Recommendations, Collections, and Contributions (oh my!)

We’ve just released AMO 5.0.9 with a ton of new features and bugfixes out into the wild. There’s something for everyone in this release, so take a look!

Add-on Recommendations

Screenshot of recommendationsSome things go great together: peanut butter and jelly, popcorn and movies, Firebug and YSlow. Starting today, when you’re looking at an add-on’s listing, you’ll see several other add-ons that are often used together. We base these recommendations on the add-ons our users have grouped together in collections.


Last month we launched our Contributions pilot, a way for add-on authors to accept monetary donations from users of their add-ons. We’ve gotten great feedback from users and developers, and are now expanding the reach of contributions on AMO. Add-ons that have turned on Contributions will now have contribution boxes in search results and category browse listings, where many installations take place.

Screenshot of contributions on home page

While these new boxes help surface the existence of contributions, some users aren’t likely to make a donation without first testing the add-on to ensure it’s useful to them. With this release, developers have a new way to accept donations after a user has tried the add-on by simply pointing users to their add-on’s new about page.


Since the launch of Collections in June, add-on fans have created almost 30,000 collections, resulting in more than 6.5 million add-on downloads. This release includes many new features to make Collections even more useful.

Logged-in users can now rate collections and sort by those ratings to help the best collections bubble to the top.

Screenshot of collection rating and usage stats link

To help determine the popularity of a collection, authors can now access a collection statistics dashboard that shows information on the number of subscribers, ratings, and downloads over time.

We’ve also added a treat for users of browsers that support localStorage (including Firefox 3.5): recently viewed collections. When browsing through the collection directory, users can now easily get back to a collection they saw, without having to log in.

Public Stats

In early 2008, we gave add-on developers their first look at the shiny new Statistics Dashboard for each add-on hosted on AMO. Currently some dashboards are public and some are private, but we’ve never offered an in-depth look at the aggregate statistics for all add-ons hosted on AMO.
Today we’ve launched a detailed view of AMO as a whole, with data on add-ons, downloads, active users, users, reviews, and collections. Go on, check it out!

Add-on Validation

Earlier this month, we announced the arrival of the new add-on verification suite on AMO. At the time it was an optional tool developers could choose to run for their files. With this release, the tool will automatically be run when uploading a new file, and the results of the tests will affect the submission process. Editors will see the test results and any flags when reviewing the add-on. We hope to help developers catch common problems before an editor’s review.

Screenshot of validation results

We’d like to extend a huge thanks to the Web developers and testers who have contributed their time and talent to this release.

16 comments on “Recommendations, Collections, and Contributions (oh my!)”

  1. Wladimir Palant wrote on

    Interesting to compare extension’s download and ADU numbers with AMO’s total counts. For Adblock Plus both downloads and pings are around 5% of AMO’s total. Element Hiding Helper’s share looking at pings is twice as high as its share at downloads. And for JavaScript Deobfuscator this is reversed – its downloads share is almost twice as high as it ADU share. Now I only need to figure out what that means…

    Altogether, the normalized numbers are less stable than what I’ve hoped for. Unfortunately, the ADU numbers don’t go back far enough – we’ll have to wait to see whether it is possible to recognize trends there. One (not very surprising) trend is already visible – Adblock Plus share rises on weekends while JavaScript Deobfuscator share falls there.

  2. Ken Saunders wrote on

    Man, all great stuff.

    AMO has always been my favorite place to get cool toys.
    Now the toy store kicks a**!

    You all rock!

  3. Pino wrote on

    The site develops fast and in a good direction, but I don’t like the new “support this add-on” links. It almost looks as if you have to pay for addons these days. I think a donation system is a nice addition, but it should not be forced on the users so much that it no longer looks like a donation but more like an obligation.

  4. BobB wrote on

    Is it possible to use the Add-on Validation system for private extensions that are not on AMO?

  5. Jason wrote on

    Why don’t you take the add-on contribution system to the next level, and allow developers to charge for add-ons? That way, there will be even greater incentive for developers to write add-ons, so that they can actually make some money rather than doing it for free (with the hope of donations). I know that in theory, add-ons are open source (since they are just XUL + JS), but this is the sort of thing that could be huge and really up the quality of a lot of the add-ons.

  6. Wil Clouser wrote on

    Is it possible to use the Add-on Validation system for private extensions that are not on AMO?

    Not yet, look for that in late September

  7. Ken Saunders wrote on

    “Ken Saunders says:
    August 21, 2009 at 11:23 am (from

    Ever hear of Windows Marketplace?


    I’m fine with developers offering pro versions of their add-ons as long as they make it clear and the ones on AMO are NOT demos and/or have disabled functionality, but I don’t want to see AMO or Mozilla ever take that road.

    The most popular add-on for Firefox is free, and from what I’ve read, will remain so despite the developer getting offers to write it for other another browser. He could make bucket loads of cash doing so, but has chosen not to as have the majority of add-on developers.

    They always have the choice of writing add-ons to be sold on Windows Marketplace.”

    To add to that, I really don’t see how $ can improve the quality of an add-on or many other things.
    Firefox is free and selling it wouldn’t make it any better than it already is.

  8. Matt wrote on

    There seems to be a bug immediately after the validation process when submitting a new addon. After all tests had passed for my addon, click continue, and I’m on a page that says “Access Denied”

  9. ycc2106 wrote on

    One cool thing to have would be to search ‘this collection’ Searching all collections isn’t very useful for the ‘addon searcher’.

  10. Brett Zamir wrote on

    Ken Saunders wrote, “I really don’t see how $ can improve the quality of an add-on or many other things.”

    To test this, you could start a service industry, do nothing besides state your goals, and see how many volunteers join to do work for you. GLWT. 🙂 Firefox only got the momentum that it has because it was based on Mozilla which was based on the for-profit Netscape, albeit an admittedly historically gracious one, which donated the code after losing the browser wars of the time. Now it has taken a life of its own, though even here it is still supported by commercial dollars through large funding by Google.

    Of course I’m not saying that fun or idealism is not a contributor as well–it obviously is (I’m an open source Firefox extension developer myself). But there have to be some alternative models to funding besides the old corporate model–where code is locked away forever and people are locked in forever–and starving programmer syndrome–where programmers who make useful products are not given a dime for making useful stuff, but on the contrary face whining complaints even from those who have not donated anything and who give poor reviews for the stupidest of reasons (thankfully AMO has a system now to challenge such reviews).

    People also often mistakenly think that all developers can make tons of money at their regular work while also having lots of time for free extensions; they think that because something is given to them for free, that it comes at no cost, and could not have been helped along by their participation or donations (or for certain uses, required payment); this is just not true.

    There are public goods, to be sure, and if a society is willing to pay taxes for it, if PBS-style, nagging can do the trick, if idealists happen to find time to work on exactly what you need, or if self-interested corporate Robin Hood battles happen to take place over some turf, you may get your free ride. But it doesn’t mean the process getting there was as smooth or fast as it could have been, nor that better things wouldn’t come otherwise.

    Beyond the welcome pure donations AMO has graciously added, I would like to see a diversity of additional options added over time as well:

    1) Addons which required purchase (ideally allowing developers to specify a demo period), though it should be a distinct category which can be completely filtered out by the user (as I feel addons depending on registration with some external site should be filterable as well).

    This purchase option should be available to developers even if their code didn’t use binaries, or even if they wish to make the code legally open source but also require purchase through the Addons site; even if some people are savvy enough to copy the code or hunt around on the net for a free version, the fact that the site compels them may ensure we developers actually get more than one small donation per 20,000 downloads–whether by people who don’t know it may be available elsewhere or people who do but just go along since they support the product and the developer’s needs, and since they have the added convenience of downloading from the trusted, freshly updating, and conveniently searchable Addons site.

    2) What I tongue-in-cheek call “Bribeware” (though maybe “entrepreneur-ware” could be a more formal name) — a request and offer system to implement specific features, giving financial incentives to both developers and users:
    a) Offer: Allow developers to stipulate a minimum amount they are willing to accept to implement some feature and let trusting contributors pledge to offer when the product is finished (with the developer specifying ahead of time whether he or the contributors will determine when the project is finished, and if it is he, the pledged donation will be compelled at the given time).
    b) Allow user(s) to request a feature for which they are willing to pay a certain amount (and to which others can add to the bounty), with the contributor specifying ahead of time whether he or the developer will determine when the project is finished, and if it is the developer, then the contributor(s) will be compelled at the given time).
    c) A social component among developers as well as users: The developer could choose whether to accept commits or patches (ideally with optional version control at the Addons site) and share profits with other developers based on offer or request conditions by such developers, but to avoid a cut-throat atmosphere on one’s own space, the ability for developers to communicate with another developer’s users shouldn’t be automatic.

    But I fully agree that it should be done in such a way that one can completely filter out paid extensions or demos while looking for free addons, filter out closed source when looking for open source, and filter out external-site-dependent extensions from those which are not.

  11. Brett Zamir wrote on

    Just to add a little more in case anyone doubts the significance of either of my proposals (required payment for open source code or a request-offer system).

    Consider the iTunes model. Many people, including myself, felt relief that we could finally pay a little to get just what we wanted, without being coerced into buying a bunch of songs we didn’t want or being tempted to download or copy illegally. We need such a model for software, including open-source software.

    I believe there are many people who, while still being able to enjoy getting for free many extensions, are smart enough to realize that their contributions–even required–of developers can get them something good and no doubt, in all likelihood, much better–and make them feel good in the process.

  12. Ken Saunders wrote on

    “Firefox only got the momentum that it has because…”

    Actually, Firefox has the success and momentum that it does because of the people behind it which includes Mozilla’s employees, the tens of thousands of volunteers developing and supporting it (for free), add-on developers (working for free), the media (free coverage and reviews), and the countless amounts of people promoting it on the Web and in the real world (also for free). Not because it was based on donated code from a for-profit company.

    I could be wrong, but I think that you’d be hard pressed to find anyone at Mozilla that would say that Firefox (and Mozilla’s other projects and products) would be where it is right now if they (a handful of Mozillians) had gone it alone with a for profit business model in mind.
    There is no way at all that they would have been able to get Firefox’s market share where it is right now and would have posed a viable threat to IE with MS’s seemingly unlimited funds, or made such an impact on the Web itself. No way. They didn’t have the capital.

    Besides, we are talking about two different things.
    Mozilla’s mission isn’t to turn a profit. They need to of course, but non-profits retain their non-profit status by putting the revenue that they take in back into the organization.
    By the way, Google isn’t Mozilla’s only source of revenue. It’s the most significant, but in case you hadn’t heard, Mozilla had been offered a blank check to take Google’s spot. We’ll just have to wait and see what happens in a couple of years. And speaking of Google, and because I’m really sick of people thinking that Mozilla owes its success entirely to Google, do you really think that Google would be where it is right now if it hadn’t been for the instant user base that they’ve gotten and continue to get from Firefox?

    Back to the main point.
    If developers were to get their own demo/pay for section on AMO that they would generate income from, then wouldn’t it only be fair that Mozilla gets a cut of the profits considering that they are providing hosting and free exposure?

    I’m fully aware of the time that it takes developers to write add-ons. I’ve been fortunate enough to have been associated with some who have made some custom add-ons or modification to their own for users with different accessibility needs. I’ve also watched a lot of those add-ons die out and become abandoned with the developer dropping off of the face of the planet because their real world needs and responsibilities needed to take priority.
    I’m a huge advocate and supporter for developers. They (you) certainly deserve compensation, respect, and notoriety. I just don’t see a commercial, pro, whatever you want to call it add-on system/store working on AMO. There wouldn’t be enough users to support it because most simply can’t afford it or won’t pay for add-ons, and it would make Mozilla look bad.
    Starting a branch of Mozilla that would be financially discriminatory towards people who cannot afford to pay for something for Firefox would turn many to most people off.
    Aside from the current tough economic times globally, there are people who have never been able to, and will not ever be able to afford such things.
    Mozilla and open source in general provides options and a balance.

    Sell add-ons. Go for it. I just don’t want to see Mozilla in the business of doing it.

  13. Brett Zamir wrote on

    Hi Ken,

    Thanks for your thoughtful reply.

    A couple things:

    1) I certainly agree Mozilla’s success owes much of its success to its volunteers. No doubt.
    2) I’m also not giving sole credit to Google either.
    3) If AMO supported a demo/pay section (again, including one which could charge for open source extensions as well), they should absolutely be able to get a cut. I have no problem with that at all. Actually, it should be good to find a diverse revenue stream both for sustainability and greater independence/concern for smaller developers.
    4) I’m not talking about Firefox itself being at all paid for, nor its crucial functionality. I’m only talking about a more diverse economic ecosystem around it. I don’t think it’s discriminatory to admit that you can get more things with money–that’s already true, so why not let Mozilla and its contributors–as well as users–benefit from it (and not make entrepreneurship available only to deep pocketed people). I say users because in addition to possibly seeing more options available and higher quality, and not at a high price if that, they would have a means of paying for customizations (which, if the product were open source, could be made available for everyone as well).

    As long as the site can filter out extensions-for-pay, I only see benefit in it, especially with my proposal for being able to make conditional offers or earmarked donations. Contributors get excited when they can get something tangible in return for a contribution.

    Look, if I could get $1 for every other download for my own humble extensions, I’d be a very satisfied guy (who could spend more time developing other extensions, including free ones, which I liked to work on and which I could probably do a better job at). There’s also a certain critical mass in building skills for ext. development which may not be otherwise in demand in the marketplace and thus not reached as high a peak as it could (I’m lucky to have a job building extensions, but it’s not like job lists are overflowing with postings for Firefox extension developers; and the skills I learn at work do not always translate into being useful for my free extensions).

    Even if you want to get 50 extensions installed, at $1 for an extension, $50 is hardly difficult even for relatively poor people (who have enough to access a computer and internet). People pay for CDs, fancy tennis shoes, etc. Even in the developing world, people (men esp.) spend a lot on alcohol, cigarettes, etc. And that’s making a big assumption that everyone will flock to charge for extensions if there is a chance. I also have no doubt that the quality (and maintenance) of extensions would improve if maintainers knew they could get some compensation for it.

    Mozilla developers might not care about this–they are employed and they might think this doesn’t affect them, outside of the fact that such a system could really boost their own support. People want to live in their browsers. That is HUGE potential. Think of all of the developers paid to make websites.

    Let’s not make it a choice between “work for the Man” (to make expensive products for a richer audience) and “work for free”, nor between “pay big for top-notch reliability and personal customization” and “pay nothing to be at the whims of part-time developers who may make cool stuff, but who do not always maintain or improve things for mainstream users”. And if designers and translators could get a cut too, things could even be cooler. I certainly don’t want to see even $20 extensions either, unless it is something like a full-featured and open source IDE or something (yes, I do think people would pay for open source if at a central site and cheap). But I think you’d see way more “hobbyists” build cool things we can only imagine without the full involvement of users through their financial support (whether through donations, earmarked contributions, or compulsory but low pay options).

  14. Brett Zamir wrote on

    One more thing… Maybe the system could even allow people to say whether they were financially needy, informing them that certain paid sites offered discounts or for free, but asking people not to be cheap-skates, as the normal prices for paid extensions were already quite low for those who did not indicate they were needy…

  15. ravi wrote on

    Just to add a little more in case anyone doubts the significance of either of my proposals (required payment for open source code or a request-offer system).

  16. Firefox User wrote on

    I have read almost the entire mozilla website regarding addons, contributions, and this blog. I’ve read that ‘during this pilot period, mozilla won’t be taking a commission on contributions’. Seeing as how you’ve created pledge drives, and other things, all good, I was curious if you’ve changed your stance.

    I found NO information.

    Please comment saying if you are taking cuts, from developers, or from paypal side, and LINK TO SUCH INFORMATION that was available online, publicly, Prior to Dec 09th 2009 – stating if you do take a cut, and if so how much, and where is it detailed – and if this information isn’t publicly available, WHY NOT?

    Also, are there any terms / agreements not publicly available (without signing up to developer forum or other account) that restrict developers talking about the amounts mozilla take, or their revenue.