MozillaZine

The State of the SeaMonkey Union!

Discussion of general topics about Seamonkey
frg
 
Posts: 314
Joined: December 15th, 2015, 1:20 pm

Post Posted May 2nd, 2017, 6:25 am

Greetings,

I am just relaying the following message for the SeaMonkey Council. It is not on the web page but it is official. If you have any questions you can either ask here or use the provided web address.

Have fun
FRG


+++ snip +++

Hello fellow users and supporters of the SeaMonkey suite.

The important part first: SeaMonkey is alive and we do not plan to discontinue it. But in continuing to deliver the best and only suite based on the Mozilla Gecko web engine, we need your help.

Lets first start where we are.

SeaMonkey 2.46 was released in late December after struggling for months with infrastructure and build issues. While still using the Mozilla infrastructure, we are mostly on our own here. We plan to release a final 2.48 based on the level of Firefox 51 soon. Again this is late due to infrastructure and build issues, but not so much as 2.46. Being based on Firefox 51 means that security patches are again not up to par with current Firefox. Believe us, we dislike this as much as you do.
The SeaMonkey project is entirely driven by volunteers working on it in their free time. The current members of the core team (count up to 7) are committed but, with all the changes, are getting slowly overwhelmed.
This means bugs do not get fixed as fast as we would like. With an aging infrastructure becoming more and more abandoned by Mozilla, due to switching Firefox building to the cloud, releasing builds does not become easier. It also means that if even one of the current key people quits, the project is in danger of becoming un-maintainable which is even worse.
Also keeping up with Firefox is becoming difficult at best. Mozilla plans to discontinue classic extensions and themes with Firefox 57 which is right around the corner. XUL, one of the key technologies of SeaMonkey, is also on the chopping block and will be discontinued in the near future. The replacement technologies, based on modern web standards, are immature and still under constant development. In the end, an almost complete rewrite of the current program will probably be needed. If it weren't for our friends from the Thunderbird project, we would now have even bigger problems.
The good news is that financially we are a little better off than last year. DuckDuckGo is now the default search engine of SeaMonkey. Every time you use it for searches in SeaMonkey we get paid. For the conspiracy seeking people out there:) Not being able to easily set the search engine in 2.46 to another provider was a genuine bug with a workaround documented in the release notes from day 1. It has now been fixed.

What we plan:

After releasing SeaMonkey 2.48 we will switch to the Firefox 52 ESR source code for 2.49.x releases. This means that the code base is more or less frozen for a few release cycles and only security updates and bug fixes will be in the releases.

The infrastructure issue has been discussed. While critical, there are no final plans yet (also thanks to lack of manpower). Thunderbird is in the same boat and we hope to work out something together. If worst comes to worst, we could ask our team member Adrian Kalla to produce our regular builds. This had been discussed earlier. It was dismissed, for now, as no crash symbols for builds would be available on the Mozilla servers.

Switching to ESR means we can work on bugs in the current tree for a while longer without having to fear that they are carried over into a release. They need to be fixed for the next ESR of course.

The most critical issue is to support web extensions in one of the next releases. It is unclear how long we will be able to support classic extensions.

There are a number of Gecko Forks in the wild. We do not plan to switch over to one of them as the basis for SeaMonkey. We think that they currently do not have enough developers themselves to cope with the changes Mozilla plans. Web technologies are also evolving all the time and we fear that they are not able to keep up.

Also, we are not planning to support any abandoned stuff like classic extensions and NPAPI plugins on our own. We will try as long as possible. But when they are gone, they are gone. The current developer base is much too small to do our own fork.

Based on how successful Mozilla is, or if one of the forks gain ground, this might change in the future.

What we need:

Setting up our own infrastructure, potentially in conjunction with Thunderbird, will cost. If you feel you can contribute towards future releases in this way, please consider donating:
https://www.seamonkey-project.org/donate/

But what we need even more is people to help out. Even if you are not a developer, you can help. For example, writing a document such as this takes time. Also, maintaining a website is not done by staring at it all day.

So if you want to help, these project areas are looking for a few good contributors:

- Development. Most code is either JavaScript or XML/CSS mixed with C++ and Mozilla technologies based on XUL and friends. In SeaMonkey not so much HTML right now, but this might need to change. The build environment makes heavy use of Python.

- Graphics: Icons and symbols need a face lift for HiDPI screens. There are plans to switch everything over to svg files in the backend which would mean a massive effort to convert existing files. With a sometimes extremely conservative user base when it comes to changes in the interface, not an easy job

- Website: Some areas are severely outdated and things like release notes need to be written too.

- Bug hunting and triaging: We could use a few more people to check out bugs and try to reproduce and categorize them in Bugzilla. We are especially short on people doing this on macOS. While we are on it. Writing lengthy threads in the news and support groups is fine but if no one actually reports them as a bug in the end they usually won't get fixed.

- Everything else not covered above. If we forgot something you can fill this slot too. Just think about it.

As a final statement, we do not think that SeaMonkey will take over the browser world any time soon. SeaMonkey is a niche product and will stay that way. Too many people are not interested in a classic suite anymore and most users are happy to use what is hip. That is OK with us. It's all about choice.

We would like to continue supporting the power users like ourselves and those who are looking for something different and flexible without reinventing the wheel with every release. We try to listen to you, our user base, for advice/orders/demands/suggestions. Of course, we won't be able to implement everything under the sun. But we would still like to implement something and stay current. It's your call now.

If you would like to support us, either send a mail to us, the SeaMonkey Council (seamonkey-council at mozilla dot org), ask for guidance in the official support groups or just pick your favorite unassigned bug from Bugzilla and start.

We are looking forward to hearing from you.
The SeaMonkey Council
Last edited by rsx11m on May 2nd, 2017, 9:17 am, edited 1 time in total.
Reason: made a 30-day sticky

Peter Creasey

User avatar
 
Posts: 396
Joined: October 26th, 2007, 2:32 pm
Location: Texas

Post Posted May 2nd, 2017, 6:41 am

Everything y'all are doing is much appreciated. The SeaMonkey Suite is invaluable for everyone who is familiar with its capabilities and performance.

Thank you!
. . . . . . . . . . Pete

barbaz
 
Posts: 1580
Joined: October 1st, 2014, 3:25 pm

Post Posted May 2nd, 2017, 8:40 am

frg wrote:DuckDuckGo is now the default search engine of SeaMonkey. Every time you use it for searches in SeaMonkey we get paid.

I use a custom DuckDuckGo searchplugin and a spoofed user-agent. Do I need to change either of those things to have you guys paid when I use DuckDuckGo, if so how?
*Always* check the changelogs BEFORE updating that important software!

frg
 
Posts: 314
Joined: December 15th, 2015, 1:20 pm

Post Posted May 2nd, 2017, 8:51 am

Probably the t:
https://duckduckgo.com/params

Its in the searchplugin. I can ask mcsmurf.
<Param name="t" value="seamonkey"/>

barbaz
 
Posts: 1580
Joined: October 1st, 2014, 3:25 pm

Post Posted May 2nd, 2017, 9:02 am

Looks like it, thanks.

If you could ask mcsmurf whether a SeaMonkey user-agent string is also needed, that would be great.
*Always* check the changelogs BEFORE updating that important software!

Frank Lion

User avatar
 
Posts: 19085
Joined: April 23rd, 2004, 6:59 pm
Location: ... The Exorcist....United Kingdom

Post Posted May 2nd, 2017, 4:53 pm

frg wrote:We would like to continue supporting the power users like ourselves and those who are looking for something different and flexible without reinventing the wheel with every release. We try to listen to you, our user base, for advice/orders/demands/suggestions. Of course, we won't be able to implement everything under the sun. But we would still like to implement something and stay current. It's your call now.


I've a couple of huge business meetings to prepare for, so I'll reply on this in full in a couple of days....possibly.

So, meantime, everything y'all are doing is much appreciated. The SeaMonkey Suite is invaluable for everyone who is familiar with its capabilities and performance.

Thank you!

:)
Last edited by Frank Lion on May 2nd, 2017, 8:06 pm, edited 1 time in total.
Metal Lion latest SeaMonkey & Thunderbird Themes - Sea Monkey and Silver Sea Monkey
"The only thing necessary for the triumph of evil, is for good men to do nothing." - Edmund Burke (attrib.)

tonymec

User avatar
 
Posts: 695
Joined: October 15th, 2004, 2:58 am
Location: Schaerbeek (near Brussels, Belgium)

Post Posted May 2nd, 2017, 5:58 pm

@frg: It is true that ~akalla builds don't get their symbols within reach of Socorro; but they are published with a crashreporter-symbols.zip. That zipfile is much too big to append to a BMO bug report, but it can be used (though admittedly not easily) by some geeks like me to translate the hex addresses in a stack dump into symbolic labels. Not perfect, since it requires some know-how that I learnt when compilation and link maps were still printed at 1100 lines/minute on 132-column zig-zag pyjama paper as a matter of course for every compile, and I suppose that such maps look like Chinese to most people nowadays (or like Cherokee script to East Asians), yet it is better than nothing.
Best regards,
Tony

libertyernie
 
Posts: 141
Joined: January 20th, 2015, 12:29 pm

Post Posted May 2nd, 2017, 6:11 pm

A little off-topic, but regarding SVG. If you can't use PNG icons anymore, couldn't you just create SVG files whose only element is an embedded bitmap? No quality improvement, of course, but it ought to work.
I wouldn't try to convert bitmap graphics to vector graphics unless it's by hand (although a new default icon set would be welcome.)

Also, thanks for adding the devtools to SeaMonkey! If it wasn't for that I'd probably be using another browser by now. In bug 1361079 I've got a preliminary patch to add a Browser Toolbox menu item to other windows - it's great for debugging extension code.

JakeMS
 
Posts: 3
Joined: May 6th, 2016, 8:46 am

Post Posted May 3rd, 2017, 7:08 am

Well, I've got a couple of projects I need to finish myself - hopefully should be finished with them next week. Once their done I'll be sending an email off to see if these guys still need help, if they do I'll try to help where possible :-D.

frg
 
Posts: 314
Joined: December 15th, 2015, 1:20 pm

Post Posted May 3rd, 2017, 9:31 am

JakeMS, you can be sure that is nothing which will or can be fixed in a week so help is much appreciated.

libertyernie. Currently png support is still there. iIjust saw a post somewhere not sure where that icons need to be svg in the future. Don't have a bug number. But what is needed is HiDPI support. gtk3 support on Linux is also not where it should be.

Arch5

User avatar
 
Posts: 1066
Joined: April 11th, 2014, 7:34 pm
Location: Melbourne Australia

Post Posted May 3rd, 2017, 6:24 pm

my guess Thunderbird will also eventually support WebExtensions if Seamonkey will be doing that ? so in short does this mean Seamonkey will also use firefox's New Rendering engine so they dont have to support Gecko themselves ?

frg
 
Posts: 314
Joined: December 15th, 2015, 1:20 pm

Post Posted May 4th, 2017, 1:11 am

barbaz

> If you could ask mcsmurf whether a SeaMonkey user-agent string is also needed, that would be great.

searchplugin should be all what is needed according to mcsmurf.

Lemon Juice
 
Posts: 764
Joined: June 1st, 2006, 9:41 am

Post Posted May 4th, 2017, 2:55 am

frg wrote:The most critical issue is to support web extensions in one of the next releases. It is unclear how long we will be able to support classic extensions.

[...]

Also, we are not planning to support any abandoned stuff like classic extensions and NPAPI plugins on our own. We will try as long as possible. But when they are gone, they are gone. The current developer base is much too small to do our own fork.

I think this is an important part because either way it looks like SM is going to lose the ability to run extensions that have full power over the browser and can interact with the underlying OS freely. Ideally, it would be great if SM retained XUL and classic extensions stuff, even if these technologies were frozen in development they would work and bring a lot of benefit. But as I understand, keeping that stuff with whatever new engine is adopted from Fx might be the main obstacle.

Another solution might be dropping traditional extensions but extending WebExtensions way beyond what other browsers allow, that is allowing to access browser's UI, low level API, etc. But that would mean the necessity to create and maintain a niche WebExtensions specification and having to convince extension developers to write stuff for SM.

Anyway, I think these things may not happen unless we can attract SM users and developers interested in this platform. This appears to be biggest challenge.
*** SeaMonkey — weird name, sane interface, modern bowels ***
Mouse Gestures for SeaMonkey/Firefox
Convert Fx and TB extensions to SeaMonkey

Arch5

User avatar
 
Posts: 1066
Joined: April 11th, 2014, 7:34 pm
Location: Melbourne Australia

Post Posted May 4th, 2017, 3:10 am


Lemon Juice
 
Posts: 764
Joined: June 1st, 2006, 9:41 am

Post Posted May 4th, 2017, 3:21 am

Or perhaps SM could join forces with Waterfox developer who plans to get funding and fork XUL to retain classic extensions support. He has an ambitious plan but worthy of praise!
*** SeaMonkey — weird name, sane interface, modern bowels ***
Mouse Gestures for SeaMonkey/Firefox
Convert Fx and TB extensions to SeaMonkey

Return to SeaMonkey General


Who is online

Users browsing this forum: No registered users and 1 guest