WordPress to Bump Recommended PHP Version From 5.6 to 7.0 By The Middle of 2017

Over the years, WordPress has been developed so that users and site owners don’t need to have a lot of technical knowledge to install and maintain it. Andrey Savchenko believes that WordPress is causing technical irresponsibility due to site owners not having to know technical details like which PHP version their site is running.

Is not knowing good for you as a site owner?

Few good things ever happened by accident and ignorance.

Running your own site takes learning and effort. It is admirable of WordPress to make that easier. It is self–serving and insincere of it to pretend you don’t have to know anything.

You have to know. You are responsible.

The post generated a healthy discussion on Twitter that included Matt Mullenweg.

Although it’s not happening at break neck speeds, there is an effort underway to increase WordPress’ recommended PHP version from 5.6 to 7.

In September of 2015, Aaron Jorbin published a post on the Make WordPress Core blog that explains what was accomplished in order for WordPress to officially support PHP 7. In the comments of the post, Rahul286 suggests that a warning be displayed that informs users they’re using an outdated version of PHP.

Samuel ‘Otto’ Wood succinctly explains why a notice wouldn’t do any good.

A notice by itself is massively pointless to show the end user, who likely neither knows nor cares how their hosting service runs.

However, it might be worth considering trying to detect the host in question, and providing valuable information for that specific host, such as links and other methods the end user can do to update themselves. Many hosts have a choice, somewhere, and if we know that, we can provide guidance.

Unlike WordPress’ Browse Happy notifications added in WordPress 3.2 that notifies users to update their browser, it appears WordPress is unlikely to perform a similar role with PHP versions.

Jorbin also commented on the article suggesting that developers reach out to their local user groups and inform users why they need to care about which version of PHP their site is using.

One thing everyone can do to help move these numbers is to talk to your local user group about why they need to care about the PHP version they run. Show them how they can upgrade, show them benchmarks of PHP 7 vs. earlier versions (especially vs 5.2).

Three months ago, Jorbin created a trac ticket with the suggestion that the PHP requirement for WordPress be increased from 5.6 to 7. Although many of the commenters agree with the move, Gary Pendergast says it’s best to wait. Dominik Schilling, WordPress 4.6 release lead, agrees.

Given that we have until the end of the year before we have to bump it, there’s no harm in holding off until major plugins are confirmed to work without notices or warnings. Until we can confirm that, I’m -1 on bumping the version for now.

This is remembering that we’re talking about the recommended PHP version. The recommended version should be providing an ideal experience for folks, we shouldn’t be bumping to a bigger number just because it’s there.

Given the PHP 5.6 support timeline, we’ll be bumping it by WordPress 4.8 at the latest, so it’d be nice to use those intervening months to ensure the UX for the wider WordPress ecosystem under PHP 7 is solid.

Major plugins like the WordPress Importer are not yet compatible with PHP 7. Developers are highly encouraged to use the time before PHP 5.6 reaches end of life to thoroughly test their plugins and themes for PHP 7 compatibility.

Mullenweg has made it clear that he will not use WordPress’ marketshare to force webhosting companies to upgrade to PHP 7 but rely on established relationships instead. In order to move the needle and get site owners to know and care about PHP versions, it’s going to take a continuous effort on the part of the WordPress community to educate them.

76 Comments


  1. I believe that bumping the minimum recommended PHP version is a good way for WordPress to influence PHP version usage, without causing folks to be left behind.

    First off, we have a practice of all unit tests passing on the day that a new PHP version is released – it was that way with PHP 7, it’ll be the same with PHP 7.1. You can always run WordPress on the latest version of PHP.

    Next, by only recommending officially supported PHP versions, there’s clear documentation that people should be running a recent version of PHP to get the best experience.

    There’s a regular argument that bumping WordPress’ minimum *required* version would force hosts to update their PHP version – we’ve found that isn’t the case. When WordPress’ minimum required version jumped from PHP 4 to PHP 5.2, it didn’t cause the large shift that was hoped for – instead, millions of sites were left unable to update. Even if it did work, throwing our weight around isn’t the way to behave in a large ecosystem.

    WordPress’ backwards compatibility (both with itself, and with the versions of PHP that are still being used) is far more important that arbitrary number bumps. If it turns out that the PHP 7 experience with popular plugins isn’t so good by the end of the year, I’d much prefer to see us delay the recommendation bump, than to cause folks to have bad experience with WordPress.

    Report


    1. I think this is just an excuse to avoid the php related decision and not sure why. WordPress is breaking BC with almost every release when moving to non BC jQuery library, changing smilies into emoji, reducing the expected quality of thumbs, introducing new end points etc.

      And it is not just the PHP version but also never actually deprecating anything, even when it is fully under WP control like link management.

      This attitude of trying to make everybody happy all the time is bad in the long term as usually you end up with bad code just because you are afraid of changing anything because it might bring down some site in southern Zimbabwe.

      People are not likely to be against upgrades, as they are upgrading, or at least ubderstand the concept, from their IOS, android windows etc. but if you don’t tell them it is time to upgrade they obviously will not.

      And if there are plugins that do not support 5.6 then it is a good time to kick them out of the repository.

      Report


      1. Yeah. Don’t get me started on the emoji thing. Arrg.

        Report


  2. I had this idea recently that WordPress should give users notice in the next version that in the next next version older versions of PHP would run more slowly. Not break, just run more slowly.

    WordPress could add 1/4 second sleep() per page load on hosted sites every time a new version of WordPress is installed on top of a 5.x version of WordPress.

    For example, if a site is on 5.2.4 then when WP 4.6 comes out it will give a full-page suggestion to the user to ask their web host to move to PHP v7.0 or the site will slow down on next version of WordPress. Then if they are still on 5.2.4 when they upgrade to WP4.7, add a 1/4 second sleep() per page load. If WP4.8 is installed and the user is still on PHP 5.2.4, add 1/2 second sleep() per page load. If after WP4.9 is installed and still on PHP 5.2.4 then add 3/4 second sleep(), and so on.

    Note the proposed sleep() duration would based on repeated updating WordPress but not updating PHP. Everyone would be given one WordPress version grace period, and the more times they upgrade WP but stay on PHP the longer the sleep() would get.

    Now I now that many people would probably hate this idea, but I bet it would get people to start caring about PHP level, and it would not affect the people who never upgrade WordPress so you don’t really have to worry about an upgrade breaking those kind of sites en masse.

    FWIW

    Report


    1. I’m sorry but that’s an absolutely terrible idea, so much so that I don’t know where to start. Besides punishing the user for something that they likely don’t have control over (other than changing hosts), it would immediately cause a well-deserved user revolt.

      Who would want to use software that would do something like that?

      Report


      1. Users have far more control than you give them credit for. Users can easily contact their web host’s support and say “I need to move to PHP 7.0.” And web hosts would get requests en masse, and so they would make it happen.

        The nail that sticks out gets hammered down. The ones that do not get ignored.

        And yes, I use and paid for SaaS software that does things like that all the time. But not SaaS solutions are run by more pragmatic people, not driven by ideology.

        Report


      2. Users have far more control than you give them credit for.

        And the correct answer to that would be to attempt to detect this is the case, and give them the necessary information to help them accomplish this feat of flipping said switch.

        Don’t punish people, but instead try to assist them. Works better, I think.

        Report


    2. worst idea ever (for some random definition of “ever ;) ). There is no reason to punish users just because they make the life of developers harder. If you don’t want to support something, just don’t, but if you support then do it properly.

      Report


      1. just because they make the life of developers harder

        Please try to carefully read the words you choose to comment on, as those were not words I wrote.

        It is no harder to write PHP code for 5.2.4 than it is for 7.0. But older versions of PHP are just at least twice as slow and less secure, which harms end-users more than it harms developers. You know, the end-users you seem to be rising up to championing for.

        but if you support then do it properly.

        Please explain what “proper” support would look like since you seem to be asserting you are the arbiter?

        Report


      2. Sorry Mike, but if the whole point is to gain some performance and better security then maybe core is right in sticking with 5.2 and letting site owners decide if they care enough to make decisions about their lower level utilities.

        The whole point of ditching 5.2 for 5.3 IMHO should be to use name spaces and gain better modularisation of wordpress code which when properly used should lead to better and more stable code.

        Each PHP version usually had a new language construct or library update that its usage can make your software better. The point of using the most advance version of PHP is to be able to make the wordpress core code better.

        Security is something that can be solved with good server admin (and we know that most security issues now are plugin/theme related, not server setting related) and PHP performance is the least important factor in a site’s performance. Those, by themselves, are not a great reasons to change minimum requirements and force people that do not want to put effort in their site to upgrade or to nag them to death.

        What does “proper support” mean? I don’t know but performance breakage in a software is usually considered to be a bug that need to be fixed.

        Report


      3. @mark k – Sorry, I can’t follow your logic in your first paragraph, nor in your paragraph about security “solved by good server admin.” No “good server admin” can solve the problems of running unsupported versions of software that has known security problems.

        And to say that the users must find good server admins or by implication they will have problems seems to me that you are “punishing” users, only in a different way.

        That said, after you denigrated me for “punishing users for making developers lives harder” you seem to be the one pineing for the developer improvements. And if you are going to admonish me for not supporting something “properly” it would be reasonable to expect you had some idea what “proper” would look like.

        Report


      4. IIRC 80% of wordpress sites run insecure versions of PHP. If your logic of “security problem” always becomes “A site is owned” then 80% of WP sites would have been owned right now. Since this is not even remotely close to happen we know empirically that running an old PHP version is not the worst thing a site might do and we know from researches that bad old plugins and themes are the main cause for security breaches. Upgrading to 7.0 will not make those plugins more secure then they are.

        And again “proper”…. When a brand sells me a milk for a week, I will be very upset if it will sell me water in the same package the next week just because someone said that cow milk is bad for humans. People know what is proper and what is cheating just because you can’t get your way by “winning an argument”, which is why you get all the negative reactions.

        Report


    3. I won’t even justify this by calling it an “idea”.

      Great way to take WP out of the game though. Thankfully not one other person on earth will ever entertain this “idea”.

      Report


      1. Exactly how would this “take WP out of the game?”

        (You have so little opinion of the value delivered by WordPress that you believe an incentive to proactively improve one’s own hosting would be all that would be required to kill it?)

        Report


      2. Ummmm….a zillion slow-as-hell WP sites out there. You don’t see an issue with that?

        Users that don’t upgrade would see an SEO hit (not to mention higher AdWords costs due to lower and lower Quality Scores) which costs them REAL money. Seriously – I run a massive AdWords account and if WP screwed me over like this I’d be done with it. You don’t see an issue with that?

        Users leaving WP in droves due to their site being SLOWWWWWWW. You don’t see an issue with that?

        I agree that it’s the worst “idea” anyone in the history of WP has ever thought of.

        Report


      3. Hi Mike,

        I get what you mean and it might work for a part of users. The issue here is that an other significant part of the users would simply not update their WordPress instance in order to save on performance. By that, these same users would undermine their websites security and put them in danger by not updating, and when they get hacked, they will start hating WP, giving it a bad reputation and pushing it out of the door. Note that both those who would have updated and those who wouldn’t would still hate it because they would feel they have had their hand forced, especially if it comes on a difficult moment in their schedule.

        Cordially, M.

        Report


      4. Hi M,

        Huge thanks for being respectful in your reply.

        I hear you. Still, I think you are projecting a scenario that is unlikely as game theory would predict otherwise.

        First, the stated purpose would be to improve security and performance, so if someone avoids the upgrade then it is on them for security problems, but WP in general. It would be easy PR spin to point out how WP is more secure because of these efforts and only people who completely ignore the issues are the ones to get hacked.

        Further, everyone acts as if this would place a huge burden on users when the reality is all it would likely take in 99% of use-cases is an email to the web host to request the upgrade. It is the web hosts that this might burden, but if most of their customers were requesting the upgrade then the web hosts would realize it is in their best interest to make it as easy as possible (this is where the game theory applies; you can’t assume the rest of the world won’t react to your actions and web hosts would react if most customers were asking for it.)

        And the people who don’t want to be bothered to upgrade? Many are probably are not upgrading WordPress either, and so they won’t be affected.

        That said, I was just throwing out an idea to see what people would say about it more to generate discussion about how to get some action on this rather than advocating for this one idea. I am disheartened though, because most responses here are about why something won’t work rather than collaborating to discover something that will.

        Report


    4. BTW, people go on and on about how great the WP community is. I beg to differ, at least in terms of tone and attitude as the replies to my comment indicate.

      So I propose an idea and instead of respectful responses with reasoned arguments as to why my idea might not be a good idea I get three (3) attacks instead:

      1.) I’m sorry but that’s an absolutely terrible idea,…
      2.) worst idea ever (for some random definition of “ever ;) )…
      3.) I won’t even justify this by calling it an “idea”…

      This is as bad as being back in elementary school. Count me as not a fan of a community where people respond in this nature to others offering up an idea.

      Report


      1. It was a really bad “idea”. Sorry if the responses hurt your feelings.

        Report


      2. Oh I can assure you it is not hurt feelings that I am experiencing. That would require me to actually respect your response. Instead I am feeling acute annoyance from having to deal with people who are both unprofessional and immature.

        Report


      3. So how should people in the community respond to you? Put out lips on your rear end?
        You need to learn that not everyone will agree with you and people are allowed to say it is a terrible idea. Every member of the community (including me) have put ideas and everyone has gotten at least one “that idea is terrible” replies.

        People can write a reply as small as two words: TERRIBLE IDEA.
        They can also write 10,000 paragraphs, it’s up to them.

        Report


      4. Thanks for validating my assertion, that many people in this community only seem to know how to response in a disrespectful (and unprofessional) manner.

        But there is nothing about disagreement that requires a person to act like an asshat.

        See M’s reply disagreeing with me, but in a respectful manner. Unlike others who responded to my comment, he (or she) is “Doing it Right.”

        Report


      5. If you don’t like people to disagree with you and treat you mean you really need to quit the internet.

        Nobody here is obligated to walk on eggshells just because you’re Mike Schinkel and you demand respect. This is a significantly less formal atmosphere here. You’re just a name among 1000 names. I doubt anyone here even knows who you are.

        Report


      6. Nobody here is obligated to walk on eggshells just because you’re Mike Schinkel

        Being respectful to someone if not walking on eggshells, and I don’t expect anyone to know me from Adam so not sure why you would position it this way.

        This is a significantly less formal atmosphere here. You’re just a name among 1000 names. I doubt anyone here even knows who you are.

        My point exactly. I don’t expect anyone to know who I am, and for that matter, who am I that matters? I am just a name among 1000.

        That said, I don’t see why asking people participating in a discussion to be respectful is a particularly outlandish thing to do. The world is a less dismal place when people treat each other with dignity and respect. Why should it be considered such an outrageous thing to ask for?

        Report


      7. I mean, your idea basically amounts to sabotaging WordPress without really accomplishing the goal of inducing users to upgrade. It should be no surprise that people who are invested in WordPress aren’t going to react well to that suggestion. You get a disrespectful reply because the suggestion to make WordPress intentionally slow is disrespectful to its users.

        There are dozens of ideas on how to push hosts to upgrade their versions of WordPress, the bluntest of which is to just make PHP5.3 the minimum version (which I honestly support). I’m sure you can think of other, decent suggestions yourself. But pulling out a bad idea, and then being annoyed that people call it out for being a bad idea, then getting mad that people call you out is far more childish than any of the responses I’ve seen in this thread.

        Report


      8. I was not annoyed that people called it a bad idea. I was annoyed that they were disrespectful in doing so.

        If people don’t react well it would more revealing of them then of me, don’t you think?

        Report


    5. What a great way to break shared servers running Apache — keep lots of sleeping fork zombies crowding the server RAM. Lovely, I’m sure that’ll be popular.

      Better idea: let people know that their site will run *faster* under PHP 7. Then let them nag their host to upgrade.

      Report


      1. @Ross – I had not even thought of that! You are right!

        Having a bunch of “sleeping fork zombies crowding the server RAM” would definitely get the attention of webhosts and thus not even reach the level of “punishing” users, as some people seem to think the idea to be.

        So YES! If WP did this then webhosts would be that much more incented to finally upgrade.

        Report


  3. A notice by itself is massively pointless to show the end user, who likely neither knows nor cares how their hosting service runs.

    Ya, but isn’t that precisely the problem? Your average user doesn’t know or care about the version of WordPress they’re running either but that’s why there’s a notice that tells them when they should upgrade.

    Report


  4. How about promoting PHP 7 from the plugins and themes side?

    The WP repository could add a PHP 7 badge to compatible plugins and themes. That badge should visually imply speed, because this is the biggest benefit of using PHP 7.

    I believe that this will drive end users to ask their web hosting companies to give them PHP 7.

    Report


    1. Excellent idea. It encourages plugin and theme authors to have their contributions up to date, it is social rather than a top down directive, and it educates end users that PHP 7 is something to pay attention to.

      Report


    2. The vast majority of plugins should already be compatible with PHP 7. Or, are you saying that you want to promote plugins that only work on PHP 7? What’s the criteria for the badge? If it’s just compatibility, then it’s not really anything special.

      Report


      1. The vast majority of plugins should

        It would be great if we stuck to actual facts not just “should”.

        If even 1 major plugin isn’t compatible and it affects 1 million+ sites then your “vast majority” comment is irrelevant and naive.

        How about we focus on the plugins that aren’t compatible and pressure the devs to ensure they are?

        Report


      2. If you want to state some facts about the number of plugins that do or do not work on PHP 7, bring them to the discussion instead of being a dick.

        Report


      3. In response to “Justin Tadlock · July 23, 2016 at 4:16 PM” (somehow there is no “reply” button below it)

        You said:
        “If you want to state some facts about the number of plugins that do or do not work on PHP 7, bring them to the discussion”

        but why don’t you apply this to yourself first?:
        “The vast majority of plugins should already be compatible with PHP 7.”

        And with this one you really went to far in my opinion:
        “bring them to the discussion instead of being a dick.”

        Report


      4. Johnny, I don’t really think it’s going too far to call someone out on their behavior when it’s a regular occurrence here in the comments.

        Report


      5. Hi Justin,

        My thinking that PHP 7 might still be an issue was based on the quote in the article above where Dominik Schilling says:

        there’s no harm in holding off until major plugins are confirmed to work without notices or warnings. Until we can confirm that, I’m -1 on bumping the version for now.

        That made me think that there might be a number of plugins that still need work for PHP 7 comparability.

        Best regards

        Report


      6. David, PHP 7.0 was released almost half a year ago. If there was any major plugin that do not work with it you would have heard about it by now.

        There is nothing bad in being conservative and wait for actual confirmation, but the lack of it do not by itself indicate there is any problem

        Report


      7. Justin, I’ve read all the comments of Ron in this thread and I don’t see anything improper in his words.

        Report


    3. How about promoting PHP 7 from the plugins and themes side

      Great idea.

      If it’s just compatibility, then it’s not really anything special.

      Marketing campaigns don’t need to be able anything technically special, they can just as easily be about awareness.

      Report


      1. Will it do the job of raising awareness? I suspect users who visit the repository would simply gloss over this badge after a few weeks of seeing it on nearly every plugin in the repo.

        I’m not sure what the current stats are, but I’m fairly certain a high percentage of users don’t even go to the repo anymore for their plugins.

        If you want to raise awareness, I’d recommend doing it in a place where users are spending their time. Maybe get the top 10 plugin authors to do something like the Browse Happy thing (is that still in core?). Or, maybe just get core WP to do that for PHP.

        I don’t dislike the idea altogether. I just don’t think the badge is the right approach. I doubt it’ll happen anyway.

        Report


      2. Just a badge is not the proposal. A badge and a marketing campaign, like Browse Happy.

        But it has to come from those in control of WordPress because otherwise it won’t show up on WordPress.org.

        Report


  5. There are two separate versions involved here. There is the recommended version and the minimum required one. The title of this article talks about the “minimum recommended version” which is very confusing. The body of the article talks about the recommended version and not the minimum required version.

    The minimum required version that WP will run on is 5.2.4 not 5.6. 5.6 is the recommended version.

    Report


  6. Three simple suggestions:

    (1) place an admin notice in the backend stating the current PHP version and link to WP page with suggestions to upgrade
    (2) email a notice immediately after the installation stage stating the PHP version and link.
    (3) put in a check for PHP version during an installation check for minimum requirements.

    Pick one or do all three. An email is already sent when an upgrade is done — so add it there. Additionally, plugin upgrade notices are available, so add it to the upgrade page.

    It doesn’t seem to be a big deal.

    Report


  7. This article is misleading. There’s no such thing as a recommended minimum. There’s minimum (5.2.4) and there’s recommended (5.6). At first blush this reads as if they’re bumping the minimum when they’re NOT. Please correct this as there’s enough confusion around this issue without adding more.

    Far from being a leader, WordPress has decided to remain a passive follower. Even though the perception is that WP has security issues, it happily runs under PHP 5.2 which has been EOL for years and no longer gets even security updates.

    There’s much pearl-clutching about throwing WP’s weight around but never any rationale for why that’s a bad thing to do occasionally, in the service of a good end. Right now, WP is allowing people to run in environments that may expose them to security exploits simply because… why again?

    Is the hosting environment complex? Sure. But face it, if WP mandated that the minimum version of PHP was the oldest supported version (5.6 at the moment), hosts would adapt. Make this the new policy and make it go into effect in, say , a year from now so that hosts have warning and time to plan and execute upgrades.

    “But some hosting clients might be running code that breaks under 5.6!” Well, 1) in a shared environment that host is opening the other clients to security issues, 2) They can segregate those clients on different servers.

    But all of this is moot. The core team seems quite good at reasons to not do things… they seem fundamentally oriented to a No based attitude. I can see advantages to such conservatism but it leads to stagnation and means the team will never be able to show leadership technically.

    Report


    1. It took me awhile to figure out what you’re saying but you’re right. I’ve edited the article title and text as there isn’t a recommended minimum version.

      Report


    2. +1

      This is not changing any time soon though. This argument has been tried many times before, but always failed :/

      Report


      1. Are you saying that since it’s failed so many times in the past that we should quit trying??

        Report


      2. Nope. Just pointing out that I suspect it’s going to take a great many more attempts before anything changes.

        Report


      3. And what are you basing this opinion on (besides it hasn’t worked in the past)? Every past failure that ultimately worked had to work the first time.

        Edison – light bulbs – 1000+ tries – I’m sure you know the story.

        Report


    3. I totally agree.

      I think whoever builds a site with WordPress *SHOULD* know something about the software and underlying importance of updating software like PHP.

      The authoritative voice of WordPress should tell them.

      Report


  8. WordPress team wants a large part ignorant and reckless site owners otherwise the 25% of the internet figure will decrease. Site owners should not have to take responsibility for their own sites, that is the WordPress message and the main selling point.

    Report


    1. Totally agree with what you said.

      Consider this also…

      Hosts and devs should also take some responsibility here.

      Devs need to ensure 100% that their plugins are compatible with 7.

      Hosts need to ensure that they have an easy way to upgrade accounts (preferably self-serve with easy instructions – a 1 click upgrade would help as well).

      Then, and only then, can WP start pushing the masses towards 7 (nagging is a good start).

      They really need to set a timeline that all plugins must be compatible with 7 and then nuke the ones that miss the deadline. WordPress is the only platform that can force change. The repo is littered with garbage plugins that need to go anyways. What better way to do it?

      Report


      1. Well WP chose to set a min requirement of 5.2 which was already obsolete when they chose it, just to not annoy site owners, but then they just got stuck there in way back when land. WP dropped PHP 4 support in 2011 and set min requirement to 5.2. At that time PHP 5.2 was already about to be EOL. If WP cared about moving forward they could have upped the min requirement every second year at least. But they didn’t and now all we get are stupid arguments saying that running a 10 year old version of PHP with no security fixes is a great thing.
        They talk about usage rates etc but we still do not have PHP per WP version stats on the WP stats page and we know from brief sample that its mainly the older obsolete versions of WP that runs on PHP 5.2. The dead sites are currently what decides which version of PHP that should be mandatory.

        Report


      2. Andreas,

        As much as I support plugins/themes/etc…being up to date…there are so many in the community that go on “you don’t need to have the latest version of x, the 2nd last version of x is fine”.

        Report


      3. I just think using stats that come from sites still on 3.2 is very missleading. This has nothing to do with sites being one version behind these are sites which are many versions behind.

        Report


  9. The article is about the recommended Version – what clearly is needed is a bump for the minimum Version – sometimes i think it’s important to use the WordPress power for a change for the future of the whole project.

    Because Hosting Providers wont act if there is no need why should they? As long as customers are satisfied ther is no need for them to change anything.

    But telling people it’s important to keep wordpress&plugins updated but keep using

    10 year old version of PHP with no security fixes

    it’s kinda weird isn’t it?

    If you announce the change properly and give providers a time to prepare and then maybe start with admin notices i think it would be a smooth transition and at some point when we the point is reached to actually break sites you can start preventing wp updates if the php Version isn’t right. – This might happen a year or more from now but it should start soon!

    Report


  10. I’m not sure why this has to be so hard. Since WordPress is the world’s most popular CMS, it makes scense that it would be running on the most up-to-date version of PHP.

    This is a message that should be taken to hosting providers, not end users. Hosting providers are always going to support the version of PHP WordPress says is required. If WordPress requires PHP 7, than hosting providers will make sure to update to PHP 7.

    Report


    1. All WordPress needs to do is this, with enough lead time AND to articulate an ongoing policy so hosts and devs know the roadmap. Me, I’d suggest this:

      Minimum: The lowest version supported by PHP. Currently, this is 5.6.
      Recommended: The most recent official PHP release.

      Remember, PHP end of life dates are publicly available at http://php.net/supported-versions.php. It’s not like they sneak up on people unannounced.

      Report


  11. I would think that having users using a new PHP version would be a good thing in general (faster speeds, more secure, etc).

    Granted, not all users would upgrade but this is something that hosting companies can easily solve.

    Selecting PHP version via cPanel is very simple.

    The only possible problem I foresee is plugin compatibility – that might be a problem because many plugins are not supported.

    Report


  12. WHM/cPanel doesn’t offer PHP 7 yet and the WP will have to wait for that to happen – at least.

    Report


    1. Nothing to do with cPanel. My cPanel have offered PHP7 since the beta days. It’s the choice of the host. And some will just wait for demand.

      WordPress is waiting for the hosting providers, and the hosting providers wait for WordPress to generate demand. This policies will never resolve. Someone has to act.

      Report


  13. I work for a hosting provider and we run Plesk on our web servers. We offer PHP 7. My WP sites run great on it, absolutely love the speed increases!

    Report


  14. i have used PHP 5.4, 5.5 and 5.6 but since moving to PHP 7 WordPress does seem to load a lot faster ,in saying that i made the transition to PHP 7 whilst my web hosting company had this still in beta development

    Report


  15. I reckon that site owners would greatly appreciate being told when their host (not themselves in most cases) has not updated their system software, be it PHP or system libraries on which PHP depends. Most currently have no clue that they are running on a platform that has not been receiving security and bug fixes. At a minimum, WordPress should be warning site owners that their site is running (limping?) on an unsupported version of PHP and as such is at risk of being hacked.

    I recently encountered a site owner who had a problem with one of my plugins. On inspection, I discovered that the site was still running PHP 5.3 (not the issue) with PCRE 6.6 (very old!) and thus didn’t handle the much beloved \K operator. That operator has been available since PCRE 7.2 in 2007. The punchline was that her host had told her that the server was well maintained and kept up-to-date! LOL!

    She was not best pleased to hear that her host was not keeping things up-to-date. I was not impressed much to discover that there are hosts pushing their old servers with no updates on unsuspecting site owners with the lie that they are new and current and shiny.

    I reckon that WordPress has a duty to at least warn site owners that they are running on really old systems, and link to a page explaining what that means and what they could do about it. That’s a minimum. Blah blah must still run on PHP 5.2 blah whatever sure, but at least let people know they are at risk because currently many don’t even know.

    Further to that, the About page has the opportunity on every WordPress update to promote the recommended version(s) of PHP and explain the benefits of upgrading. It could even explain why PHP 5.6 is still the recommended while PHP 7 could be better for their site. On sites running less than the recommended version, a bold notice could be displayed at the top of the About page warning site owners that they are not running a recommended version.

    Sitting back and hoping that site owners will discover a page on a site they never visit is not going to help site owners. Most quite simply don’t know to look, and trust their hosts to provide them with the best. Many hosts have not earned that trust.

    Report


  16. I have to say I’ve found WordPress and Automattic lacking in encouraging PHP upgrades, it has been very underwhelming.

    Instead of Mike’s idea of artificially slowing pages, why not just quantify how slow php is in its current incarnation compared to PHP 7+.

    I’d like to see a nag notification on every release saying something like: “X WordPress users have upgraded to PHP 7+ since the last release. PHP 7 runs x times as fast as your current version, learn how to get an upgrade .”

    Report


  17. For the hell of it I upgraded to PHP7.

    Increase in performance = ZERO in my case.

    More secure is worth it though I suppose. At least it didn’t break anything.

    Report


  18. This is a form of bullying. Which is what Google has done regarding mobile friendly websites. Update your website to mobile friendly or we are going to penalise your site.

    You are saying no different. Upgrade your php or we will penalise you.

    Bullying

    use superior strength or influence to intimidate (someone), typically to force them to do something

    Report


  19. Being practical… let’s suppose that the recommended version is bumped up from 5.6. What sort of impact is this expected to have? Does the data show that it’s even worth discussing? Only 28% of sites currently run on the currently recommended version or later, according to https://wordpress.org/about/stats/ .

    Of course, we can’t enter an alternative universe and discover what percentage that would be if the recommended version was something less (or didn’t exist at all). But, I’d suggest that 28% at this stage means that, whatever is recommended, will not translate into a useful outcome for developers. The WordPress website could recommend anything it likes, ranging from running PHP 5.6+, to feeding slices of Dutch cheese into your power supply every 5 minutes. But if the vast majority pay no attention to the advice (and presumably a good chunk of that 28%+ are running PHP 5.6+ for reasons that have nothing to do with WP recommending it – PHP 5.6 has had a stable release since August 2014), then, well, I suppose I ask myself why I even invested time in typing this out! :-)

    Report

Comments are closed.