Bug #11686

Replace Pidgin: refine blueprint

Added by Dr_Whax 2016-08-20 13:59:08 . Updated 2020-03-30 23:56:11 .

Status:
Confirmed
Priority:
Normal
Assignee:
Category:
Target version:
Start date:
2016-08-20
Due date:
% Done:

0%

Feature Branch:
Type of work:
Research
Starter:
1
Affected tool:
Deliverable for:

Description

Refining the blueprint to replace Pidgin hopefully one day.

team: drwhax, sycamoreone


Subtasks


Related issues

Blocked by Tails - Feature #15816: Can we stop including an IRC client by default? Confirmed 2018-08-19

History

#1 Updated by sycamoreone 2016-08-23 12:35:59

  • has duplicate Bug #11702: Define requirements for an XMPP server for our support channel added

#2 Updated by sycamoreone 2016-08-23 12:36:26

  • is duplicate of deleted (Bug #11702: Define requirements for an XMPP server for our support channel)

#3 Updated by intrigeri 2016-08-27 10:56:00

  • Target version set to 2018

#4 Updated by intrigeri 2017-09-28 12:13:05

  • Target version deleted (2018)

(as per updated roadmap)

#5 Updated by anonym 2017-09-28 15:05:43

What is going to be refined, exactly? I sort of would like to know, so I know if I should spend more time on tracking Tor Messenger.

#6 Updated by intrigeri 2017-09-28 17:05:28

> What is going to be refined, exactly?

IIRC the main open discussion that was left is the one that’s documented as “TODO” on the blueprint. There’s been an almost finished discussion about it on tails-dev@. But I don’t see it referenced on the ticket nor on the blueprint, so you’ll need to find it in the archives to see if there’s more there.

Also, I really don’t remember if the specs on the blueprint have been discussed at all yet.

#7 Updated by intrigeri 2017-09-28 17:23:47

Also, it would be nice to have info on the blueprint (for the main canditates) about :

  • underlying technologies being used such as:
    • UI toolkit, which matters for accessibility, touch and Wayland support (e.g. in the not so distant past we did things like adding QT4 apps and now that’s technical debt we have to deal with)
    • programming language (i.e. is it memory safe? is it something statically compiled and if it is, how are security updates managed?)
    • any chance it works with Flatpak/Portals or similar sandboxing technologies?
  • envisioned update model (i.e. in Debian?)
  • for multi-protocol clients, compatibility with modern mobile messaging (or lack thereof, or realistic plans for it)
  • project status and long-term viability
  • which other OS ships it by default

I’m not saying we have to research all this for all candidates right now: looking at one or two of these criteria might be enough to drop some software from the list of candidates :)

At some point we’ll need to look at UX (e.g. desktop integration, how OTR and similar are integrated) too but the above technical criteria are easier for devs to document, and if we can drop a candidate or too this way, less UX evaluation work will be needed.

#8 Updated by Kurtis 2017-12-11 05:10:56

I created an issue with Gajim to ask them to help Tails understand their project better so that we can make a more informed decision about whether to keep it as a candidate or not: https://dev.gajim.org/gajim/gajim/issues/8796

#9 Updated by Anonymous 2018-01-15 11:11:52

Updated the blueprint with Debian status of the proposed replacements.

#10 Updated by Anonymous 2018-01-15 11:20:09

  • Starter set to Yes

Next steps:

- Add the requested information for https://labs.riseup.net/code/issues/11686#note-7

- Discuss the requirements & update the blueprint accordingly (Send email to tails-dev)
- Discuss the TODO item: “would a pair of two separate client (XMPP and IRC) also be okay, or are we only looking for a single client that can do both? In fact, it is not even clear if Tails needs to contain an IRC client at all, after #tails and #tails-dev have been moved to XMPP. ” -> This sounds like a discussion to be held on tails-dev or during a monthly meeting. Please initiate this discussion and track the outcome.

@Dr_Whax: do you still intend to work on this?

#11 Updated by Dr_Whax 2018-01-24 21:27:11

u wrote:
> Next steps:
>
> - Add the requested information for https://labs.riseup.net/code/issues/11686#note-7
> - Discuss the requirements & update the blueprint accordingly (Send email to tails-dev)
> - Discuss the TODO item: “would a pair of two separate client (XMPP and IRC) also be okay, or are we only looking for a single client that can do both? In fact, it is not even clear if Tails needs to contain an IRC client at all, after #tails and #tails-dev have been moved to XMPP. ” -> This sounds like a discussion to be held on tails-dev or during a monthly meeting. Please initiate this discussion and track the outcome.
>
> @Dr_Whax: do you still intend to work on this?

Yes, to all of this. Thanks!

#12 Updated by Kurtis 2018-01-25 00:20:36

@Dr_Whax: I got some good responses related to this from Gajim here: https://dev.gajim.org/gajim/gajim/issues/8796

The Debian mantainer for Gajim also responded.

Here’s the debian repo: https://packages.debian.org/stretch/gajim
Here’s the debian repo for the gajim OMEMO plugin: https://packages.debian.org/stretch/gajim-omemo

Please let me know if you’d like me to ask them any more questions.

#13 Updated by Kurtis 2018-01-25 00:28:16

i’m not extremely educated on debian repos, but I think I posted the wrong one. I think this is the one that Tails would want: https://packages.debian.org/stretch-backports/gajim

#14 Updated by intrigeri 2018-03-01 08:18:58

  • blocks Bug #8574: Test CoyIM in Tails added

#15 Updated by Anonymous 2018-08-17 15:27:00

ping?

#16 Updated by intrigeri 2018-08-19 10:59:01

  • blocked by deleted (Bug #8574: Test CoyIM in Tails)

#17 Updated by intrigeri 2018-08-19 11:11:26

  • blocked by Feature #15816: Can we stop including an IRC client by default? added

#18 Updated by intrigeri 2018-08-19 11:12:47

I’ve moved the IRC support discussion, that I’ve started 2.5 years ago, to Feature #15816. I’ll make sure we lead it to a conclusion. Then we can come back here.

#19 Updated by Dr_Whax 2019-07-07 15:18:25

  • Assignee deleted (Dr_Whax)

#20 Updated by cypherpunks 2020-02-22 11:54:45

Out of that old blueprint, it seems like dinoIM is the thing that survived and comes with OMEMO, which seems to be the standard nowadays.

#21 Updated by syster 2020-03-30 23:56:11

>Out of that old blueprint, it seems like dinoIM is the thing that survived and comes with OMEMO, which seems to be the standard nowadays.

also gajim. Works well in tails, and has a stable candidate in debian.