Automattic needs to improve for the WordPress.com.
And I think that’s bad, because (my opinion) the majority of users of WP.org is people like me, webdevs that sell sites to clientes and need a truly CMS, and not a “powerful” software for blogs.
You’re right! From what I see, Automattic judges a lot of suggestions by their 80/20 rule, so if THEY think it’s not useful for at least 80 percent of users then they reject it.
I hope they will be more brave again in the future and just implement the stuff that’s needed: take the search functionality for example – wasn’t touched since 2.x releases and it’s really outdated. There are awesome plugins like “Relevanssi” for that but this is a purpose of core. They need to add something like a little search API for easily add CPTs, plugin-specific stuff etc. This way it keeps lightweight but still flexible!
Major issue with good search is that it needs to modify database scheme to include full text index, to add extra indexes. Whatever deals with database structure needs to be in the core. And to be honest making good search doesn’t take a lot of code. I made something like that for a client a while back as a part of custom WP plugin, and it was all in 500 lines of code (including color coding results).
I’ve been thinking alot lately on this, and I think it’s time to fork this thing already, make a “WP PRO Version” and use something like backpress to still rely on the good parts of it (code essentially). I know it’s a pretty bold call, and I dont have all the dev expertise needed, but who knows who else more jumps on board.
Oh no, no, no… not a pro version. I remember when that happened to NotNetNuke and it was the kiss of death for ‘fun’ projects. It divided the community into the group of hard core developers who could take a chance, and the suckers (everyone else) who had to buy. Add ons were soon focused to the pay-platform and usability drifted apart.
I’m gonna call out again for the staging area since I’m not in the last few steps of the projects and with all the tweeks we’ve done -everywhere- I have a huge problem getting it over to my production site. I still haven’t resolved it really and look forward to stomach-aching testing at late hours.
Wp Pro version is already out there in a different form.
I have no grudges for WP is great software given away free.
After version 2.8 sites have becom S L O W and getting S L O W ER
At the same time Automattic invests few millions in WPENGINE hosting service whose USP is we would get your site FAST and FASTER without all those plugins
Of course the solution is damn costly and limited to pay by number of hits in case of overuse
There is no conspiracy. Automattic made an expected and commendable business move.
But this whole package in place of a professional edition defeats the ‘freedom’ concept touted by WordPress, and one of the big reasons why many of us switched over in the first place.
One more thing, I’d like a working importer to get posts across from blogger.
Some great points! I think the more people that get involved and put their points forward at WordPress.org the better and stronger the WP platform will become.
This is nice list 🙂 WordPress isn’t perfect, but I don’t think that there is a real alternative actually. So, I’m happy what I got for free 🙂
Great list, I think everyone would also like to see a dynamic image resizing API.
I’ll add another one regarding media management.
WordPress currently only allows to insert a image in a post using HTML, which leads to links to specific image sizes.
ie: img src=”image-300×200.jpg” width=”300″ heigh=”200″
That is a serious pain in the ass when you switch to a theme with different column widths and also leaves you with thousands of orphan images if you change thumbnail sizes.
A simple solution could be inserting the images using shortcodes with references to thumbnail sizes (like galleries)
ie: [img id=”234″ size=”big”]
My biggest complaint is how hard it is to turn the category pages into useful content pages. I feel like I have to break core functionality just to add a decent description. With the excerpts functionality, category pages can be so much more than the dumping ground so many seem to consider them.
I’ve not really had issues with the search function. But, it seems that it should definitely be modified to include custom fields, either by default or as an option.
I would love to see a new strategy for images. In some ways, the date thing makes sense. But since images with the same name are appended with a number anyway, it seems redundant.
As to templates, I would love to have two default templates with each release. A fairly complete one like Twenty-Twelve and a skeletal/clean one that designers could build from scratch.
And, I’m all for improved posts management.
Yup. Sounds like a collection of (some of) the major gripes I’ve had about WordPress core to date.
We have to remember that the majority of core wordpress developers work for automattic, and their main business is wordpress.com. So from that perspective wp 3.4’s focus on theme modification makes total sense. And the lack of attention to the other wonderful items mentioned in this post also makes total sense.
We as a community need to take a further step and move from want lists – which are great – to how are we actually going to support someone coding this and getting it into core. It’s not that core developers are averse to any of this, it’s more a matter of business focus for them. So if we code it up in collaboration with core developers it will happen.
I suggest we use an existing crowd funding service to propose some of these great ideas to the WP community and see which ones get funded, then hire a quality team to build them.
I’m CRAZY about this idea!
this seems to be the best place to start it out at http://www.fossfactory.org/ plus we’d need a person or team to manage it.
Comments are closed.