2 ms·
> Just about all of the major browsers earned their reputation for being blazing fast, tiny, and using little memory. Not like Opera though who made performanc
by webmobdev 6y ago
> Just about all of the major browsers earned their reputation for being blazing fast, tiny, and using little memory.
Not like Opera though who made performance and small size (and power optimizations) as their core philosophy.
(This was the browser whose popularity even frightened Microsoft, whose infamous Internet Explorer had I think nearly 80% market share at the time. So much that they deliberately ensured that their website broke on Opera. And it forced them to reluctantly focus on the Trident engine that powered IE again - till then, they had been ignoring IE because Microsoft was rightly genuinely frightened of competing on and with the web).
Starting with the mindset of making a fast, small and efficient engine that could run on constrained systems was the reason Presto outshined other engines. Later this extended to optimising their code for lower power consumption too.
That's the main difference with Gecko, Chromium / Blink and Webkit (to a lesser extent) who did focus on optimising for speed but were a bit more lax on the memory and power aspect and only started addressing it after users started complaining (and ofcourse, only after they made mobile platform a priority). I still remember how we users grumbled when Opera installers crossed the 10 MB size :) ... but Opera still managed to keep us hooked by introducing a plethora of features while still trying their best to maintain an efficient code base and deliver it in a smaller package. Compared to it, equivalent versions of Firefox and other browsers at that time (and even today) seem very bloated in comparison because it's not a priority to them.
> Then real-world pressures kick in -- it turns out that although people love small and fast, that doesn't do them much good without support for X (or being able to handle condition Y), and there are many, many values of X (and Y).
Sure, but this doesn't apply to Opera. Opera kept their high standard and core philosophy intact till the last version before management failed them. It maintained feature parity with competing browsers till the end, before it was sold off and dumped for Blink to cut costs.
Opera didn't fail because it couldn't keep up with web standards or in introducing new features, but because of poor business management.
If Opera's founder is to be believed, they just couldn't allocate enough resources for keeping track of, and patching broken popular sites that don't follow web standards. (This is the reason he gave for abandoning Presto, and selling his company, and then choosing Blink for his new Vivaldi browser).
Opera the business failed, but their product was top notch and better than their competitors till the end. If I remember right, I stuck with it till 2017-18 before I moved on to Firefox and later to Ice Cat, Pale Moon and Safari.
> We [I work for Mozilla] end up mainly competing on different metrics, assuming we all keep up with "good enough" speed and size.
Yeah, that's spot on - that's what I've always felt about Firefox, that Mozilla's philosophy is to be "good enough", and that's why to us consumers the product (Firefox) feels "average". There is nothing much to feel excited about in terms of performance or features. That's the area (performance and features) that Opera excelled in.
> And yes, it was a deliberate decision, but not for the profit motivations you're suggesting.
I am sorry but I don't believe this - it was very much for profit too. I do believe it was to ensure the project isn't forked easily to build a competing product. Just look at Goanna / Pale Moon struggling to maintain quality and feature parity now. (Ofcourse they are also struggling because of the design decisions they made).
Chromeless Firefox was a very interesting project. Killed before it could become popular. Vivaldi browser is based on a similar idea and its enjoying some success.
My point is that it is a smart decision from a purely business perspective - look at all the relatively successful fork of the Chromium browser. And browsers based on webkit. There are many people who don't like the current direction of Mozilla and / or Firefox and they would gladly jump on to decent a fork if it were easy to do so.
> It was based on maintainability and developer velocity, when we realized we were not going to out-compete Google with engineer headcount. We had to simplify and reduce our nonessential complexity overhead.
That's MBA manager speak - lots of buzzwords that don't really convey anything meaningful. ;)
My perspective is that as an open source project, Mozilla shouldn't sacrifice product quality for short-term gains. They should fix bugs rather than focusing on new features or products even if they lose ground to other browsers. Their focus should be on quality rather than trying to compete with a corporate product and keep pushing release after release just for the sake of competing and "staying relevant".
> Why throw so much shade? ... we have very little to gain from keeping any technical or architectural stuff private, and quite a bit to lose.
That is exactly why I found it interesting that an open source project was asking volunteers to sign NDAs! (https://medium.com/@smsnobin77/firefox-build-macos-f1f53d643e81 https://medium.com/@smsnobin77/firefox-build-macos-f1f53d643...)
As for the snark against Mozilla - your perception is right. I do have a certain level of antagonism against the foundation. It stems from the following reasons:
(1) They are a million+ dollar foundation and still produce an "average" product. Comparing them to other open source projects with much, much smaller budgets that produce high quality softwares really irks me.
(2) Some of their recent management decision range between foolish to greedy. E.g. Forcing Pocket on unsuspecting users (I strongly suspect some members where bribed to push this integration). E.g. Including closed source DRM in the browser (if users want to download a plug-in to watch DRM protected media, then they should have a choice to do so. Google and Cisco, if I remember right, now have a monopoly on this because of Mozilla).
- smnthermes 6y ago> Just look at Goanna / Pale Moon struggling to maintain quality and feature parity now. (Ofcourse they are also struggling because of the design decisions they made What design choices? If they would ride on Firefox release cycle, Pale Moon would lose its advantages.
- webmobdev 6y agoTrue. I meant that some of the struggles that they are facing in keeping parity with current Gecko development is because of the goals they have. I use Pale Moon sometime but it's performance on mac leaves a lot to be desired. I also worry about security issues on it ...
- fabrice_d 6y agoAbout the NDA part: some community members have the possibility to sign an NDA and get access to some Mozilla Corp information before it is public. This is absolutely not required to contribute to the code base. About DRM: Mozilla fought against it, and caved when it became clear they could not win. But Firefox users still have to opt-in to install the binary blob (eg. Widewine), it's not bundled in the browser.
- webmobdev 6y ago> About DRM: Mozilla fought against it, and caved when it became clear they could not win. While I do believe there was genuine opposition to this move within the community, I think the real decision makers had already made up their minds and just made a show of "discussing" it. I feel they did it under pressure, and perhaps were even induced with promises of more revenue, from Google. And where was the question of really caving? All they had to do was ask Google (or any other company) to develop a browser plugin for Firefox and let the user decide if they wanted it (when a user encountered it on a site, produce a pop-up alerting the user to the plug-in required). And maybe then we would have some real competition in this space too as a new plug-in war would have developed ... > But Firefox users still have to opt-in to install the binary blob (eg. Widewine), it's not bundled in the browser. That's not true - it's been opt-in by default for a long time. See here - https://imgur.com/a/iCTmnzN https://imgur.com/a/iCTmnzN ... I think since EME free versions were released, at some point a policy decision was taken to make it "opt out" in the usual general releases, as a technical person would rather opt for the EME Free version or go through the options and disable it if they so desired, where as a non-technical user might get confused about EME or DRM or plug-ins ... I fear that the Google DRM binary blob has actually made it much easier for them to uniquely fingerprint every Firefox user with EME / Widevine despite all the ad-blockers and anti-fingerprint extensions or techniques being used. In other words, I fear that Firefox (knowingly or unwittingly) has aided Google in killing privacy on the web.