Pale Moon testing

[Rich3w]

Given some the of issues with the latest milestone of Pale Moon:
I'd like to contribute to the test strategy of Pale Moon however, I can't find seem to find any test packs, regression or new functionality tests that are run every release.

Can someone please link me to them? Sorry if I seem dumb, but I can't find them anywhere. Maybe link me to the team lead that does the testing?

[doofy]

Given some the of issues with the latest milestone of Pale Moon

Which ones are those?

[New Tobin Paradigm]

It's called real world testing using the unstable channel. Download and install the unstable channel and watch the commits and issues on github and test the browser out in the real world.

We prefer this instead of some form of automated testing suite crap because it would be an endless cycle of writing new tests and fixing old ones. Just look at Mozilla they have thousands of tests in their suite and they constantly spend time writing, rewriting, and running tests as if that makes their browser good.. Given you use Pale Moon I am sure you see where this is going.

Suffice it to say, at the end of the day, it isn't worth a damn if a controlled test passes for any feature or change if in the real world it doesn't work as expected on some website.

We would rather spent our limited time and resources making the thing work correctly in the real world than trying to get one of a hundred thousand green checkmarks on some panel.

Is shit gonna slip by from time to time? Yes.. But that already happens anyway everywhere because targeted controlled tests can only test for targeted controlled conditions.

So if testing and validation is something you are interested in helping with then get on the unstable channel and watch github commit log and issues and just bloody well use the thing to test in the real world before those hit a release channel.

[Fedor2]

Try build with debug enable, you will see many things for the testing.

automated testing suite crap

Can it be all removed from the code?

[New Tobin Paradigm]

Eventually, save for layout reftests and spidermonkey tests. Those are still useful.

[Rich3w]

It's called real world testing using the unstable channel. Download and install the unstable channel and watch the commits and issues on github and test the browser out in the real world.

We prefer this instead of some form of automated testing suite crap because it would be an endless cycle of writing new tests and fixing old ones. Just look at Mozilla they have thousands of tests in their suite and they constantly spend time writing, rewriting, and running tests as if that makes their browser good.. Given you use Pale Moon I am sure you see where this is going.

Suffice it to say, at the end of the day, it isn't worth a damn if a controlled test passes for any feature or change if in the real world it doesn't work as expected on some website.

We would rather spent our limited time and resources making the thing work correctly in the real world than trying to get one of a hundred thousand green checkmarks on some panel.

Is shit gonna slip by from time to time? Yes.. But that already happens anyway everywhere because targeted controlled tests can only test for targeted controlled conditions.

So if testing and validation is something you are interested in helping with then get on the unstable channel and watch github commit log and issues and just bloody well use the thing to test in the real world before those hit a release channel.

Any software developer worth their salt in the real world will just be astounded by that response. Live proving comes after system testing which you evidently don't do.

You know you depend on Mozilla testing and finding bugs as well as for the code that you needed to rebase to get a working modern version of Pale Moon. That you don't also test all the changes you've made in a reproducible manner is, quite frankly, ridiculous. I guess there are real world developers and hobbyists.

I'll leave it there as you know full well that isn't how software should be tested. You just lack resources. I wish you well.

[fillerup]

where are these 'test strategies' you promised to 'contribute'? oh wait.. there's none, cos being helpful isn't your actual intention. nope, it's another attempt to disparage the PM team while hiding behind the facade of trying to be helpful. too bad you're so awful at it, everyone can easily see through your BS

[Isengrim]

That you don't also test all the changes you've made in a reproducible manner is, quite frankly, ridiculous.

Reread the above post. He said that they do not use automated testing strategies - that is not the same as not testing at all. Automated testing is merely one of many testing strategies, not the be-all end-all solution to software testing.

I think what's more ridiculous here is that you're implying that Pale Moon is completely untested, which could not be further from the truth.

If you want to see clearly documented test plans and a well-maintained automated testing framework, you're welcome to put them together. This is, after all, a community effort.

[Moonchild]

Don't waste your breath on Rich3w. They obviously don't understand that immediate testing as things are implemented is part of normal best development practices. Automated tests are good for just one thing: catching regressions caused by code changes in parts related to what was changed, if they aren't caught already in the development stage. Everything else is and will remain work for actual (human) testers, who with their wide range of different setups and workflows will give us the real-world feedback that no automated test can ever hope to achieve.
This also means that these automated tests are, for all the immense work that they require to write and maintain (a single line of code changed can require dozens of changes to these tests), completely unable to catch anything that isn't explicitly tested for, which has bitten Mozilla many, many, many times over ("tree is green, ship it" -- "oops we need a chemspill release"). This would even be 100x worse if they didn't rely on people on the beta and nightly channels to report issues found (the same as our unstable channel users).

[New Tobin Paradigm]

If Mozilla testing was SOOOO spectacular why did we spend 9 months once again fixing and improving such a "perfectly tested" Mozilla codebase to a point of being at best acceptably viable. It only took a few months to port and adapt Pale Moon to the codebase.

Besides, if all anyone wanted was "business as usual" why do we feel we need to exist? Surely, Mozilla with its hundreds of millions must be doing a better job than we are. For that matter, why are YOU here, Rich3w?

[Rich3w]

I NEVER mentioned automated testing. Why are so many people trying to distract from the discussion that basic repeatable and auditable tests are not being done on Pale Moon.

Of course testing is done as code is developed. Then it is further tested by someone other than the coder in a reproducible manner. Often at more than one level.
This is a basic requirement of any software development. Apparently this project is an exception and knows better.

If Mozilla testing was SOOOO spectacular why did we spend 9 months once again fixing and improving such a "perfectly tested" Mozilla codebase to a point of being at best acceptably viable. It only took a few months to port and adapt Pale Moon to the codebase. ?

No doubt you can show us all the bugs you raised with Mozilla while fixing their code so they can benefit. Or have you just grabbed their code and given nothing back as Moonchild has previously accused Mozilla of Ref?
And if you are honest, you are adding Pale Moon to the Mozilla codebase. There's far more Mozilla code there than changed code.

Besides, if all anyone wanted was "business as usual" why do we feel we need to exist? Surely, Mozilla with its hundreds of millions must be doing a better job than we are. For that matter, why are YOU here, Rich3w?

Ask yourself where would Pale Moon be without the last codebase you used from Mozilla? You offered (meagre) rewards for desperately needed functionality to keep PM relevant. It proved way too complex and another rebase of the code had to happen. Although it was a nice spin that Mozilla had done such a bad job you only 'just' noticed

And, maybe, just maybe I have experience you could use. Or rather could have used.

No doubt there will be a smug reply based on a single point that the cultists of Pale Moon can lap up regardless of the facts presented.

[satrow]

The above missive is coming to you from a Windows 10 user, as if MS are now the paragon of thorough testing before release...

[doofy]

And, maybe, just maybe I have experience you could use.

And, maybe, just maybe, you could offer it instead of whining.

[Thehandyman1957]

So what is the point of this conversation anyway???

If all you came here to do was rank on the efforts of others then
I think it's time to bugger off.

No doubt you can show us all the bugs you raised with Mozilla while fixing their code so they can benefit.

As if they would care? I could just hear the conversation right now.... Hey Moz, I just found a bunch
of bugs in your code from version???, what, what's that you say??? Your not working on that version anymore???

Ask yourself where would Pale Moon be without the last code base you used from Mozilla?

Last time I checked, it was still open source! Why don't you ask the same question to everyone
who has added to the Linux code base to make their own operating system?? Does that make them less
than you or others?? Your logic eludes me.

This really just sounds like a Moz fan boy trying his best to cause trouble. Good luck with that!

[Sajadi]

And, maybe, just maybe I have experience you could use. Or rather could have used.

No doubt there will be a smug reply based on a single point that the cultists of Pale Moon can lap up regardless of the facts presented.

I give you instead a nice advice. Listen carefully boy:

There are:
Chrome
Firefox
Edge
IE
Chrome variants
Simple Firefox rebrands

Now pick one, leave and enjoy your life without wasting anymore time at a place which is inhabitated by - by your own words - cultists. Simple, isn't it? Oh, and when you are on your way out, please close the door in a silent way

Because that is really the question, why are you wasting time at a place you do not like and spend time using a browser you do not like. As Tobin already said:

For that matter, why are YOU here, Rich3w?

[New Tobin Paradigm]

Two last things I wanna add before this topic falls into the ether for being entirely pointless.

We have bugs on bugzilla and the response is hardly ever positive.

As well as, if Pale Moon is a cult .. Apperently we did it wrong because cults typically suck people in and never let them leave.. Where we at the core only hope to be useful and regularly advise people to use what works best for them which may not be our stuff.

Welp, that's all.. Well not really but gotta stop at some point and now seems like a good place, least as far as my real world testing can tell anyway.

[moonie]

Man such vitriol, all over a browser. Like they're saying, if you don't like it, don't use it, and use something else. At least people here are willing to offer advice and help to the best of their knowledge. There are bigger problems in the world, and most likely in your world, no offense.

[Moonchild]

Why are so many people trying to distract from the discussion that basic repeatable and auditable tests are not being done on Pale Moon.

If you feel this is an essential addition to the way we are supposed to run our project to meet your standards, then what's stopping you from getting involved and doing this testing? Oh, that's right, you don't -actually- want to contribute, only complain about how we're not doing it "right".

*walks by in an overall, whistling, stringing up a rope with a sign "Exhibit closed"*