|
|
Subscribe / Log in / New account

LineageOS 22.1 released

Version 22.1 of the Android-based LineageOS distribution is out.

We've been hard at work since Android 15's release in September, adapting our unique features to this new version of Android. Android 15 introduced several complex changes under the hood, but due to our previous efforts adapting to Google's UI-centric adjustments in Android 12 through 14, we were able to rebase onto Android 15's code-base faster than anticipated.

Additionally, this is far-and-away the easiest bringup cycle from a device perspective we have seen in years. This means that many more devices are ready on day one that we'd typically expect to have up this early in the cycle!

Last, but not least, we even had enough time and resources to introduce not one, but two new exciting apps! The first one, Twelve, will replace our aging music app, while the other one, Camelot, will let you view PDF files.



to post comments

Shocking

Posted Jan 1, 2025 1:15 UTC (Wed) by jkingweb (subscriber, #113039) [Link] (9 responses)

A built-in PDF viewer? We truly are, finally, living in the future.

I detested Eleven last time I'd used it, so I'll be interested to see if Twelve is more of the same, or a nice new direction (or, perhaps, a terrible new direction).

Shocking

Posted Jan 1, 2025 10:48 UTC (Wed) by butlerm (subscriber, #13312) [Link] (8 responses)

Considering that Chrome and most other browers have arguably been taking major steps backwards in the performance and desktop integration department for about a decade now a native PDF reader rather than one written in interpreted flavor of the month Javascript sounds like welcome progress. Flash, ActiveX, Acrobat Reader, Internet Explorer, native PDF rendering and native binary plugin support are good things that have all been abandoned to slower and more portable alternatives. HTTPS and DNS are not exactly setting records in the performance department either - especially when better performing alternatives have long been understood or available. Web caching has been going backwards in recent years as well even though that problem requiring independent caches for independent websites would be easy to fix by adding an intermediate layer of caching with the appropriate timing considerations to protect end user privacy.

Shocking

Posted Jan 1, 2025 12:48 UTC (Wed) by rsidd (subscriber, #2582) [Link] (6 responses)

From the release notes:

« The app uses the same Jetpack PDF library used on recent Google Chrome and Files by Google releases, so there’s really nothing outstanding to mention about it: it’s just a simple PDF reader »

And Flash and ActiveX were good things? Seriously?

Shocking

Posted Jan 1, 2025 16:35 UTC (Wed) by butlerm (subscriber, #13312) [Link] (5 responses)

If I am not mistaken Jetpack PDF is only used on Chrome for Android. I believe Chrome for Windows still uses their Javascript implementation which no longer works on Windows 7 since they download it from the Internet and it uses newer Javascript enhancements that the last version of Chrome for Windows 7 does not support.

Flash and ActiveX were and are well designed and of great value to a large number of users, web developers, and software vendors. Flash is no longer supported by the major browsers primarily because it was too heavyweight for a 2007 era Apple iPhone, had a series of security vulnerabilities that Internet Explorer 6 could not handle, and there are plausible if slower and less efficent alternatives in the form of SVG and Canvas. Security vulnerabilities would be easy for someone like Adobe to solve using a combination of static analysis, fuzz testing, careful review, and safe C/C++ compiler technologies if not porting to a relatively safe language like Rust.

Active X is even more useful if you carefully control which ActiveX controls can be activated on a web page, obtain them only from trusted sources, filter them by domain, and if the developers, distributors, and vendors apply the same techniques that would be necessary to make Flash or any binary browser plugin safe to use. Safe C/C++ has a lot of potential even for the Linux kernel even though it is still an active area of research insofar as C and C++ compilers are concerned.

And the best part about ActiveX or comparable binary plugin technology such as Netscape plugins is the ability to closely integrate with desktop applications such as Quickbooks from Intuit or various desktop email and calendaring programs. When Microsoft dropped support for Internet Explorer and Windows 7 it made my job much harder and required my primary client to develop a much more byzantine, slower, and more difficult to install alternative at substantial expense.

Our ActiveX control used for synchronization with Quickbooks had to be rewritten in the form of a Win32 C++ bridge program distributed and installed through a different channel and a browser extension that Google has approval authoritity over and makes a dozen different demands regarding. Mere users and system administrators are not allowed to install Chrome extensions unless they go into developer mode which is not even extension compatible with the same software released through the Chrome web store.

We also previously used Flash for a sidebar menuing system for an online web application and had to abandon that implementation and write two different alternatives in Javascript that did not work nearly as well including one from ExtJS and one coded from scratch And of course Adobe and Adobe Coldfusion do not support Flash anymore either so our use of the Flash gateway and the more efficient Flash binary message system was out as well and had to be replaced with code that used XML RPC over HTTPS in the modern Javascript style as well.

Furthermore as developers and designers of a general purpose construction management and construction accounting system we wanted to integrate with numerous software packages or interfaces that have traditionally only been available on the desktop including Quickbooks, Planswift, AutoCAD, Crystal Reports, Microsoft Excel, Microsoft Word, Adobe Acrobat, and a large number of desktop only accounting or construction accounting systems. It is much harder to do that these days and somewhere along the line the Chrome people have approval authority over everything.

So we are not particularly fond of Apple, Adobe, Microsoft, and Google effectively killing or refusing to support so many technologies these days including many of their own products. Microsoft for example used to recommend that everyone integrate the technology behind Internet Explorer into their applications and now they have abandoned it completely, run campaigns advising people not to use it, and no longer support it even though there are a number of companies that would like to use it indefinitely and would pay good money to do so.

So any version of Quickbooks Desktop more than four or five years old no longer runs properly on modern versions of Windows, Intuit has been trying to kill off Quickbooks Desktop over the objections of their customers for about ten years now, and have them much a much inferior version of Quickbooks Online, and pay them yearly subscription fees that are much higher than what most small businesses used to pay to purchase versions of Quickbooks which were stable and usable for years at a time.

It is also unfortunate that Postscript, PDF, and high quality print oriented rendering and display including support for page numbering, page headers, and footers, rendering html cleanly across page breaks and flowing textual content from one page to a continuation on a different page, and a number of other things that just about every business oriented web application on the planet and many others could use and benefit from is a third class citizen on the modern web at best. I am not sure why none of the major browser developers seem to think that is a problem worthy to solve although Apple seems to have made some moves in that direction. The general attitude among web developers and software as a service providers is either hold print output in contempt or require PDF generation to get anything usable to print cleanly.

In that regard and a number of others the modern web is a much more hostile environment for developing business and other applications than it used to be and there are so many things that would make it much better that have been abandoned and would be relatively easy to fix, or added as commonly supported extensions to html rendering engines and CSS. There is a long list and among them I wonder why every browser on the planet does not support Display Postscript, why Microsoft gave up so quickly on Silverlight and dot NET on the web, and why Oracle has basically abandoned support Java applets.

Shocking

Posted Jan 2, 2025 1:41 UTC (Thu) by rsidd (subscriber, #2582) [Link] (2 responses)

This site is LWN not WWN. Just saying.

ActiveX was a Windows-only technology and not a web standard, and in those days Microsoft pushed such things not just to make the lives of windows users easier but to actively break standards and lock in users to Microsoft systems.

Those days are gone, thankfully. The end was not because people wantonly decided to break your proprietary ActiveX controls and make your life hard. It was because the rest of the web evolved well beyond IE6 and Microsoft found itself unable to keep the browser up-to-date with Chrome, Safari, Firefox.

Shocking

Posted Jan 2, 2025 5:24 UTC (Thu) by butlerm (subscriber, #13312) [Link] (1 responses)

Microsoft Windows is not exactly a trivial market in the business world and there is absolutely nothing stopping anyone from implementing ActiveX, COM, or DCOM on any reasonable platform including MacOS, Android, and IOS. COM and DCOM are based on DCE which was developed by the Open Software Foundation in the mid 1980s.

And the Netscape plugin interface is an effective substitute anyway which was used by PDF and Flash plugins alike before Google decided to kill it off. There used to be a plugin that implented ActiveX on Windows using a Netscape style plugin. And that of course became useless after Chrome dropped support for binary plugins in favor of much slower and much less efficient Javascript extensions.

Of course if no one cares very nuch that their software runs ten times as slow they can stick with Javascript and XMLHttpRequest, which has got to be one of the least performing RPC technologies ever invented. A 200 ms startup time is typical for that, making it roughly ten times slower than something like DCE/RPC mostly due to the latter skipping the incredible overhead of starting a TLS connection.

Also, the only reason why Microsoft struggled to keep Internet Explorer up to date is because for at least five years they had adopted the strategy of making minimal improvements if any to IE 6, and IE7 because Bill Gates rightly feared that the web browser would become a rival platform and wanted Microsoft to have as little to do with it as possible. Sticking their head in the sand did not help matters.

And it is not like a company like Microsoft has to struggle anyway - all they need to do is dedicate enough resources to it which they refused to do, preferring to chase Apple and its thirty percent share of iPhone app revenue instead, another thing they did remarkably poorly by trying to make one platform handle both touch oriented and mouse oriented applications well.

Shocking

Posted Jan 2, 2025 7:03 UTC (Thu) by Cyberax (✭ supporter ✭, #52523) [Link]

> Microsoft Windows is not exactly a trivial market in the business world and there is absolutely nothing stopping anyone from implementing ActiveX, COM, or DCOM on any reasonable platform including MacOS, Android, and IOS. COM and DCOM are based on DCE which was developed by the Open Software Foundation in the mid 1980s.

Sigh. COM/DCOM by itself is simple, they are basically just a standardized way to lay out a vtable. However, all the ActiveX components heavily depended on Win32 API, and it was not easy to implement.

Also, I absolutely HATED the software that used ActiveX components. It was always crashy, slow, and hard to use. It always was much better to have standalone versions of utilities that could be launched normally, rather than bastardized browser-based abominations. Windows Update was a _really_ good example of this.

> Of course if no one cares very nuch that their software runs ten times as slow they can stick with Javascript and XMLHttpRequest, which has got to be one of the least performing RPC technologies ever invented.

I don't think I've ever seen an ActiveX component that was not slow and buggy at the same time. Java Applets were the only technology that was even worse.

> A 200 ms startup time is typical for that, making it roughly ten times slower than something like DCE/RPC mostly due to the latter skipping the incredible overhead of starting a TLS connection.

While I hate most webapps, 200ms is an unusually long pause for modern webapps.

Shocking

Posted Jan 2, 2025 17:28 UTC (Thu) by linuxrocks123 (subscriber, #34648) [Link] (1 responses)

Why didn't you just write a Win32, .NET, or Java application to do these tasks? You're having to fight the web browser so much that it seems like you might be using the wrong tool for the job. There's nothing stopping you from opening a TCP socket and getting / posting HTML from a program other than a web browser, and it sounds like you control the server and so wouldn't have to deal with a lot of corner cases.

Another thing you could do is distribute your own Chromium-based web browser with the extensions you wrote preinstalled. That would get around the "I don't like Google reviewing my code" problem. But, really, it seems like you're going to a lot of trouble just to avoid writing a real program rather than a browser extension.

Shocking

Posted Jan 5, 2025 8:02 UTC (Sun) by butlerm (subscriber, #13312) [Link]

I actually do plan to write a web browser. I would consider starting with Chromium except I prefer to consider a problem from first principles before reading the source code of anyone else. By the time I see the source code of someone else I want to rewrite or make major changes to it and that usually does not go over well whether a project is run by committee or by a benevolent dictator for life.

But releasing a series of patches for something like Chromium or making a fork of it is certainly something I am interested in even if none of those patches ever get merged to the upstream version. A fork would solve most of the problems I am complaining about - with desktop integration in particular.

Shocking

Posted Jan 27, 2025 18:22 UTC (Mon) by Donieck67 (subscriber, #175152) [Link]

Palemoon has been still NPAPI support


Copyright © 2024, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds