09.21.20
Gemini version available ♊︎Guest Post: The Worrying State of Political Judgement in Free Software Communities
Original version in Spanish, to be found here
Summary: A look at what Mozilla has become and what that teaches us about the Web and about software
month ago, David Teller published a blog entry explaining in great detail the process behind the controversial XUL+XPCOM elimination from Firefox upon its “57″ version being released.
That decision, as I was saying, was controversial because it implied losing the huge add-on ecosystem Firefox had: one of the main reasons for using Firefox in the first place, instead of just using Google Chrome. And it was so controversial that, in fact, there are still people who are angry about it three years later, even though these people just stopped using Firefox back then. In my echo chamber, everybody saw the move as a shot in Mozilla’s own foot. And the reasons justifying the move were the ones we’re getting used to in contemporary informatics: “speed”, “security”, and “what users want”.
That last part may be a bit unfair, because a good chunk of the justifications were about the many difficulties Mozilla had to face in order to continue to sustain Firefox development. But my point is that those difficulties will be there no matter what they choose, and that’s why I’ve excluded development from that list. Maybe there’s a whole other debate to be found in all this, but it is not the one I’m interested in right now, and so I just let the issue be dealt with here.
“That looks a lot like what started to happen in the late ’90s with the Windows ecosystem.”Teller’s post tells us stuff about historical details of XUL+XPCOM, things that happened around the Web, and the problems Mozilla has faced while sustaining Firefox, in the face of a migration towards another system as was widely decided. His post is excellent and definitely a recommended read — so much is like this, that it became a somewhat popular focal point and an object of debate, getting to have a comment section at least as interesting as the post itself. Now, for a few weeks I wanted to write about it…
I’ll get right to the point: Teller wrote about “competing with Chrome” in his post, and different people argued about that. It got up to a point that Teller ended up editing the article (with an edition note at the end), replacing the “competing with Chrome” phrase with “as fast, as stable and as secure as Chrome”. And I believe this detail is in the core of a very general problem.
One can see, for example, that the first comment available is from Jeremy Andrews, who maintains Pale Moon in Solaris. And Andrews argues against this idea of “competing with Chrome”, on “why would somebody do something other than competing with Chrome”. He says the following:
(…) I’m doing it for the people that have been left behind since about 2007 when the iPhone and Facebook changed the world. The people that still mostly use their desktop PCs and like being able to tweak or customize everything. People who largely feel that they’re being asked to accept that the freedom and choice of the early Internet is being phased out in favor of security, top-down decision making, centralization, and lack of real choices. (…)
Daniel Eriksson then responds this other commentary:
(…) Up until 2010 I was always excited about new technology since it always gave me new possibilities and made it possible to do more in a way that suited me, and then that changed. Now I worry about new releases of software, fearing what useful function or option might have been taken away this time. (…)
That looks a lot like what started to happen in the late ’90s with the Windows ecosystem. Who didn’t start to keep old installers from previous versions, or even portable versions (which back then was just copying the install directory), because new versions were a problem in many ways? I must still have (somewhere) some CDs with a Winamp 2 configured as I pleased, because newer versions were crashy and asked for extra resources. And this practice (keeping old versions) entered in times of crisis when a thousand bits and stuff suddenly had online installers and dependencies, and thus version checks or even protocol changes made it stop working, leaving no other choice but to install newer versions. And this kind of stuff is happening in Free Software today: with “user-friendly” Gnome using UIs (or even program names) that are every day becoming more dumbed-down while breaking old stuff, hellish dependencies at the heart of entire systems (I’m looking at you systemd), QT going private, x86 being deprecated as if it were no longer in use anywhere… and, of course, an obese World Wide Web that no longer allows you to read a simple blog entry in a netbook given so much JavaScript for notifications and tracking. Not a good path.
“Let’s please keep aside for a while the whole privacy issue in this, as it’s absolutely secondary to my argument: the problem here is political and epistemic.”But the “competing with Chrome” mention raised a debate full of exchanges that I recommend you check out. I’m interested in the justification behind “competing with Chrome”: telemetry. I’m not sure if that’s a technical or a commercial name for the thing, but it means “data from the users”. Let’s please keep aside for a while the whole privacy issue in this, as it’s absolutely secondary to my argument: the problem here is political and epistemic.
Epistemic, because all evaluation criteria for reality is being contrasted against data clouds that replace it. Political, because I suspect this is at the core of all the bad changes in last few years in software communities in general, and Free software in particular.
As a programmer, and as a person of science, I understand the value of data. But as an activist, and as a person of art, I also understand its limits. Data is just a single ingredient of several ingredients needed when constructing a map of reality. The others need to be taken at least with the same priority as data. Today, in every aspect of technology, and even in sciences as well, we seem to be living within some spiral and recursive tendency between the act of compiling data and generating any idea of possible future around it. And this is seductive, not just because data is powerful and a thing of our times, but also because it heals or even resurrects our devastated modern search of objectivity: that security which only some unquestionably legit truth can give in order to guide us across the ocean of uncertainties that is the future. Data brings access to a very peculiar way of truth: Aletheia.
“What does Mozilla need telemetry for? And looking at this from the “competing with Chrome” perspective, much more than “what users want”, what this question seems to bring about is, “Mozilla wants to behave like Google”.”But it’s a mirage that previous generations already faced, at the same time they rediscovered how possible worlds and possible futures are also linked with hopes, ethics, and principles, that can and sometimes should take a distance from data instead of embracing what it says. And here’s where politics and art have some lessons to teach.
There are the bias issues. Yet, very frequently bias is evaluated as a defect in judgment, which has the consequence of getting away from objectivity; and I’m more likely trying to vindicate it. Biases do exist, and even when it’s true that biases have palliatives, they don’t translate data into full objectivity either. When we add to that the detail that software development is not necessarily a science, there’s room for the question of what’s the deal with the presentation of objectivity in data, or even data itself.
What does Mozilla do? With its software, with its users, with the data…
What does Mozilla need telemetry for? And looking at this from the “competing with Chrome” perspective, much more than “what users want”, what this question seems to bring about is, “Mozilla wants to behave like Google”. And this is troubling. But not because of “data privacy” (the boogeyman of contemporary progressives in informatics), but because it’s politically aberrant for Mozilla’s history.
Mozilla is supposed to be a non-profit foundation, that used to be a champion of a free web and a great empowerer of users. Mozilla was the one rising up and fighting against Microsoft on the Web, achieving what was the utterly unlikely outcome back then — a triumph that is invaluable today: making the Web not Microsoft-centered. Firefox fought against Internet Explorer in an endurance battle for an entire decade, until Apple and Google entered the ring and finished any hope of Microsoft controlling the Web, for good. And also, let us remember the days when all of us had to make our web pages IE6-compatible, when banks and state offices were forcing us to use IE to access their systems, when a good chunk of the Internet didn’t work without Windows-dependant plugins, or how close we were to get that horrible ending, how lunatic it sounded for a state or any other future. We owe to Mozilla our eternal gratitude and respect for having battled that battle in the way it did. Damn… we should have epic songs about it, for future generations to remember.
For a time (that lasted years), Google recommended Mozilla Firefox for their Web sites, un-recommending that way Internet Explorer. Eventually Google Chrome appeared, based on Safari’s code, for only years later having its own engine/code. Google then started to centralise more and more its Web operations around Chrome — to the point where today it is the de-facto new IE (and, not-so-ironically, even the renamed IE now has Google’s code). But today, unlike back then against Microsoft, Mozilla seems to want to follow Google’s steps instead of fighting it: it takes their browser as a reference instead of having a critical stance. And in the middle of that scene is… data: something that back then didn’t exist, at least not in the way we know it today.
And the thing is, data effectively indicates that users prefer Google Chrome. But that happens in the same way that data from 20 years ago would have suggested the same about IE. Most likely it would also tell us that IE was faster than Firefox here and there (like starting up, as it was integrated well within the OS), or even that the infinite technical problems IE had could not matter less to people (given that IE was used for a much longer time than it should have been tolerated by anyone, and not precisely because of “user experience” or any other metric like that). I’m sure many of us had friends who used the web without ad blockers, and thus they experienced that as the only way there was to use the web, and of course they do that with Google Chrome. And, yeah, I know that counts as “data”. But it’s kinda pointless for what we want from the Web, isn’t it?
However, my problem is not that there can be biases in the interpretation of data: my problem is the actual interpretation Mozilla chooses. Because 20 years ago Firefox would have interpreted “we need to do something against Google Chrome, or it will swallow the whole web in its culture otherwise”; and today it reads like, “we need to be like Google Chrome”.
And here’s where we need to look at Mozilla from some safe distance. Because that blog entry from Teller was posted on the same week Mozilla was laying off hundreds of people. And this post of mine is being written the same week Mozilla cancels yet another couple (two) of services. And this, of course, is deeply related to Teller’s mentions of development and maintenance costs, as stated in his post.
Mozilla is behaving much more like a for-profit business rather than a non-profit foundation. It looks at data in the same way any other enterprise: looking for revenue. It’s sharing the same biases private business do, because it is following the line of “where and how” to make money; and that logic always points to hegemonies to the detriment of minorities (with the notary exception of economic elites).
“Mozilla is behaving much more like a for-profit business rather than a non-profit foundation. It looks at data in the same way any other enterprise: looking for revenue.”And the thing is, the problem is real. This financing problems, and the costs of operation once reaching a certain scale, are not problems experienced exclusively by Mozilla. It’s the same thing that leads Canonical to making deals with Microsoft, then Red Hat being sold to IBM, and the deplorable current state of the Linux Foundation. All free software communities are — year by year — more besieged by financing needs, as the direct result of their operations’ scale/growth. This is because Free software had indeed won many battles, maybe even entire wars, and this is the cost of those triumphs: this is the logic of centrality in capitalism, that free software communities in general don’t seem to be facing as a crucial issue.
However, I expect more from Mozilla, and there’s this point where I start to seriously worry. Because Mozilla is a political reference, and clearly it doesn’t seem to find a way around this. Let me take a deviation for a few seconds, so I can put it in terms of concrete examples.
I know nobody who turns their cable or ADSL modem off during nighttime, or ever. That means that, even when it’s not being actively used, unlike in the dial-up times, our houses are all day long connected to the Internet. That means they’re basically a little potential datacentre. What stops us from serving contents from our very houses? Technologically, the changes needed are almost silly; the real limitation is entirely cultural. And that culture leads us to that; any common person, or even advanced user, today can’t have a clear idea of where to go if they wished to have their own Web site somewhere. In the same way, it’s frequently argued that the “Internet isn’t free” and that it has maintenance/infrastructure costs for being like we know it: fast, 24/7 online, accessible from all around the world, etc. Yet, why does the Internet has to be like that? Why can’t there be Web sites that work just under/during certain hour ranges, like any other human operation that precisely gets cut in time periods in order so save costs and to respect sane work conditions? Why couldn’t my personal Web site work only when I say so, from time x to time y, which is basically when I turn on and off my personal computer inside my house? Why should I guarantee that somebody from Hong Kong or Norway or Ethiopia can connect to my Web site, if I couldn’t care less about the ability to make it happen? Why can’t my Web site be available only to a certain local community that I choose, while having the option of also hosting my site internationally on Amazon or Google or whatever? And why does everything have to be fast? What’s the big deal with waiting for 30 seconds or even a minute in order to access some content, if the important thing there is the content (and not its speed)?
More very basic ideas: the Web is obese, let’s make it lose weight. Why can’t there be other mainstream hypertext languages, more text and styling focused (instead of “structure”) and less complicated to maintain for browser makers? Why not wind up having as a parameter “it must work well in third world countries” or “in 15 years old hardware”, rather than being all the time behind every silly novelty for-profit business make? Isn’t it true that out there exist millions and millions of people in need of stuff like that, given that the thing now seems to be to looking for “data” and “markets”?
“Mozilla’s case is representative and symptomatic: it’s conceiving the Web as an space for market before culture; or even worse, reducing culture to market.”These are fast, accessible, almost silly questions, with very simple answers, which give different ideas corresponding to many different futures for the Web. Apply those same questions to our messaging services, e.g. “who compiles which data through which means and towards what ends”, or “how do we get informed about what stuff, et cetera.” Anybody can think of stuff like this. However, Mozilla, one of our champions of old in the defense and creation of a user-oriented Web, an organization that could be working on stuff like this and easily get results in the shortest time, today puts its efforts into trying to do what Google does. And that’s in a big way just because it has bills to pay. But also there’s a lot of it because of the people that form the Mozilla teams, and the people that form our communities; because in the last 20 years lots of people got inside, and even lots of younger people from a whole different generation. That means people with very different cultural and political formation, which not only generates dissent and needs but also shifts from original or foundational principles. From there, things start to operate with lots of human factors, but it’s also a strong vector of corporate cultural influence. That way, we suddenly have lots of people thinking that Microsoft “is no longer evil”, that increasing speed in things is a need, that we must face politics with a frenetic impetus that leads to very little space for critical reflection (and thus things from ultrapolarization to the RMS cancellation), and that frequently confuses or conflates “novelty” with “progress”.
Mozilla’s case is representative and symptomatic: it’s conceiving the Web as an space for market before culture; or even worse, reducing culture to market. When one does that, what such bias cuts off are many absolutely crucial facts to think about when envisioning a better future Internet, and even better informatics in general. Facts like that are important to many, many of us, who do our stuff without any lucrative spirit, “for free” (as in both free beer and freedom), and that’s most likely a very good and large chunk of current informatics which work well because of that. Facts like “profit logic” are not the only a human logic, and there’s room-full of people all around the world very eager to work on thousands of initiatives, if the conditions are adequate, and without meaning important costs for Mozilla; and that logic, which is very closely related to Free Software history, looks much more closely into the works of activists and artists rather than “producers” or “employees”. Facts like that include the observation that the Internet from 20 years ago was very different, and today we have other big players involved: today, access to the Internet is legally considered a right all around the world. Why doesn’t Mozilla focus on working closer to or more closely with nation states as a revenue source, while working on lots of initiatives related to technology rights, and generating “data” from that other perspective? Apply that to Latin America (where I’m from) and we’re talking about hundreds of millions of people (not at all a small “market”) who also need “solutions” (and not in a commercial sense), and moreover don’t have the same problems as people from the US (which I guess may be the main origin of all the data Mozilla gets from telemetry). Wasn’t the Internet an international thing? Right? Then why are Internet not-for-profit organizations behaving like the whole world needs another Google? Why not even reach out to the UN in order to get funding, in exchange for work on human rights initiatives related to the Internet? The Internet is a thing of relevance to the UN since decades ago (until now), and Mozilla has a curriculum to show off.
From that point of view, the very real financial need looks more like an excuse, and the problem is the political path they’re taking much more ferociously, even before the financing factor. And this is a critique that also applies to any Free Software community. The “as in free beer” is not just a clarification, nor a joke: wherever the money comes from is a big problem, and a political one. Because if we’re slaves to money, when our software gets into the news, then the next step is to become another monster; “not-as-in-free-beer” looks like a very shy way of saying “for-profit”. And this thing will keep on happening again and again and again, until we as the community face the very core problem of our relation with capitalism itself, and perhaps the question of what our stance on it actually is. I believe this is part of the crisis Free Software is dealing with right now.
“…any financing or initiative evaluation has to come with political principles as parameters.”But there’s room for a clarification here, regarding Mozilla. Asking Mozilla to come to Argentina to fix our informatics problems is unfair and absolutely out of place: it should be Argentinian groups — the ones reaching out to Mozilla if they want that to happen. Yet, that “competing with Chrome” impetus from Teller’s article gets Mozilla very far away from any possibility of dialogue with any actor other than an economic leviathan: because that’s against what it’s pretending to compare itself. When the enemy was Internet Explorer, even when it’s true that Firefox worked notably better, that wasn’t the reason all of us Mozilla promoters defended it for, but its role in a better future for the web. Today, when I CAN’T honestly say something like “Chrome works better than Firefox” (as “better” is a much different concept than “some animations are smoother”), it seems that “working better” is the only metric to look at. And that is not the case. That’s wrong, actually.
The thing happening with Mozilla then is something to worry about, because it looks to me like the same as what happened to other references from older ages. And this is something that has a solution in politics rather than in software or in money. Our communities need referents, with a clear political vision, showing the way: any financing or initiative evaluation has to come with political principles as parameters. And this is especially needed in order to guide all the youngsters wanting to be a part of their generational changes: an absolutely necessary guide if we don’t want things like the cancellation of Stallman to happen again in some other way (that brutal disinformation campaign from enemies of free software was successful and effective among younger people, and we didn’t had a strong and sound response from Free Software referents in defense of RMS). Our communities and political organizations just CANNOT be SO sensible to corporate influence, and while that keeps happening there’s no debate about any software or any “data” that could protect us from the next corporate operation against our rights. We need guiding words for organizing resistance, much more quickly (faster) and much ahead or before our need for financing.
As a closing note, just as an observation, I believe it is pretty much graphical why I’m writing this. I read that Teller’s post during my lunchtime break on a working day, and I tried to write a quick response in the comments section. Then I happened to realise that the blog had Medium as their comments technology; and I happen to have a Medium account that I actually think I created to answer another Mozilla employee’s blog post, only some years ago but didn’t remember the password. So I asked for a password reset, and hoped to let the thing work for another time (when I get the password/access back). Two days later, after several tries, I still didn’t have my password, so I gave up and went to use a third party ID service: Medium offered Twitter and Google, as well as several other options. I also have a Twitter account that I never use, so I choose that; but at login time, Twitter told me that Medium “needed” to access some private data of mine — stuff that I don’t remember right now (or cannot recall exactly what was it), but I do remember it was scandalous: something like “my private messages”, or “my contacts list”, or stuff like that, which in no way I would have allowed. So I went back to log in with a Google account, which I also almost never use, and this time Google offered me two links about privacy policies and data collection that I frankly just ignored while feeling defeat. By that time, Teller’s post was already edited — specifically the the part that I wanted to comment on (the one about “competing with Chrome”), and so I had to change my comment before posting it. But even then something else happened: the next day I went to check if anybody answered my comment, but the comment just wasn’t there. I published it, and it wasn’t anything rude so I don’t think it got moderated, so it had to be shadowbanned in some way: I didn’t bother to log in again and check it out.
So, here’s my point: if for having a dialog with somebody from Mozilla I have to enter into a blog hosted in GitHub (Microsoft), allowing a third party to access my data (Medium), to even having to log in to that third party system using another third party credentials (Twitter or Google), and even then end up censored somehow… if Mozilla’s people don’t see a problem there, or ever tried to say something about “what people do” or “what data tells” in front of that… then I’m afraid we should be very worried about the state of political judgement in our political organizations. █
ᶃ Gemini Space
This post is also available in 
Content is available under CC-BY-SA