XSF Discussion - 2020-02-04

yay

a colleage brought a proper XMPP sticker from fosdem

I can finally get rid of my ugly square one

jonas’: 🤘

when one googles for "xep-0391" google thinks it was published in 1969... because the date in the example is 1969-07-21...

Hah

Having looked into this, I'd strongly recommend staying away from it if you value your sanity.

Oh neither of Google and DuckDuckGo understands the dates on prosody.im still? I thoguht I got at least one of them to understand 😞

oh yes 0202 (entity time) is even better. the publication date is right (2009), then continues with a date from 2006 and a random paragraph about jabber:iq:time. Clearly googlebot is drunk

IIRC there's no sane way to set a 'last modified' time on a web page.

Most of the various ways you can set a date is "created" or similar

fippo: I think it just finds the very first full ISO timestamp (with a T in it), which for XEP-0202 is indeed 2006-12-19T17:58:35Z. And it is not a random paragraph, is it? It is the first paragraph of the main text?

Heh, relevant reading made it to the top of HN: https://tools.ietf.org/html/draft-wkumari-not-a-draft-06

heh

the SCAM work team has taken op the challenge to generate some leaflets, folders an the like that provide information about XMPP. For starters, we'll re-use content from the website. I'd like everyone to review the text on the website to see if the content is still relevant/up-to-date. Of particular interest: https://xmpp.org/about/technology-overview.html https://xmpp.org/about/myths.html https://xmpp.org/about/faq.html but maybe other pages too.

I welcome your PR's 🙂

(thank you)

If we had team mentions, I'd tag @commteam.

But I'll have another look myself, too. Great idea.

oh hey, a mention!

what was that about?

nyco: just what was written above by Guus about leaflets.

Today in the news: MLS encumbered? https://datatracker.ietf.org/ipr/4015/

oops

I remember that (likely) this patent, Qrypt Inc. and MLS where on HN(?) a while ago

flow: interesting, I haven't time nor energy to read the patent right now, but if it is patenting double ratchet in 2016 then it seems to be quite futile to me...

winfried, I think the relevant part is the epoch key exchange (instead of with every message like signal/omemo/olm do)

They do reference the signal protocol as prior art in the field, so copycats of signal protocol like omemo/olm don't need to care that patent

well that's clearly worse than GPL

Does MLS use epoch keys? (still no time nor energy to confirm myself)

moparisthebest, if I understand Dave's mail correctly, he agrees there are no GPL issues with omemo per se. he seems to be fine with omemo being based on signal protocol as long as there is documentation of how to use olm for that (so there are easy instructions how to do omemo without using gpl libraries) - that however can't be a requirement for a XEP in general IMO (it should be perfectly fine if there is only a GPL implementation as long as the only reason there is no non-gpl implementation is the fact that nobody wanted to write one)

winfried, that's what I understood, but I also haven't read the current spec

sorry, I was mainly joking that "only GPL users are guaranteed legally to be able to use this spec" is clearly a better situation than "some company owns a patent on it and no one can use it"

moparisthebest, usually you can still use patented stuff as long as you don't do it commercially, but yeah I agree patents are worse than GPL in general

flow: you mean this one? https://news.ycombinator.com/item?id=21954006

larma: the commercial use does make it an issue for us. Unsure about the IETF.

Some discussion around it: https://mobile.twitter.com/AlecMuffett/status/1213356702399115267

is there a definition of "commercial" somewhere btw

Hey, an ex-colleague

not one across all jurisdictions, of course patents don't apply across all either

to know in what context we're using it

pep., the problem is in the phrase "as long as you don't". Everything that follows is an incumbrance.

https://xmpp.org/about/technology-overview.html links to https://xmpp.org/rfcs/rfc7622.html which returns 404 ...

That's likely to be true for more links to the RFCs then. ✏

https://xmpp.org/rfcs/rfc6120.html and 6121 work ...

why do we not link to the ietf?

could probably work up a rewrite to 302 to the ietf on any of those

pep.: Probably to be 100% sure the link won't go dead? ;-)

and how's that working out :)

We never hosted 7622 in the first place.

Only 3920, 3921, 3922, 3923, 4622, 4854, 5122, 6120, 6121, and 6122.

Well we link it.

Indeed we do.

For those not in summit, does anyone have a fever, cough, bad chest etc?

He means: for everyone that is not in the summit muc, but attended the XMPP summit and/or FOSDEM in Brussels last week.

That, yes.

dwd, something something virus?

well, I do have a cough but unless webex now transmits diseases, that’s probably unrelated.

nope you only get the DRM disease from that

pep., I have potential 2019 Novel Coronavirus. It's fun.

Interesting.

Tbh I was expecting some kind of epidemic at FOSDEM :)

(#notspreadingfear)

oh dear

dwd, I hope you get well soon

I shouldn’t have talked about how to get rid of europes hackers by poisining mate at conferences too loudly in the other room, it seems /conspiracies

It happened.

Well, thankfully I'm informed I'm very low risk.

Once more ;)

Also, I believe this weekend was the start of the 'regular' flu season in Belgium, so there's that.

Maybe it's because of FOSDEM? :P