Browsers! and how we develop for them

Discuss requests for comments/changes posted in the Issue Tracker for the development of phpBB's style.
Forum rules
Please do not post support questions regarding installing, updating, or upgrading phpBB or modifying styles of released phpBB versions. If you need support for phpBB please visit the Support Forums on phpbb.com.

If you have questions regarding creating styles please post in Styles Support & Discussion to receive proper guidance from our staff and community.
Post Reply
User avatar
hanakin
Front-End Dev Team Lead
Front-End Dev Team Lead
Posts: 939
Joined: Sat Dec 25, 2010 9:02 pm
Contact:

Browsers! and how we develop for them

Post by hanakin »

There is a lot of information/dis-information out there surrounding all the browsers. Which ones to use, which ones are most popular etc... The aim of this post is to be sort of a one stop shop for all vetted accurate information surrounding all the browsers we see as valid and support worthy with reasoning and justifications. As all of this is subject to change over time and one person is not fallible, please respond with any info you may have that may conflict anything in the post. We will review it and update this post as necessary.

The intent of this post is to keep it updated and eventually hang it on either.com or in a blog post or something.

phpBB Support policy
To maintain proper balance of coverage and sustainability we will support all major browsers with greater than 1% Global Usage plus the last major versions. This is across all devices/platforms.

Tiered Support
We maintain two tiers of support for these browsers. The first tier is those that we test against. These are major browsers with market share above 3% or is the default browser for a major operating system. The second tier are browsers that we support through use feedback/bug reports. These browser we will not test for but we will patch. These are all browsers that we support that are not in the first tier.

This gives use extremely efficient development testing with greater than 85% coverage of our supported browsers. Here you can find a complete list of all our supported browsers Supported Browser List

Given those browsers in the list above, here they are broken out by Tier

Tier 1 (Mac/PC Only)
Chrome (Most Recent)
Edge (Most Recent)
Firefox (Most Recent)
Safari (Most Recent)

Tier 2 (Mac/PC/Linux/IOS/Android)
Everything else from the link above that does not fall into tier 1

* We consider IE 11 an outlier and will no longer be fully supported in 4.0. in 3.3 we currently treat it as Tier 3 which is Tier 2 but we reserve the right to not patch certain bugs


What is a Major browser?
By saying major browser, we do not support what we call edge case browsers. Certain browsers keep themselves above 1% global usage by employing shady practice, or targeting certain restrictions in there favor. Examples of this are browsers that are forcing the user to use them by default without an option to change the default. We see this as an edge case and reserve the right to not support it if we see it as a problematic browser. Another example of this is ad/bots. Certain browsers gain there usage by running through hard-coded background processes on certain devices that load through a specific browser without the users control. Thus we also reserve the right to not support a browser solely on its usage % but also based on its reputation.

What is a Global Usage and how is it determined?
We rely on several industry standard tools to ensure that all of our code is compliant with our supported browser.
These tools rely on usage statistics by StatCounter GlobalStats. This is the only official statistics. They are updated yearly! which is very important for reliable data. Individual stats can and do flux from month to month, especially when a new release of a popular browser with alot of marketing behind it hits the streets like IDK Microsoft Edge 80 (Chromium) . This is a perfect example of why we rely on stats from this source.

This browser which is actually despite the name not based on chrome! Instead just like chrome & safari it relies on https://webkit.org/project/ which is an open source Web Content Engine for the web. Chrome and now this new edge just applies its own skin and exts on-top to this standard. Infact there are other similarities in some of these browsers as they can also use the same exts at times as well such as the develop tools which is also another open source project.

Back to the main point why this is a good example for why the data can flux. This browser was given a lot of media based on Microsoft to webkit in this build. Given its release date which was just a few months ago in January. We are now seeing articles saying that it has surpassed firefox.

But is this accurate? Its actually misleading and too early to tell. Given that its so new a lot of individuals are still riding the hype and testing it out. Its also pulling a lot of users from IE11 which is also inflating those numbers. Not to mention that this year is the mass adoption year for Windows 10. Meaning a lot of individuals are for the first time getting access to edge which is also inflating those numbers. All of this and more leads to less than stable data. Thus its better to wait until the end of the year to get better results. In-fact Microsoft is well of ware of this I am sure which is why they chose to release when they did.

Where does this information evryone is talking about then come from? It comes from Net Market Share. Which is another site dedicated to usage stats. However its currently not the standard for the global development community until an api is developed for the tools that we leverage. Whats the difference the difference is mainly in how they interpret information. According to this article https://www.zdnet.com/article/net-marke ... you-trust/, NetMarketshare on the surface is more representative of a clean average weighting every user the same. StatCounter weights heavy users more which gives you a better representation of what browsers you may see in you websites logs. This is why if you look on StatCounter you will not see Edge with a higher percentage then firefox. One other difference is in the browsers themselves, each has a list of what browsers they get results for that may differ.
Donations welcome via Paypal Image
User avatar
AbaddonOrmuz
Registered User
Posts: 7
Joined: Wed Jul 02, 2014 9:44 pm

Re: Browsers! and how we develop for them

Post by AbaddonOrmuz »

That's good news.

I can't wait to refactor some of my extensions to use ES6+ (arrow functions, modules, classes, fetch(), etc.) and stop using polyfills explicitly for IE11 :lol:
Post Reply