I was e-mailing with @CynthiaReynolds the recently and there is one topic I find important to bring up here. It’s regarding the chat communication channel at Slack which I find to be challenging the idea of an OpenFoodNetwork and limiting the potential of OFN’s growth and flexibility (see The Open Organization by Jim Whitehurst). Also I find it potentially dangerous because of e.g. (quote @CynthiaReynolds):
“OFN has gotten a free non-profit Slack plan, but yes archiving will be an issue, it was discussed and we chose to stick with it”
I think the archive issue, data ownership, gated community and lock-in of users are all big problems for both the short & long-term sustainability of OFN. Open channels and communication protocols which respects and upholds the same philosophy as OFN stands for will more easily get e.g. hackers more interested to contribute to the software.
So I’d like to start the conversation on the main alternative which can strengthen OFN’s credibility as an OpenFoodNetwork, foster community, avoid future pains and enable easier growth.
XMPP (aka Jabber) - The Open Standard since 1999 (and what runs e.g. Facebook Chat & Google Chat)
About: https://xmpp.org
There are numerous free and public servers to host chat rooms. (can’t add a 3rd link here as a new user)
There are also numerous great clients for Windows, OSX, Linux, iOS & Android
Example of Webclient https://conversejs.org/
There are more alternatives to present here but XMPP is the most long-term sustainable choice at this point of time and it’s constantly being developed, works great with other services, group chats and will never have to cost OFN anything.
I’d like to add that there are open source alternatives to Slack that more closely mirror what it provides, and might make for a smoother transition. A few options that came up:
I agree that Slack is not really open. And I really like XMPP with the open protocol, lots of native clients and plugins like OTR to securely exchange secrets like passwords.
I’m not sure about the archiving though. I think that chats are like face-to-face communication. The said word is volatile. The results have to be documented somewhere else (e.g. Discourse) if they should be preserved for later. I like the freedom of just saying what comes to my mind instead of being aware that it will be publicly available 30 years from now and someone may get it wrong in a different context. On the other hand, as soon as a channel has a certain number of people and is public, it’s good to think about what you are saying anyway and an archive would be really good for transparency and future research.
The two alternatives Rohan found look promising. Mattermost would probably be the easiest transition. The downside is the cost of hosting and administration.
It is my understanding that we have been recognised as a non-profit and therefore have free indefinite access to Slack’s ‘Standard’ plan, giving us unlimited archives, so I’m not sure that there is an issue with archiving…
Data ownership is potentially an issue, but I personally don’t really have a strong feeling about it.
As far as I understand it, there is nothing stopping us from opening up our Slack community for anyone to join the conversation. The gated-community thing is purely a manifestation of the way we are currently using Slack, not a limitation of Slack itself.
@matt (welcome!), could you explain what you mean by ‘lock-in of users’?
I guess my final thought is that a transition would not be zero cost, we have an existing community with existing conversations, which is working quite well and costs us nothing. We would need to migrate these existing conversations over to a new platform or accept losing them. Setting up a hosted solution would take time, and as @maikel says, ongoing maintenance and hosting would be an additional cost.
Hi all, thank you @matt (and welcome!) for opening the topic.
I personally share @oeoeaio view on it. I agree that “free software” is not “open software” and that data ownership is something we will need to tackle at some point, but the nature of conversations here in Slack for me is, like @maikel said, chatting, nothing we want to document. Documentation is another topic, and we have Discourse for that, anything we need to document go on it.
I value more the easiness of cooperation that the “purity” of our behavior, looking for purity is what unfortunately can kill communities and unperfect solutions are the core of agility, purity can kill action and block the flow of energy as it requires more energy to get into it. Today lots of people use Slack (I do for 4 communities and sincerly I’m tired and depressed by the idea of learning and switching to a new tool again… I have a deep pain feeling), it’s pretty easy to get contributors on Slack.
I would rather put my energy on developing the features that our users need than switching chat room, as a priority…
This being said, maybe @sylvain will have say here? I think you are working on something regarding Slack?
Yes indeed! Happy Dev is a for-profit organization, so we don’t have any special plans for Slack.
We are in the process of migrating to a Rocket Chat solution, hosted by our hosting team. We are also offering this solution to our partners and plan to offer it to more people in a near future. So if OFN is interested, we can probably figure something out.
It is built on an open protocol called Matrix which offers tonnes of great functions like voice and videocalling, integrations like GitHub & IRC as well as design with simple guest access.
Disclaimer: I’m not connected to any alternative, just an advocate for sustainability through openness.
There is maybe not an issue with archiving right now but could be when OFN potentially migrates to another service.
Data-ownership is important in the sense of OFN work material either not being OFN’s or otherwise not being open-source, but instead being Slack’s.
With the lock-in of users, well, basically, one example is this: If Slack all of a sudden decides on doing something that affects OFN or me personally in a negative way - then I am entirely powerless to do anything about because I bought into their total control and I can’t fix it even if I am technologically knowledgeable. As an allegory it is like buying a unique car which is only sold by one manufacturer which has total control over the supply chain. I get the car for free but when it breaks down and I need to repair it, it is too late because the replacement parts are unique, only sold by them and costs way much for my wallet to handle.
As a new user here I cannot post more than 2 articles/post so I’m posting another here about one guy arguing for Riot as a priority before Slack:
Hi @MyriamBoure, nice to meet you online this time!
I agree with you that there is an importance of flow, and maybe it is extra important to OFN right now. My intention with starting this topic is not to force a fast decision, rather to have an low-pace ongoing dialogue about it and evaluate when the timing is better to potentially do a switch. I’m sorry to hear about your pains about platforms, it was never the intention to cause any harm by opening this topic. I opened it because I see a need for being proactive to prevent harm being done to OFN in the long-term future. Slack is currently fragmenting the communities who believe in openness so I have heard friends who are feeling pained and depressed by that current development.
As I see it there is no longer any dichotomy between flow vs purity with the alternatives mentioned above, like with the possibility @sylvain presented.
Also, https://Riot.im basically looks and feels like Slack and it is even easier to get guests / contributors on board. Rocket.chat seems ot integrate with Matrix as well, just another client since it is open and integrating with Matrix.
Thank you @matt for being so empathetic and understanding This is what I love in this community! I do agree with all what you said, happy to see what would be more aligned and take it low pace. I’m a bit stressed and trying to make the project work in France, sorry if I sounded a bit “closed” to opening that discussion, that’s not what I meant
hi all. completely agree with @matt in spirit and @oeoeaio and @MyriamBoure in practice. we’re all a bit drowning at the moment and contemplating attempting to move everyone from a channel that’s actually working really well hurts. But all for it in the longer term if/when is headspace and someone to take it on as a project, to set it all up and manage everyone across with minimal disruption - especially if is to something that is similar enough to reduce / remove learning curve.
I have another project that is just starting to use slack and am open to trying something else. @matt if you’re able to make your clearest recommendation on the closest to slack open option, preferably that doesn’t require me doing anything overly technical, I’m open to giving it a go (@serenity thinking we could move farmplanning)
You’re so welcome Myriam, it’s my duty to serve the community in a friendly manner. True pleasure. Empathic and deep level listening is key to healthy relationships and organisations in my world! And it is deeply connected to openness and curiosity.
Great, happy to see it happen in low pace. I feel with you and find myself in a similar situation currently developing and delivering projects until start of 2017. That’s yet another reason to why I find it important to find a sustainable and holistic solution through a well-function and inclusive process that is aligned with OFN. Don’t worry, written communication is tricky so I tried to mind the gap about your reply and see it from your perspective.
My reasoning for the communication is that one of OFN’s best and Unique Sales Point is openness and transparency. That is an important differentiator and opens up for a “blue ocean” in the food system. This in comparison to the competitive “red ocean” of many proprietary supply systems can be a big thing for people and producers.
Hi Kirsten,
Thanks for your spiritual support! I have full understanding for the practical bit. As mentioned above let’s take it slow pace both on a decision and process now in hectic times.
To have a project leader on the migration process is definitely a great point.!
Currently my recommendations list in general based on my experience & research (links provided in previous posts) is as follows:
XMPP (technically most secure, open and the true open standard)¹
Riot & Matrix (2nd most open and secure, perhaps most “user-friendly” but not an established standard)
Rocket.Chat
Mattermost
Zulip
I don’t want to include Slack as a recommendation on the list because of previous reasoning, which can be complemented if necessary…
¹ Everyone using e.g. Facebook & Google chat are using XMPP which is running under the hood as I have understood it.
For info I’m using rocket chat at the moment for another project. It’s ok and could be a switch option, but doesn’t have the same Slack UX in my opinion
Great to hear! I am using https://riot.im which UI & UX feels a lot like Slack, I can really recommend trying it. It’s not as glossy but glossy does not equal good. It gets stuff done! Keep trying stuff and we’ll find something out!
I recommend checking this with the most up to date functional and technical classifications about real-time communication tools Digital Communications Protocols - Google Sheets I obviously root for Matrix as you can notice since a while back in this thread.
I obviously root for Matrix as you can notice since a while back in this thread.