There is a lot of "shift-left" emphasis out there, but ultimate their is a conflict of interest problem. I think there's no need to go beyond the "we are losing money" explanation and QA is considered more of a cost center.
Heck, the burden can be shifted to users! But then you might not have so many users in the future. :(
Very low engagement out of QA. I'm not sure if they just have a way they like to work, or they think that if they exist then no bugs make it to prod (don't laugh, plenty of our twitter adverts get comments from folks thinking you don't need any monitoring if you have a tester on the team or if a dev writes unit tests).
Understand how the end user experiences your software. That's the actual source of the truth.
This in itself is a not a problem, coming from an oncall minded company. You take care of code you write and you're given time budget to KTFLO.
It was a point of nerd pride on Usenet how well your operating system could handle a Netscape crash back in the day.
"...de-emphasizing QA departments..."
I feel like we (the industry) have forfeited. I did a stint as a SQA manager. Coming from a dev background, I took it very seriously, totally immersed myself in the domain, transmuted from skeptical to true believer.
I honestly don't know what to make of today's state of affairs.
For example, today's business analysts often do many of the tasks we used to associate with the QA role. Testing, verification, liaison with customers. Did we just rename the role?
Did "Agile" smother QA? Until very recently, I've never heard an Agile explanation for how to do QA/Test. I mean really do it, not just wave your arms. The "Test Into Prod" thesis, strategy, whatever, is the first intellectually honest, actionable, constructive (criticizable) methodology I've seen which is tailored for our new market realties.
I've never understood Agile. My teams were way more "nimble" (to use a different adjective) using PMI, critical path, iterative, lightest weight decision making, front loading work, managing risk, and so forth. All the battle hardened time proven stuff people untrained in project management pejoratively call "waterfall".
One correct criticism of all failing methodologists, including Agile, is lack of feedback loops. The "throwing over the wall" of work downstream. We designed feedback loops into our processes, some of what today would be called CI/CD. We were definitely not waterfall. (Another is managing transaction costs, something Agile has manifestly failed to do.)
Rant over. Sorry. Now get off my lawn.
The person asking for a features should absolutely be the one signing off on the feature.
A fuzzer is not going to replace unit tests or good SDLC, which often involves QA.
If there's anything they're going to be in the history books for, it's going to say Rust, not Firefox.
Mozilla is very good at making browsers. Making browsers is incredibly hard and it's amazing that Firefox is competitive with Chrome given a fraction of the development resources Chrome has.
Mozilla has made some big mistakes, but so have the other browser vendors. It's easier to brush over your mistakes when you have an ocean of resources and market power.
Given that the embedding API would have had to change dramatically due to e10s pretty soon after 2010 (and maybe again due to Fission), I think 2010 was probably a bad time to promote a stable embedding API.
Embedding doesn't rank high on my list of Mozilla's mistakes.
Elsewhere in these comments people have estimated that Google pays at least 1000 people to work on Chrome. That's about the size of all of Mozilla, and a lot of those Mozilla staff are necessarily not working directly on Firefox --- you need HR, accountants, marketing, etc. Also, Google pays its developers significantly more than Mozilla does, on average; Mozilla developers tend to get big raises when they move to Google.
And that's just direct spending. Historically Google has done a lot of Chrome marketing on its Web sites, which is prime advertising real estate that would cost astronomical amounts of money if it was for sale. And historically Google has paid hardware and software vendors to preinstall Chrome, which is also expensive, though I'm not sure how much that happens these days.
One contribution we can always use is people writing and talking about rr. One of the biggest things holding rr back is so many people just don't know it exists (or they know it exists but they don't appreciate what it could do for them).
Quality and popularity are rarely correlated, but such metrics are even further confounded when you have monopolistic factions like Google and MS pumping huge money into user acquisition for their competing platforms. It's ridiculous to imply that Firefox is being rejected by users just because Google has been dumping millions into getting people switched to Chrome (and doing so not only with marketing dollars, but increasingly with bully tactics).
I would love someone to estimate an actual figure for total Chrome marketing spend including in-kind. I'm guessing it would be well over $10B.
I tried to use Firefox recently. It leaked 28 GiB of RAM on x86_64 GNU/Linux with no extensions except uBlock Origin. Happened a few times over the month whenever I visited JS-heavy websites. Never had that happen with Chromium, which runs through megs of JS like butter.
Wouldn't it be nice if an experienced browser dev team maintained a privacy-oriented libre version of Chrome (without manifest v3, sync, and all that trash). Or should they keep doing what they've doing and make the best pro-privacy browser that no one ever uses except indirectly through Tor Browser.
And further contribute to the browser monoculture?
> I tried to use Firefox recently. It leaked 28 GiB of RAM on x86_64 GNU/Linux with no extensions except uBlock Origin. Happened a few times over the month whenever I visited JS-heavy websites.
And you've reported this, I assume?
> Never had that happen with Chromium, which runs through megs of JS like butter.
Really? I explicitly avoid Chrome on my computer because it can't handle the web without chewing through my RAM.
> Wouldn't it be nice if an experienced browser dev team maintained a privacy-oriented libre version of Chrome (without manifest v3, sync, and all that trash). Or should they keep doing what they've doing and make the best pro-privacy browser that no one ever uses except indirectly through Tor Browser.
I would like the experienced Mozilla team to continue to work on their pro-privacy browser than a decent number of people use.
This is a total non-response. They're not actually under an obligation to report a bug and use their time - Mozilla, however, is obliged to make their browser work properly, and to ensure the team that's working on it is properly staffed/resourced/competent.
Your response is downplaying someone's actual concern by acting like it's wrong that they don't spend their spare time participating in open source software development.
Furthermore, this "browser monoculture" argument is ridiculous. WebKit and Blink are both open source, and Mozilla is increasingly becoming what Opera was back with Presto: a lone engine with quirks that nobody wants to waste their time working around.
If Mozilla gave up Gecko tomorrow and forked WebKit (a la what Blink is) I don't think I'd bat an eye. This is like what junior programmers wind up learning at some point - nobody cares what the code looks like, just that it does what it's expected to.
I say this all as someone who's contributed to some Mozilla repos, has a massive personal investment in Rust (and has read a good chunk of Servo, and liked it), and has been a fan since the Mozilla Suite days.
Anything that broken would be fixed, almost instantly, had they reported it. I am not very amenable to people who discuss their very extreme personal anecdotes that they have failed to even put even the basic (some may even say courteous) amount of help and instead talk about their experience like it's typical.
> Furthermore, this "browser monoculture" argument is ridiculous. WebKit and Blink are both open source, and Mozilla is increasingly becoming what Opera was back with Presto: a lone engine with quirks that nobody wants to waste their time working around.
Open source does not ensure that a monoculture will not develop; both WebKit and Blink are financed by billion (trillion?) dollar corporations that have almost complete control over what will or will not be worked on or merged in. Every engine has its own quirks and slant on how it interprets the web standard: if WebKit was the dominant browser today we'd all be even more hesitant to roll out WebM or WebGL 2.
> If Mozilla gave up Gecko tomorrow and forked WebKit (a la what Blink is) I don't think I'd bat an eye. This is like what junior programmers wind up learning at some point - nobody cares what the code looks like, just that it does what it's expected to.
Who exactly are you likening to "junior programmers"?
> I say this all as someone who's contributed to some Mozilla repos, has a massive personal investment in Rust (and has read a good chunk of Servo, and liked it), and has been a fan since the Mozilla Suite days.
And you know my personal history on this pretty well too, but you'll notice I'm advocating for Mozilla in this case.
Doesn't mean much since Google controls what features are planned, implemented and included in Chromium.
There just aren't enough users out there to stick to a bad browser for religious reasons. It has to actually be better.
The browser market is as free of a market as you can get. Chromium has won. It had an advantage of being written mostly from scratch with lessons from the failures of Firefox and IE. It's basically too late to catch up. While trying to catch up, they will lose what little market share they have left, and the result will be an undisputable browser monoculture. What Mozilla can offer is a different frontend (or "userspace" if you will) to Chromium. That is the realistic approach.
> I would like the experienced Mozilla team to continue to work on their pro-privacy browser than a decent number of people use.
desktop user agents: 8% and dropping, about to be overtaken by Internet Explorer (lmao)
4.9% with mobile included
<1% mobile only
https://en.wikipedia.org/wiki/Usage_share_of_web_browsers#Su...
Given all the UA-sniffing and purposefully sending broken or degraded sites to various browsers going on nowadays (including on Google properties), no, it's not.
Chrome's main advantage was/is the limitless ocean of resources of one of the biggest three companies in the world, and incessantly being featured on the front page of the biggest (90%+) search machine in the world.
> Chrome's main advantage was/is the limitless ocean of resources of one of the biggest three companies in the world
Yes, this is the reason why it became the best. It had loads of talented developers deployed on it for years and years. You can either complain about it or be happy that it happened and you have access to the fruits of this labor for free.
Of course this is not the experience of the vast majority of Firefox users, so it's pointless to suggest that this means something about Firefox's overall quality. You can easily find people making the same sorts of complaints about Chrome.
Sadly, one of the drawbacks of opensource is that people can take your code and butcher it, and users will blame you.
Some mornings I'm beachballing for my whole first coffee. Eventually I just wind up power cycling the mac if I can't open a shell prompt with which to `pkill -9 plugin-container`....
> pkill -9 plugin-container
Ahhh, ok - that's not FF, that's a crap plugin (likely Flash). It's the browser equivalent of blaming Windows when a crap taiwanese driver gets it stuck. You could try uninstalling all that crap.
Firefox often appears to use large amounts of memory in top / htop, but I believe that's just reported address space allocations. The RES column gives a more accurate depiction of memory usage.
Maybe it's poor Linux support, I have a distressingly high 4 digit number of tabs open on a Windows box and I don't think I've seen it go past 8gb with multiple weeks of runtime.
Edit: Win10 Pro on a Xeon with 48gb ram available
I hardly ever have more than 10 tabs open, and aggressively close everything I am not working with. I also shut down my browser twice a day (2 working locations) and never restore the previous session. I do bookmark some pages, but as a matter of fact I notice that I hardly ever refer to my bookmarks. I don't have the feeling that I am missing out on anything.
Fail to go back and clear them out, mostly. Most were left open because of something relevant at the time, so I mostly need to spend a little time going through and nuking or nothing. There's been little friction due to leaving them open so it hasn't been a priority.
Pretty much the same thing that leaves some people with tens of thousands of messages in their inboxes (I deal with someone who does that and it makes my teeth itch, so my inbox isn't so bad).
As a programmer who has spent significant time with performance work, having useless tabs in a browser would hurt me. But that Google has to search through a bit longer list of messages I can accept as the typical wastefulness of computing these days. (I am old enough to have done time-sharing on 4 MB with 11 other students on their VT100)
Type '%' (no quotes) followed by parts of the URL or title or both in the URL bar. Searches only your open tabs. I can usually find exactly what I want quite quickly.
I must be quite an outlier, because even 10 tabs open at once sounds like a distressingly high number to me.
It has a fantastic FF extension.
Sadly, the sites out there have changed far more than any of the browsers have, and not for the better. They load a ton of crap that does not contribute to the value delivered to the user, but instead contribute value to the site owner and the dozen layers of intermediaries in between user and site owner.
FF -- fix text sizing. As an example, old-style Reddit renders wonderfully on Brave and terribly terribly small on FF mobile. Another tester site is BOFH.
1) Not using the recommended performance settings. Setting 'Content Process Limit' to 2 removed more than half the memory usage IIRC. There are other settings that you will need to go to about:config to change like what size images are cached to memory etc.
2) Use the 'Auto Tab Discard' extension. You will have to configure this to not discard tabs on certain sites or you may lose info you have typed into a webpage. Fortunately this is easily configured by right clicking a tab.
At the moment Firefox is using 700MB to display 5 tabs. 4 HN and on Reddit tab. Given that these sites are close to just text I find 700MB absurd, but that is just the web these days.
Not sure what was going on with your experience, but I use Firefox on x86_64 machines with far less than 28 GiB of RAM daily and several extensions (including uBlock Origin), and I've never had anything like that happen. I'm not sure how JS-heavy the sites I visit are though, as I don't have any issues, so I don't really pay attention to that.
One is a layout engine, one is a JavaScript interpreter.
In many ways test automation tooling has become much easier to use, develop, and manage.
But I suspect the larger driving force is that it’s (arguably) a cost center for an org. The burden of ensuring software quality can be shifted to devs and PMs, though usually with mixed results.
For Mozilla, axing quality and security first is a bad look when those are crucial aspects of a privacy-first company value.