Highlights From Matt Mullenweg’s Q&A Session on Product Hunt

On September 22nd, Matt Mullenweg participated in a question and answer session on Product Hunt. Product Hunt is a site that lets users share and discover new products. In the session, Mullenweg answers questions like what advice he’d give to his 20 and 25-year-old self and how he combats burn out. Here are a few highlights from the session.


Nikhil V – Hello Matt, with the upcoming release of the WP REST API in WordPress core, what do you hope to most see built with its capabilities?

Matt Mullenweg – I’m hugely excited about the REST API! Once the infrastructure is in, I think we’ll see really widespread adoption from plugins that right now are creating their endpoints in very ad-hoc ways. Once we iterate a bit more on the core content endpoints and authentication I think we’ll see a lot more specialized interfaces built on WP as like a content kernel, an engine powering a wide array of applications and interfaces you’d never imagine having WordPress behind them today.

Erik Torenberg – If you could go back in time and give advice for your, say, 20-year-old self, what would you tell yourself? How about 25?

Matt Mullenweg – To my 20-year-old self I would say to be okay with things building up over time. Something I didn’t appreciate until recently is that there any many productive decades ahead with which to build the things that I feel need to exist in the world. One of the best things I did then was avoid any press or capitalization on my age (to the extent I could) because youth is an ephemeral asset and just a novelty in business.

To my 25-year-old self I would tell him to slow down and think about health more, I burned the candle at both ends that year. I would also say to not get as caught up in mailing list and IRC arguments and discussions, just focus on building.

Sydney Liu – In the early days of WordPress, what was the 20% that got you the 80% of the results?

Matt Mullenweg – There’s no simple answer there, if I had to pick one thing it would be luck.

Kristof Bernaert – How do you see WP about 5 years, or how do you want it to see?

Matt Mullenweg – In 5 years I think that WP will be infinitely easier to use for both power-users, developers, and newbies. You will be able to have a full experience of core, plugins, themes, docs, and support in several dozen languages that will be as large or larger than English. There will be over a million people making their living on top of the platform. Most people will be creating from touch devices, and their content will mostly not come from keyboards (virtual or physical) as an input device.

Tys Bradford – Your accomplishments with WordPress have literally changed the face of the internet and enabled people without programming skills to make beautiful and functional websites. Do you think this could be replicated ever with mobile applications (iOS/Android)? I have seen a few companies try, but none have come close to delivering in the way WordPress is able to.

Matt Mullenweg – Thank you! The ecosystems and app stores for iOS and Android right now are too locked down, and the discoverability and overhead of apps is too distorted, for an approach like WordPress’ to work for mobile applications. The environment for the foreseeable future there is going to favor more centralized apps, my only hope is that at least a few get really successful (including ours) that aren’t advertising-driven.

My Question Was Not Answered

Unfortunately, my question was not answered. However, I’m going to ask it in this post with the hope that he’ll answer it in the comments.

You’ve spent a lot of time, money, and energy trying to create a great mobile experience in WordPress via dedicated apps. As more of the WordPress backend becomes responsive, how will this affect (if any) the priority level of developing the mobile apps versus putting that time and effort into a 100% responsive WordPress backend?

Mullenweg answers a lot more questions than what’s listed above but those are the ones that stand out to me. You can read the entire session on the Product Hunt site.

Would you like to write for WP Tavern? We are always accepting guest posts from the community and are looking for new contributors. Get in touch with us and let's discuss your ideas.

47 Comments


  1. Regarding your question Jeff… there’s always question time at “State of the Word” at WordCamp US. :)

    Report


      1. I thought posting it here was Plan B?

        Report


  2. This comment from Matt totally baffled me and I would love more explanation:

    We’re experimenting more here in the coming year. The things we’ll emphasize are the things that make your plugin a good citizen in the world: support your users, don’t upsell, update frequently, have multiple contributors, translate, test frequently, be responsive to security issues.

    My emphasis added on “don’t upsell”. What part of trying to make a living from selling WordPress plugins makes me a bad citizen? This statement baffles me, not only because the freemium model is what so many of us use, but it’s the model that pretty much every Automattic product from WordPress.com to their plugins has used.

    Does Automattic intend to make all upsells for WooCommerce, Akismet, Jetpack and VaultPress free in order to be a “good” citizen? I doubt as that would be insane, they need to make money like the rest of us. Freemium is a great way to make money while providing something of value for free.

    It seems like there is some bias against using WordPress.org to make a living that confuses me. What’s wrong with monetizing free software?

    Report


    1. Do you think maybe he’s referring to the Dashboard? Don’t put a bunch of alerts and messaging about Pro versions of plugins?

      Report


      1. That’s a good guess, Chris. Worst offenders in terms of upsell in the dashboard that I see on a frequent basis (working on lots of sites): WPMUdev, Jetpack, Yoast.

        Really upselling should be left for the Settings screen where one can discreetly offer additional features or in the Description in the repository (“additional features available in Pro”).

        Maybe Otto or Mika could whip us some more guidelines.

        Oh wait, those guidelines already cover a lot of this miscreance:

        It is fine to include an Upgrade prompt on the plugin admin page, but not throughout the blog. It is acceptable to embed a widget on the dashboard but this should be the same size as others and be dismissable. It’s fine to put an error message at the top of the admin for special cases, but it should be linked to a way to fix the error and it should be infrequent. Any form of “nagging” is absolutely prohibited.

        Yoast in specific has been nagging us and all of our clients throughout the entire site to remove the harmless RSS Footer plugin and replace it with full blown upselling Yoast WordPress SEO.

        Note that if you do include what we consider to be “advertising spam”, or attempt to game somebody else’s advertising system, then we will not only remove your plugin, but also report your code to the advertising system’s abuse mechanism as well. We do not react kindly to spam.

        WPMUdev is running full ads surrounding their plugins. I believe All in One SEO does something similar. In the case, of WPMUdev, Samuel and Mika can’t shut it down as we’ve actually paid to have our dashboards hijacked. All in One SEO is another matter (WordPress.org repository). Yoast does something similar (WordPress.org repository).

        Let’s hope these rules will be applied equally and across the board, no “some animals are more equal than others”.

        Any other good examples come to mind?

        Report


      2. There are plenty of examples of plugins being overly aggressive with upgrade prompts. There isn’t anything keeping non-repo plugins from doing so. .. but plugins found within the repo should follow the guidelines.

        What I see frequently as well are plugins which are in violation of section 5 of the Detailed Plugin Guidelines:

        Trialware is not allowed in the repository. It’s perfectly fine to attempt to upsell the user on other products and features, but a) not in an annoying manner and b) not by disabling functionality after some time period. Similarly, you cannot “cripple” functionality in the plugin and then ask for payment or provide a code to unlock the functionality. All code hosted by WordPress.org servers must be free and fully-functional. If you want to sell advanced features for a plugin (such as a “pro” version), then you must sell and serve that code from your own site, we will not host it on our servers.

        There are plenty of plugins, some found right on the first few pages of the Popular listing, which violate this. It’s a very tricky rule to work around when developing a freemium plugin.

        On this one in particular I feel it is important for us all to play by the rules because the development costs associated with doing so can be significant. It’s not fair that some developers spend extra time and money to do it right, while others do not.

        Report


      3. I’m unaware of any trialware plugins in the repo, and I wrote those guidelines.

        If you have a problem with any plugins, feel free to email me or plugins at wordpress org and I’m happy to explain the details of why or why not, and take any actions necessary. We’re fully open to input. :)

        Report


      4. Is the entire quoted section in reference to trialware? I though that was only the first sentence.

        The sentence starting with “It’s perfectly fine to attempt to upsell…” is a new sentence, right?

        Report


      5. @Piet, I don’t understand your question. The entire guideline is talking about trialware and crippleware. That’s why it starts out talking about trialware. Guidelines are generally about a single topic. This isn’t a mix-and-match or a choose your own adventure type of situation here. Paragraphs should be taken as a whole, not in pieces.

        Report


      6. I agree with you Jason about crippleware. Which plugins in particular were you thinking of?

        Report


      7. I’ll follow up with the repo team privately about it… I don’t want to name names publicly and be wrong about my interpretation of the rules.
        What I see though are plugins that have disabled/sleeping functionality in their code which is hosted in the repo. Whole big swaths of the plugin. Major features. Maybe as big as headlights on a car. Kind of needed, right?

        Those features are disabled unless you visit the developers site, sign up for an account, and pay for a license. There is then an activation button in the plugin which pings the key server, sees that you’ve paid, and gives you access to these other features. I would argue that is a violation of *you cannot “cripple” functionality in the plugin and then ask for payment or provide a code to unlock the functionality. All code hosted by WordPress.org servers must be free and fully-functional. *

        In the case of our own plugin (Postmatic) we abstracted any functions which are part of our premium plan up onto our own Rails server. We then built an api. Once you are a premium user our plugin relies less on WP and your webhost and more serves as a gateway to our own server… where all of the good premium magic happens. We built it that way (a) so the gazillions of free users do not put stress on our own servers, (b) to give our paying customers a guaranteed quality experience that *just works* and (c) to follow the repo guidelines. Free users use the limited capabilities of the server they are hosted on, Premium users get to overcome any limitations on their hosting account by tapping into the the infrastructure we provide.

        Again. I don’t want to start a public witch hunt….. but there are benefits to us all being accountable to the rules.

        Report


      8. Jason, I don’t think without naming names publicly progress can be made. I understand it’s more politic to divide the cake behind closed doors but to my mind lacking grievously in the transparency to which an open source project should aspire. Otto’s active presence in the comments here is a comfort, as he is one of the principal arbiters (and authors) of these rules.

        To be honest, as much as I like Postmatic and believe at your current price point you are offering tremendous value to even the medium traffic WordPress site, Postmatic falls into another category undermining real OSS and independent web publishing: serviceware. I.e. instead of just letting my server handle the hosting in a freestanding way, the site owner is obliged to sign up and share all user data with you and cannot just move the data back to his or her own hosting.

        I know the plugin moderators at WordPress.org have their own ideas about serviceware while Automattic has their own very liberal notions on serviceware. In your favour, the loophole in OSS and GPL2 which Automattic are exploiting is the one you’ve chosen Jason. Hence you are unlikely to run into any issues, as it would shut down the main commercial plays of the project’s owner (well sponsor at this point).

        Sorting out fair play in the commercial realm will decide whether WordPress remains a viable long term independent publishing tool or becomes just another paid CMS like those which it replaced ten years ago. What we are debating here are existential questions for those of us deeply invested in WordPress.

        Report


      9. Hey Alec,

        It’s not my place to decide if a plugin is in violation or the repo guidelines or not. That is up to Otto and others and I feel that due process by the mods is in order before I start pointing fingers. I could be wrong from both the technical as well as interpretive angles. It’s not my job really. If anything comes of it I’m sure you’ll know. If the guidelines were more explicit or there were well document precedents of violations I’d feel more confident.

        Postmatic falls into another category undermining real OSS and independent web publishing: serviceware. I.e. instead of just letting my server handle the hosting in a freestanding way, the site owner is obliged to sign up and share all user data with you and cannot just move the data back to his or her own hosting.

        You’re completely wrong here. Postmatic does not touch your user data, share it (even with us), retain it, or even think about it.
        You are not obligated to do any of those things with us. And there is nothing to move back to your own host because it never went anywhere in the first place. If you are running Postmatic the only data of yours that flows through our servers is diagnostic log data of incoming comments and subscription confirmations which are retained for 30 days to help with support tickets. You can read about it in our privacy policy. In fact, i’m going to put the tldr; here:

        You own your data, and your relationship with your users.
        We’re just here to help.
        We don’t profile your users, or interrupt them with advertising, or sell your data in any way. Fantastically-magical email-based commenting is our only product — not you, your users, or your data.

        We choose to operate on an affordable, subscription-based revenue model with a generous free option. You and your users can feel free to post and comment on your site without the worry of being profiled, farmed and sold down the river. That’s not us, and we won’t have any part of it.

        The only data that makes it to our servers is the data that we need to be able to reliably deliver emails for you (only true on Premium, actually), and even that is governed strictly by our privacy policy. Our commitment to openness and WordPress best practices keeps your data in your hands. Postmatic is built on the native WordPress comments and user systems to make sure that what’s yours, stays yours.

        To be clear: we use native WordPress comments and users, which live in your local database and always will.

        Report


      10. Strangely enough WPtavern is using Postmatic but comments are not native but loading (slowly) in javascript. Looking into the source code, I see Epoch there.

        We’re going a bit off topic. I hope you are able to remain true to those ideals of the sanctity of user data. I’ve seen so many services fail that test over time (especially during changes of ownership) that it still concerns me.

        PS. If your servers are US-based, you won’t be able to keep those promises even if you want to as anything on your servers belongs to the government at all times. Concrete example: in the case of commenting software, there could be political sites where people prefer to comment under aliases.

        Report


      11. WPTavern is using Postmatic (for comment notifications) but also our other plugins Epoch[1] and Crowd Control[2]. Epoch is a javascript replacement for the comment template that came with your theme. You didn’t notice that posting a comment here has gone down from about 12 seconds to 2 ;P

        Just like Postmatic… Epoch is fully native as well. It changes *how* your comments are rendered, *not where they come from*. Your data is yours, we just add a pretty layer to the top. Tavern could switch off Epoch at any time and go back to using the comment template that came with their theme, or use Jetpack, or something like wp-discuz or de:comments. The data is safe, local, and untouched by them all.

        I think that Disqus has done so much damage in the commenting space that everyone is so reactionary about the whole thing. We hear nothing but complaints about us being another third party commenting system. We’re not. We’ve just done a lot to make native commenting easy, fun, and perform like… well.. not native comments.

        And you are right: anyone that thinks that their data is safe when it is flying around via email hasn’t done their homework. We make no claims that email is secure. We once had a user get very aggressive with us about the contents of their comments flowing through our servers. My response was *why would i spend 2 years building machinery to read your comments when instead I could have just visited your website?*.

        Thanks for the nod, Jason

        Report


      12. Jason, you don’t have to guess. Point me to the perceived problem, and ask. It’s that easy. We look at these all the time. We won’t judge you based on asking. If we disagree, we’ll try to explain why. All plugin review team members are smart dedicated people who do their best and don’t compromise on issues or principles. That’s why I chose them. :)

        I’ve been very picky on the plugin review team, mainly because it gets sensitive security information. But I definitely trust their judgement and respect their views. Your concern will be handled properly, but perhaps not timely. It’s a small team. ;)

        Report


      13. A public list of cases and resolution would go a long way to levelling the playing field Otto. While I respect your team and appreciate your enormous efforts to keep feedback loops short, I do feel (and have told you so in email in detail) that you are not always maintaining a level playing field. Ironically Yoast of all people among among the non-Automattic properties seems to be given a lot of wiggle room in both reviews and in actual plugin behaviour.

        Report


      14. Alec, my plugin SO Cleanup Yoast SEO is steadily becoming more popular with almost 50% active installs of 2k downloads since end of April this year. My best guess is because it gets rid of all the annoyances you have summed up, and then some. Check it out.

        Report


      15. Piet, that’s a very cool idea. I didn’t know about it as we normally use our own Simpler SEO (no annoyances of course, intelligent defaults, very safe for client use). For some of our external clients, your plugin will be a godsend. Thanks for the heads up.

        Report


      16. Glad you like it Alec and thanks for your kind review of the plugin :)
        Yeah, I got so sick of it at one point (that was the time when the about WP page was redirected to the Yoast SEO about page and nags couldn’t be clicked away), that I said “till here and no further”.
        It seems that with every update team Yoast thinks of new ways to annoy its userbase, so I have to keep a close eye on their updates too.

        Report


      17. I stopped using the Yoast plugin because of those annoyances you talk about, Piet. I really don’t know why that plugin is so popular.

        Report


      18. premium member of wpmudev here, i actually really appreciated their dashboard, seems to me to be update notices, quick convenient access to support forums and direct individualised support from staff, quick download of all their products etc. i dont look at it as advertising. its convenience. possibly people with a lesser plan gets advertising of products they don’t own ? not sure because i am not on that plan. i have noticed nags and ads with other plugins rtmedia and many more but its all appreciated and welcome to be informed of their updates if i use and appreciate their products. obviously not everyone will agree, but devs have to live and make money for their hard worked products that is seen in the dashboard ONLY because the product is actually being being used LOL. downside to this is white label requirements for clients etc. just my 2 cents :-)

        Report


    2. From what I can tell, Matt has always very reluctantly tolerated product-based WordPress businesses. It seems like he much prefers for people to make money with services.

      Report


    3. I’m with you on this one, Josh. Our business is built on offering awesome value, in the repo, for free… without crippling the product at all. And what we’ve created is a tremendous gift for the WordPress community – it solves an almost decade-long problem while also giving any WordPress site something no other blog platform has. So, IMHO, that’s huge. And its free.

      We then upsell some icing on the cake such as providing huge volume email sending, inlining, and other services that your $5 hosting plan can’t offer. Same plugin, your stuff just goes through our api instead of out via phpmail. And we’ve put everything on the line to create it.

      If that model becomes a problem than thousands of WordPress sites lose the ability to have awesome comment notifications with support for replies. It’s back to the stone ages of one-way contextless subscriptions.

      Postmatic is just one example of hundreds of plugins (yours included as well) that provide amazing value. For free. But need to make some money back in order to exist.

      Otherwise… what. A fragmented plugin ecosystem where the innovation is outside of the repo because it needs to produce a return? More commercial plugin marketplaces which are disconnected from wp-admin and fragmented all over the place.. with terrible customer experience?

      Big topic… but if WordPress is truly shooting for 50% we ought to start talking about it now.

      Report


    4. He was specifically talking about plugin search results. Upsell-heavy plugins tend towards lower ratings and bad reviews and annoyed users in the support forums. If we start taking things like that more into account, well…

      We’re gathering lots of useful data on plugins via reviews and other feedback. It only makes sense for us to count that as a factor in promotion of plugins, no?

      Report


      1. That’s interesting that upsell-heavy plugins tend to offer a lower quality of service and experience. I would completely expect the opposite.

        But how does this process extend to plugins which have terrible reviews, upset users, a bad experience, but are not selling? That are 100% free? There are an equal number that fall into that category. Are you foreseeing a general reckoning for poorly performing plugins?

        Report


      2. I don’t make predictions, I just point out things I see and things I think. We currently don’t give ratings much weight in search results. A revamp of search might give them more.

        Report


      3. Then why have a ratings system that can be easily abused? Seems counter-productive to have a feature that doesn’t add much weight to WordPress’s supposed quality assurance?

        Report


      4. I’ve dug into this a bit (because ratings make a difference to our bottom line) and it seems that the repo mods have given a lot of time to making sure that if the reviews system is abused, there are checks in place to notice irregular patterns and take the appropriate action.

        I’m sure it’s a cat and mouse game and that there are people still gaming it all the time in low-level ways. There are not any written policies surrounding the reviews system so I don’t doubt that most of the abuse is just a marketplace at work….business doing what business does. The repo and reviews system weren’t designed taking into account commercial and upselling plugins. All of this will probably need some attention sometime soon.

        Report


      5. @Jason

        …if the reviews system is abused, there are checks in place to notice irregular patterns and take the appropriate action

        I think that as long as plugin authors can rate their own plugins this “system” is completely dysfunctional.

        Also users using the review system instead of support should be caught I think.

        Perhaps an idea to add a report button or something like that. No doubt that such a button will be abused too, but it is very easy to “punish” abusers of that button I would think.

        Report


      6. Man I really hate people leaving one star reviews when really they should have just created a support ticket. A way to report that would be so fantastic.

        Report


      7. So should we expect Akismet, VaultPress, JetPack and WooCommerce to be punished in the ratings for having upsells? That would be really dumb, as those are all good plugins.

        Report


      8. Hi Josh,

        Being a good plugin and owned by Automattic or another made man should not be a license for mayhem. We don’t want to end up with some animals more equal than others.

        Surely promotion can be done discreetly on a plugin’s settings page without poisoning the user experience.

        Whatever we allow Akismet, Vaultpress, Jetpack and WooCommerce should be allowed any developer. Keep that in mind.

        Report


  3. Jeff’s question is ironic considering responsive web app driven Twitter app has become painful for me to use on my iPad Air 2. It’s laggy and simply doesn’t perform as well as native apps.

    As a user I hate web apps that masquerade as native apps and are nothing more than web sites loaded in a native app wrapper. My issues with the Twitter app are a great example.

    Facebook’s iOS apps used to also be responsive web apps in a native app wrapper. They also used to perform horribly so they had to backtrack and make them truly native.

    Web sites and web apps should be responsive. They should be accessible on mobile. But no matter how many times someone tells me that building a mobile app using web technologies instead of as a truly native app my experience as a user tells me otherwise. My experience as a developer agrees.

    But some people will agree to disagree on this one. I can’t stand it. Give me a legitimate native app. Leave the web for a browser.

    Report


  4. Hey,

    Thanks for including my question Jeff! I definitely enjoyed that AMA yesterday and I’m glad I got an answer. It was also great to see the many other awesome questions, not even related to WordPress, that Matt was able to answer.

    And +1 with David, you always have the State of the Word to ask yours :)

    Report


  5. There was one other tidbit that caught my eye in response to this question:

    * A cool internal beta of WordPress :)

    What is this cool internal beta of WordPress?

    Report


    1. Well, the question was asking about phone apps, so I assume meant the beta version of the WordPress phone app. There’s apps for iPhone and Android. The Android beta is pretty nifty, but Matt probably uses iPhone a bit more.

      Report


    2. Good question, I don’t have any idea what that could be. The earliest version of the WordPress mobile app was considered something like an internal beta of WordPress. It’s far from that now.

      Report


  6. Without upsells, there is no freemium model. And very few good devs would bother to make things that were helpful to the community at large – and support those things – if they weren’t collecting something for their time and effort somewhere. People can’t just give everything away, that’s madness.

    Report


    1. Kim, no one is saying there shouldn’t be pro plugins. Whether free versions of pro plugins should be able to take over the dashboard, the plugins screen, the admin bar with advertising is what is in question.

      Without advertising people would still buy groceries (basics), people would still choose leather seats for their cars (luxury features). There are other places to promote a plugin than in-your-face in a WordPress install.

      As WordPress goes more and more commercial, we’ll need stricter and stricter rules. The only way to enforce any kind of compliance is via inclusion (exclusion) from the WordPress repository.

      Report


      1. In fairness, any kind of everyday purchasing started as some form of advertising. You still need to make a decision as to what grocery store or leather seat seller you go to, and advertising is a part of that process.

        Report


      2. Advertising is fine but not unlimited inside a WordPress install. That’s like putting up ads inside someone’s windshield. Hopefully the plugins team at WordPress.org will say enough is enough and start enforcing the existing rules more stringently.

        Report


      3. While I agree with the premise, how would you like to see devs being able to upsell? A simple link within the plugin description on the plugins admin page, or something different? One of the things I see often is great devs not being able to maintain development due to underfunding or non-upsell opps.

        Just curious how this could happen while still not offering space within the dashboard?

        Report


      4. Danny, as I mentioned earlier, the plugin’s own setting page is a good start. Or an upsell (as many of freemium authors already do) in the plugin description. Really the dashboard should be out of bounds to both plugin authors and to Automattic. WordPress’s starting premise is free OSS CMS not a billboard for upsells.

        Report

Comments are closed.