Free software tools for staying in touch

They just apparently spent their money on building an enterprise-worthy app instead of marketing.

https://help.webex.com/en-us/n80v1rcb/Cisco-Webex-Available-...

Also, what is your basis for saying Zoom has the best performance?

I would've assumed they draw upon the same cloud resources; but based on our experience: 1. Client Webex has been basically unusable since day 1 of 100% remote work (~3 weeks ago) 2. Webex through my employer account works 98% of the time for 85% of the people.

I don't know where the bottleneck is with our client's webex - do they have lower tier account (unlikely), or are there some self hosted aspects or authentication aspects that are giving us trouble.

But my lesson learned is - different people will have drastically different experience with a tool for unpredictable / confounding variable reasons :/

But again, the point of the article isn't about which software is free as in cost or easiest to use. It's about taking an ethical stance despite sacrificing things like a pretty UI.

Mumble: Mumble is amazing. I used it extensively playing EVE Online and what can be done with mumble with regards to external authentication and integration is really great. As an enterprise solution, if I have to weigh my time versus expenditure, it's not worth it though, I'll go with MS Teams or Slack.

Asterisk/SIP: I actually ran an asterisk PBX with a self made IVR solution for a project at my previous employer and handled a lot of calls with it and saved a load of money compared to what the callcenter would have charged. Asterisk is great but we are back to the time/expenditure calculation again. I can buy a PBX solution and pay someone to maintain it for far less then me setting up Asterisk. I'd rather just buy a PBX and a support contract. For IVR the calculation gets more complicated.

Jitsi: Jitsi meet is really great. It's right in that sweetspot where Skype begins to fail for decent conferences (>4 people) and where I'm invested that much that I want someone to blame and a service contract if my meetings get large enough. If you exceed that threshold (in my testing it has been somewhere between 15 and 30 participants) you want to pay someone for this service or run your own server/videobridge (still curious why BigBlueButton isn't mentioned here anywhere in all these Zoom topics). As an educational institution with my amount of users, hosting a videobridge or bbb the hosting budget alone would have exceeded our Zoom licensing cost not to mention my time. At least in Germany there is also the issue with not having a contract with the provider and problems this creates with regard to the GDPR.

Jami: I don't actually know this one.

OBS: Nothing can touch OBS. It's an incredibly flexible tool and while it could use some usability improvements it just really is that great. With it's usability downfalls it's also not something you can suggest to a casual user, which is kinda sad. Someone please figure out whatever Zoom is doing with the virtual camera background. Setting a simple picture as background without a green screen frankly is one of their killer features. No one talks bout it but so many people are using this that it has got to be one of their killer features.

XMPP: Is a weird mention for me, since it's not an application but a protocol. Jabber sucks, most XMPP clients suck for normal users. I'm not aware of any decent enterprise solutions.

IRC: Well, IRC is fucking great. I spend so much time on IRC in my youth and I'm fairly certain my earliest linux contact was an IRC bouncer on a friends machine. In the end it's kind of a weird mention as well though, since it's a protocol and not an application. It is also a protocol that is kinda caught between the "I want link preview, VoIP, file sharing, whatever" crowd and the "why is that noobs client spamming my channel with un-interpretable messages" crowd. IRC with a spec upgrade, even if it would induce a permanent netsplit would be great.

I'm not gonna talk about the rest. Nextcloud is great, and with Univentions app server (https://www.univention.de/) is a usable solution that just suffers at scale. Of course I could set up a univention server and nextcloud and imap email and ldap and lots of other things mentioned above but it would cost me a significant amount of money and time and I can have Office 365 basically for free. So, considering I'm spending tax payers money, and I'm in the EU and MS claims to be GDPR compliant... what is my incentive to do any of this?

If I'd have one complaint about Zoom, I'd would be that we had one student unable to join from Iran (which is either because of his ISP blocking certain connections or Zoom actively blocking users from Iran). Zoom excludes users from Iran in their ToU but we assumed that would just be for paying users and have a currently open ticket to figure out what could be the issue.

So, this is my take. Basically if you have the time and money and want to tinker most of these OSS solutions are great and worth it. If you are operating within budget or time constraints there a valid reasons to choose a commercial provider.

GNU seems to assume that average users are willing to tinker and struggle until they get something to work. Average users won't do this. If users are demanding things like animations and transitions, GNU software which can't even get onboarding right will always struggle.

Personal experience:

Bad: Mumble is a nightmare to admin. It has confusing popups and so forth when joining a server. UX is archaic.

Bad: OBS is the wrong tool for the job (however amazing it is when used for the right job). UX is archaic.

Bad: I've always been pretty bullish on XMPP. It's a really great solution that's still looking for a problem to solve. That being said, I have yet to find a client that has the polish of modern chat apps.

Bad: IRC. I still fail to log into freenet some 40% of the time, and I'm an technical user. NICKSERV is a complete DISERVice to the platform. Having to switch to alternate nicks if you disconnect is bizarre.

Middleground: Riot (not mentioned) needs UX work, but comes close.

Middleground: Installed Jami. The onboarding is uncharacteristically competent. Haven't used it for a call.

Good: Mattermost (not mentioned) has a daft name (which users care about!), but competes with Slack pretty nicely.

Good: Discourse is pretty damned incredible.

Good: Etherpad, but has limited use for average users.

The reason is that people do not generally think philosophically or strategically about the software choices they make. The FSF does and, time and time again, their deep consideration has proven prescient. It will be the case with Zoom too.

That's really a weak argument: bad is not better than unknown. And in this case unknown is not unknown.

Jisti is OSS, has been reviewed as such, and has been around since way before zoom was even invented.

* Traditional Web Servers (Apache 2 or nginx)

* https/ssl

* modern web browsers

Running a videoconference software that just works in your browser over https solves a large majority of security concerns. The video and voice is encrypted over SSL between the clients and the server. It's not end to end encrypted because that's not really possible, but the only place its unencrypted is in your browser and on the server you own and control. If you don't trust Jitsi's ability to password protect rooms, you could put normal http password authentication in front of your jitsi install -- which would break the mobile app but work just fine in your browser.

Lots of the concerns of Zoom's security are resolved by using readily available standards and software instead of depending on proprietary black box solutions that turn out to send encryption keys through china.

EDIT: As kardos notes below the video actually runs over udp on port 10000 so it is not encrypted over https. It looks like video is encrypted[1] but I'm unclear what method it's encrypted.

1. See Security section of Readme https://github.com/jitsi/jitsi-meet

To get full use of it back then, using high-quality hardware SIP phones was a necessity. These days I think that in-browser phones will work very well for most people.

Interconnection with telcos went from unsanctioned to orderable to being widely marketed during that time. If you know what you're doing you can save a lot of money. If you would prefer to have someone else do the work, the virtual PBX vendors are pretty good.