This article is patently ridiculous. As if the article's central claim was not dubious enough, it opens with this:
> "Linus Torvalds is perhaps best known as the creator of Linux, but he has arguably had a bigger impact as the inventor of Git"
I had to check that this wasn't an April fool's day post. Linux powers a large proportion of servers on the internet, and has done so for a long time now. The popularity of Git is a relatively recent phenomenon.
> "Mozilla has seen better days. There was a time when it was indispensable to web freedom."
It's more indispensable to web freedom now than it has ever been.
> "there was real concern about the web's future with its primary gateway owned by one big, proprietary company"
I guess they've never heard of Google Chrome? A browser run by a company that's far scarier than Microsoft ever was in that period. The implications of a Google Chrome monopoly in the browser market on user freedom and privacy far outweigh the moral implications of Internet Explorer's monopoly. Not to mention the implications for web standards. Microsoft uses Blink in their Edge browser. If Mozilla were to concede in this area, Google would become the de-facto owner of web standards.
Internet Explorer wasn't just morally wrong. It was a performant browser with extremely broken standards support. Web developers had to choose if they want to support the broken IE6 or the much more logical and easier to develop for web standards. It was a painful long time for web devs with lots of lost developer productivity until Firefox and Chrome came to the rescue. The amount of money MS lost for not continuing IE development is staggering though.
This is somewhat disingenuous. IE was trying to strong arm web standards in much the way that chrome does today. Just back then, few folks would have it from ms. That said, div, span, xmlhttprequest... All of those came from ie, if I am not mistaken.
Now, I confess I have grown to not like divs and spans. That said, I am very amenable to "border-box", which is kind of hilarious in historical context.
I also remember IE actually supported fieldset and legend in sane ways well before netscape. Which was very annoying.
Actually, my memory is essentially that the IE team was doing decent work. The Netscape team was faltering. Only reason folks didn't like the IE team is the rest of MS was acting in very bad faith with clear embrace and extinguish as a goal for the browser market. Which ultimately left Opera as the biggest loser who otherwise had an amazing product.
A long time ago I wrote an extensive comment--with tons of documented references--here showing how broken everyone has the story of IE: Microsoft was in fact the savior that came to save us from the world of Netscape... Mozilla was great, and had to save us from the new world of post-antitrust Microsoft (which now had a dominant platform they were too afraid to maintain); but Netscape was just as much of a bully as anyone else :(, and so the W3C was pumped about MS.
(edit:) I just remembered that many years later I gave a talk that contained a follow-up to some of the thoughts about Microsoft (it is a big part of the talk, but not the beginning or the end).
(edit again:) OMG people seem to be enjoying this comment, but I stupidly linked to the wrong old comment up there (and even that link is weird: the best part of that thread is a few parents up from that link)... and it is so sad as the actual comment was one I was so proud of... here it is:
> Actually, my memory is essentially that the IE team was doing decent work.
Yes for a while there they were doing a great job of implementing Microsoft's Embrace, Extend, Extinguish philosophy by intentionally avoiding web standards conformance and instead focusing on proprietary features (ActiveX plugins etc) in attempt to make IE "better" than the competition
This is actually compatible with what I was saying. The ie team was building a decent product. If it was being sold, it would have been a healthy competitor in the field. Instead, the company pushed it for free to basically burn the market.
"It was a performant browser with extremely broken standards support."
Microsoft licensed the source code to NCSA Mosaic in order to prodcuce Internet Explorer. What were the "standards" at the time Mosaic was released.
What never seems to get mentioned in these threads about Chrome and Mozilla is that Google hired a couple of Mozilla Firefox developers to kickstart Chrome.
Perhaps I am wrong but I sometimes think there would be no Chrome without that help from former Mozilla employees.
There are some really insidious errors in this comment. It's close enough to the truth that it runs the risk of being taken at face value, despite the errors.
> Microsoft licensed the source code to NCSA Mosaic
Not quite. They used Spyglass Mosaic, which shared only the name. Spyglass Mosaic was a from-scratch implementation.
> What never seems to get mentioned in these threads about Chrome and Mozilla is that Google hired a couple of Mozilla Firefox developers to kickstart Chrome.
Not quite. Early Firefox development resembled Linux: multiple industry actors collaborating on a single codebase—pretty unlike the way it works today, where a single (bumbling) corporation is almost totally responsible for marketing, product direction, and development. Google was one of the biggest contributors to Firefox development. The Firefox lead was a Google employee being paid to do just that.
When Google decided to do Chrome, they pulled their engineers off Firefox, with parallels to what later happened with the WebKit -> Blink fork: an initial cooperative effort, followed by their chafing at the overheads of collaboration and the lack of total control, followed by breaking away to do their own thing.
If there was any amount of poaching/brain drain that should be mentioned but no one really talks about, it's how many ex-Mozilla Corp folks left for Facebook and is partially responsible for the engineering culture there.
Spyglass, Inc. licensed Mosaic from NCSA. Microsft licensed Mosaic from Spyglass. Microsoft did make some modifications to Spyglass's "from-scratch" implementation in order to produce Internet Explorer.
Sounds like Mozillians are not so concerned about user privacy if they went to work at Facebook.
> Spyglass, Inc. licensed Mosaic from NCSA. Microsft licensed Mosaic from Spyglass. Microsoft did make some modifications to Spyglass's "from-scratch" implementation in order to produce Internet Explorer.
I don't know what point you're trying to make. A reading of your previous comment would lead people to believe that there is some part of NCSA Mosaic lurking under the hood in IE. But IE is based on Spyglass Mosaic, not NCSA Mosaic, which are not the same thing.
The point I was making is that there could be no Internet Explorer without a license to Mosaic, either from NCSA or from Spyglass. Similarly IMO it is arguable there could be no Chrome without Google first having employees work on Firefox.
Firefox as a browser is rapidly becoming irrelevant, and it's not "indispensable to web freedom" when they tear down privacy by implementing the Chrome team's privacy disaster APIs in full such as AudioContext and others that are used almost exclusively by fingerprinting scripts running on DoubleClick's distribution network.
Mozilla is complicit in this behavior, and they cave time and time and time again.
The prime difference of course is that web games actually play audio, whereas most abusers of this never play audio. Why does a site with no audio need this information with no notification or permission from the user?
I also beg to differ, AudioContext has a horrible security model (it's non-existent). Browsers don't even care if the website even needs it.
> It's more indispensable to web freedom now than it has ever been.
"Changing these dangerous dynamics requires more than just the temporary silencing or permanent removal of bad actors from social media platforms. Turn on by default the tools to amplify factual voices over disinformation. "
I was conscious of this when I wrote that. It's a shame that the Mozilla of today serves as the thin line separating us from an effective Google monopoly on web-standards. I know that Brave's userbase has grown recently, however I refuse to see using a Chromium-based browser as a solution to ensuring web freedom.
> I had to check that this wasn't an April fool's day post. Linux powers a large proportion of servers on the internet, and has done so for a long time now. The popularity of Git is a relatively recent phenomenon.
Linux might be used for the majority of web servers, sure, however Git is used for version controlling the significant majority of software produced today. I would argue something that is used by nearly every software development team/company across the entire planet has a much greater impact than something that is used by nearly every server across the planet.
Linux also sees more heavy competition from the likes of Windows and *BSD, where git has essentially dominated it's market with things like CVS, SVN and Mercurial becoming increasingly abandoned, and only niche proprietary systems like Perforce still competing for usage
It's got a big impact, but mercurial was released pretty much the same month as git. Git got more popular, likely helped by Linus already being known. If we didn't get git, we'd use something else without issues. And you can use something else right now. Linux on the other hand just doesn't have an alternative equal in features.
I'm not sure I agree. git has a particularly lucid underlying data model which is the reason why it has endured so long (despite its objectively awful ux), and still sets the bar for cryptographically secure SCMs.
"Heavy competition" is overselling the threat that Linux faces from Windows in the Internet server space. I'd go with something like "Linux roundly dominates, save for some legacy environments that have largely been picked off by cryptolocker breaches over the last several years."
> Linux also sees more heavy competition from the likes of Windows and *BSD,
No it doesn’t. Neither of them are even a consideration in most server scenarios at tech companies. Anecdotally, every company I’ve worked at used Linux in multiple forms (severs, embedded, etc) but not all of them used git (Google is the most well-known example).
Every tech company I've worked at used Linux for servers as well, however that doesn't mean I'm ignorant of the huge number of (predominantly large enterprise) corporations relying on the Windows Server stack. Also I would argue there are significantly more servers, embedded devices and especially networking infrastructure running *BSD than you give it credit for, not to mention every consumer device manufactured by Apple.
How many open source projects or tech companies can you cite using an SCM suite other than Git, except for Google which you already mentioned?
The huge disparity in usage makes it pretty clear Git is the only widely relevant SCM suite today. GitHub and GitLab are arguably the most active and central hubs of open source communities, and they're based on Git. Bitbucket discontinued Mercurial support a while ago. How many popular / impactful public code hosting repositories are primarily using SCMs other than Git? I'd be surprised if you can name a single one other than Google's now discontinued Google Code service.
It depends on your success measure. There are way more Android users for example (most smartphone users), or people who interact with Linux servers (basically every person on the www), than people who use git (developers, a niche profession). The first two categories contain billions of humans, while the last category includes maybe a couple dozen million or so.
> Linux powers a large proportion of servers on the internet, and has done so for a long time now. The popularity of Git is a relatively recent phenomenon.
Isn’t Linux become popular in recent years too? At first they use Unix to run servers, you can even encounter some sites running Unix nowadays.
I kind of agree with the saying “Linus has a bigger impact as the inventor of Git”, because we can use git for Windows/Mac OS/BSD software, while Linux can only run Linux Apps.
It's just silly. Get outside your desktop bubble. Linux powers literally billions of smartphones, a similar amount of embedded devices, millions of servers in large datacenters, etc etc.
git is one VCS among many, even if it is the most popular and the best in many respects. It's one of the many tools used by pretty much a single profession. To me it's a bit like saying that a circular saw has had more impact on the world than Linux.
Rust and Firefox are important in very different ways. Firefox is still critically important to the Web as a standards-based platform, which is really important to stop client applications collapsing into Google/Apple walled gardens. People just care less about the latter than they should (and did).
If you look at where they invest their resources, it's Firefox.
It's almost funny because Mozilla critics are about evenly divided between "Mozilla needs to move beyond the browser" and "Mozilla needs to focus on the browser". And both groups are adamant that their view is obviously correct.
I think not. The state of Mozilla is mostly the result of decisions they made in the past that are no longer operative, and factors that were never under Mozilla's control.
Mozilla bears a lot more responsibility (including ongoing responsibility) for its self-inflicted wounds than this common, comfortable narrative says it does.
This. I was so sad to find out that despite having a phenomenal CSS support, FF does not (easily) allow to switch off CORS, which makes it impossible for me to use at work.
Less than two months ago NASA landed a rover on Mars which is powered by Linux. I struggle to see how git (as important as it has been), can really do better than that.
And sure, Rust is great. It's lowered the barrier-of-entry to systems-level programming significantly compared to the alternatives. But, comparing it to Netscape/Firefox, and coming to the conclusion that Rust is Mozilla's greatest contribution is ridiculous.
In a time where Microsoft was much more hostile to the tech industry and open-source in general, the landscape of web-browsers was an absolute mess thanks to Microsofts insistence on developing their own way of doing things contrary to accepted standards.
Without Firefox and the work Mozilla did pushing back against Microsoft, there would be no Chrome. And while this may be a bold claim, there may have never been a NodeJS and the Javascript renaissance that the industry has seen over the last ten or so years.
From my perspective, when Chrome burst onto the scene and gave us Firefox vs Chrome, it did feel a little bit like Microsoft vs Mozilla, except this time the monolithic corporation won, and we're all still sitting here wondering if we are better or worse off for it.
Maybe I'm old or maybe I just don't get it, but this seems like we're rushing to throw away historical contributions because they aren't as cool and popular as the hot new stuff.
Firefox is less than what I think that it could or should be - but it is the last shield against the wholesale takeover of the primary face of the internet by an advertising monopoly.
The thing that scares me most about mozilla is the question of which long term revenue generation mechanism best guarantees their ability to continue to develop and mature software the allows for a user focused internet software ecosystem (which WASM is forming of the web).
Sorry for the lack of punctuation there, but I couldn't figure where the fuck the commas should go.
Rust may be Mozilla's greatest _industry_ contibution, but for society I think it's not even close.
Anyhow ¯\_(°ペ)_/¯ Rust is a great example of good engineering. There is a lot of old ideas put together in a way that makes the whole greater than the sum of its parts.
The title litterally means "Rust, not Firefox, is Mozilla's greatest industry contribution", like noone is talking about society contribution.
I don't understand what are you talking about.
> "Linus Torvalds is perhaps best known as the creator of Linux, but he has arguably had a bigger impact as the inventor of Git"
I had to check that this wasn't an April fool's day post. Linux powers a large proportion of servers on the internet, and has done so for a long time now. The popularity of Git is a relatively recent phenomenon.
> "Mozilla has seen better days. There was a time when it was indispensable to web freedom."
It's more indispensable to web freedom now than it has ever been.
> "there was real concern about the web's future with its primary gateway owned by one big, proprietary company"
I guess they've never heard of Google Chrome? A browser run by a company that's far scarier than Microsoft ever was in that period. The implications of a Google Chrome monopoly in the browser market on user freedom and privacy far outweigh the moral implications of Internet Explorer's monopoly. Not to mention the implications for web standards. Microsoft uses Blink in their Edge browser. If Mozilla were to concede in this area, Google would become the de-facto owner of web standards.