XSF Discussion - 2024-02-26

Hello. Last week, we discussed here about a XEP that i'm going to submit. I have a question: is there any policy concerning author nicknames? I did not find any information about that. John Livingston is actually a nickname, and i don't know if i can submit with this nickname as author, or if i have to use my real name? (and even if nicknames are allowed, i still hesitate... nobody knows my under my real name...)

The closest that I can find on this is in XEP-0001, which states: > Author Information -- first name, last name, email address, and Jabber ID are all required and must be provided for all authors (to simplify the text we use singular "author" in the remainder of this document, with the understanding that there can be multiple authors)

Yeah there isn't much justification

It doesn't say anything explicitly about the 'correctness' of the supplied data.

Legal question, so for Board to get an answer to, I think. Absent legal advice I’d certainly be uneasy at the idea of doing “legal things” (like copyright assignment) under pseudonyms.

That's how I feel too. If there's an important reason to allow for pseudonyms, perhaps we can think of a way to allow for pseudonyms used publicly while for example the Secretary keeps a private register of pseudonym-to-real-name mapping?

Maybe cross that bridge only when we get there, and if we need to go there.

> That's how I feel too. If there's an important reason to allow for pseudonyms, perhaps we can think of a way to allow for pseudonyms used publicly while for example the Secretary keeps a private register of pseudonym-to-real-name mapping? Or a public key? Who want to be anonymous can provide a public key, and incur the risk that if anyone steals the private key then they lost their idea ✏ ↺

> Legal question, so for Board to get an answer to, I think. Absent legal advice I’d certainly be uneasy at the idea of doing “legal things” (like copyright assignment) under pseudonyms. I've asked that in the past just to get a "meh"..

I don't recall this coming up before, but I recall membership has come up a few times, to which the answer is not "meh", but that we need legal names

> Maybe cross that bridge only when we get there, and if we need to go there. We've long crossed it, just people in here don't care much

i dont understand why xsf needs to do copyright in the first place personally. it makes everything more complicated

However it's not clear to me that those names need to be public, as we have traditionally done

There's plenty of reasons why someone would want not to use their legal names

pep., I agree

MSavoritias fae.ve, it does have advantages, like the original author not being able to turn cloak and decide that everything should be rm -rf'd.

not to mention that to redistribute e.g. XEPs, we *do* need the approval of the copyright holder.

i wasnt thinking of leaving the copyright with the person that wrote it

i was thinking just having the documents be public domain

define public domain.

and you still need assignment to the "public domain" (whatever that means) by the contributors

cc0 or https://unlicense.org/

MSavoritias fae.ve, not all jurisdictions have the concept of "public domain", so that can't be relied upon

> MSavoritias fae.ve, not all jurisdictions have the concept of "public domain", so that can't be relied upon for sure. raw public domain doesnt work

Whenever public domain (or cc0) wouldn't be available/valid I guess the xsf's licence may not be either anyway ✏

CC0 could work

you do need the assignment still but probably you dont to publicize the names no? you can just submit a xep under public domain

not a lawyer of coure and stuff

Seems to me the important bit is the assignment, not the name at the top of the XEP.

> not to mention that to redistribute e.g. XEPs, we *do* need the approval of the copyright holder. By reading XEP-0001 and other documents, i understood that we transfer the ownership of the XEP to the XSF... so... The only issue, is to be sure that the real author transfers the rights to the XSF.

"The developer submits a specification to the XMPP Extensions Editor [11] and agrees to transfer ownership over the protocol (but not implementations thereof) to the XSF."

(XEP 0001 intro)

I'm not denying that there might be good reasons for people to wish to be anonymous - but for a consumer of documentation, having an anonymous author does not necessarily inspire confidence.

the only moments I've ever even looked at the author names was to figure out whether it's plausible that the authorship could be handed over for inactivity.

I think I've only looked at authors for XSF stuff, rather than consuming a XEP.

I'm looking for author names to find out who to ask questions. But, if I'd read a XEP and notice that an author name is obviously a pseudonym, I'd pause.

Guus, and if it's not obvious? Who cares?

If you can contact the author and they respond, I think that's what matters

It seems that way to me.

also, nobody ever asked for my ID ;)

you'll never know who I truly am!

Indeed

I do care, to some extend. To me, explicit secrecy instills a sense of distrust.

Guus, doesn't have to be secrecy, ftr

In my case, you will never find me under my real name... Only under my nickname (which is the name of my individual company btw) ✏

e.g. Anglification for convenience?

secrecy is not the same as privacy

or mismatch between gender identity and legal name

or that

(I think the right word is anglicisation, but I like my version better :) )

Oh, this isn't a hill for me to die on at all - but I do want to point out that allowing pseudonyms also comes with some downsides - at least to me. That doesn't need to be a reason to reject using them at all, as far as I'm concerned.

haha

> makes mental note to always use a convincing-sounding realname-like name when trying to social engineer Guus "Kev" seems to have worked fine for me :)

well, yes. That's why no-one tries to social engineer anyone with an unbelievable sounding name. There _is_ value in this.

why would you want to ask the author of the xep for help tho? isnt the xep and implementation supposed to be complete enough so that the author is not needed

i mean thats why we write the xep no?

I've had people ask me questions about XEPs I've written, FWIW.

MSavoritias fae.ve, sometimes you need to clarify stuff if it turns out there are more than one interpretation to a paragraph for instance. ✏

hmm ok

imho the IPR re-assignment needs to have a solid legal base so implementors have a clear legal situation. just because we haven’t checked IDs or other official documents in the past doesn’t necessarily mean we should weaken that process even more

I think i will submit my xep with my nickname, and clearly say in the email that it is a nickname, and i'm ok to change if required. So that XSF member can discuss and decide.

I understand that people want to remain anonymous. but when I implement a XEP I want the license situation to be clear

John Livingston: if this is important for you, then please do, because then it immediately is a relevant topic to address. If it's merely a way for you to force a debate, please consider that we're already thin on resources.

i don't want to force a debate

ok, good enough for me.

I thought this topic was already discussed. I see that it is not as simple as i thought.

All that'll happen is that the Editor will put 'needs Board' on the PR after you open it, and then push it up the food chain.

Which seems sensible for all parties involved, I think?

I think there should be some decision one day. But if you are thin on resources, i will adapt, i don't want to add you extra work load

I'm not convinced it's worth the XSF's resources to get legal advice on it, but *shrug*. Board can decide whether to get legal advice, reject it outright, or call it a non-issue.

If Board call it a non-issue, it's an extra four clicks for the Editor or whatever, so NBD.

If board was to call on this, it might be good to update existing documentation to reflect the outcome.

Wouldn't it be weird that following the "maintaining a XEP" guide (https://xmpp.org/extensions/xep-0143.html#maintain), my commits will be from my github account (where i use my nickname), if the original XEP has my realname as author? How will you check that i am the real author??? ✏

I think the assumption has been that when people assert that they are who they say they are and that they agree to the IPR policy, that they're not going to commit fraud, but ... *shrug*. I just follow the process :)

There's been instances where updates to XEPs have been flagged as 'need author', right? So, it might be possible that your github account triggers Editor to ask your email account for verification that you as an author are OK with the change, leading you to respond "well, that actually was me in the first place".

seems complicated...

I'm just trying to say that it will probably automatically work out.

yes, but this is not optimal, and will require some energy for everyone :/

I used to put github handles in changelog entries in order to match things up

Significantly less energy than having this discussion, I think. :D

Guus, as former editor, disagree.

email roundtrips are the worst.

I stand corrected.

Could the fact that "John Livingston" is the official name of my company change anything to this copyright issue? ✏

we might consider extending the ent file to include the github handle

That will definitively doxx me.

I see

but won't the same happen if you link your github handle to a protoxep submission with the other name on it?

I have some reason to not make it too easy to associate my nickname and my real name

oh, I don't doubt that, trying to find solutions here.

(possibly create a secondary github account is a way to work around those?)

(yes, but this means give my real name to microsoft...)

Didn't you already do that anyway/

if you have a primary github account?

(I'm also not sure if you need to provide your real name when signing up to Github, but am totally willing to accept that it is)

Hum, possible that i already gave it, when i registered to github sponso. I have to check. ✏

Oh, with Github Sponsors, money suddenly is involved. That immediately makes it more likely to me that some kind of identify verification applies.

Thanks all for your thoughts on this subject. I will think it some more before submitting.

> Oh, with Github Sponsors, money suddenly is involved. That immediately makes it more likely to me that some kind of identify verification applies. As i said, my work is only known under my nickname. Even NlNet (one of my sponsor) uses my nickname for public messages (and they know my real name for the financial stuff).

they accept that? that makes me hopeful for me too then :D

MSavoritias fae.ve, yes! And they ask if they can (or can not) use your real name.

thats amazing! nlnet just became a lot nicer than i thought

there are really nice people

+1, they're great :)

MattJ - as a Boardish person, what do you think is sensible to do here as a first step?

I think John should just submit the proposal, and we figure stuff out from there (personally, I don't think there is anything special we need to do, but Board can discuss it)

I suspect at most we would need a second (private) confirmation of copyright assignment, but I doubt we need to go that far

Ta.

MattJ: should i precise in my submission mail that John Livingston is a nickname?

No need

Yes please.

:D

Go with Kev's answer :)

Means I know Editor needs to get Board to say it's OK.

Fair enough

I'm fine with Board then saying "Of course it's ok", but I think legal liability questions belong with Board, not with me.

From my perspective, the only reason this is an issue is because John Livingston (kindly) brought it up. Otherwise we would have accepted it without question. That will always be the case unless we start demanding identity proof along with XEP submissions, which is rather excessive if you ask me.

I'm also fine with a direction from Board to Editor to not care about such things :)

^^'

I think it would be a bigger problem if John Livingston was not the actual author of the document

As things stand at the moment, the Editor has Authors confirm that they're good with the IPR policy, which involves (I thought) their name, but maybe I misremember.

(It also has them assert they're the actual author, unless I greatly misremember)

Would we have acted if John Livingston had chosen to identify themselves as Johnny the Magnificent though?

Guus, probably. Whether rightly or not, I don't know. Most names can be valid legal names :)

https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-believe-about-names/ obligatory

Which is why your argument of "it doesn't give me confidence in the document if it's clearly a pseudonym" is problematic, because there isn't a reasonable "is clearly a pseudonym" test.

MattJ In the UK probably, but in DE laws are horribly restrictive.

Yeah, quite the opposite in the UK

MattJ: problematic or not, it does affect things for me, and plausibly other people, even if there's no accurate way to test for it

But I'm coming over a lot stronger than what I actually feel over this

Sure. I understand what you're saying, but it's your problem and not the XEP author's if you don't think their name is legitimate :)

I'm happy to accept that there are good reasons for the XSF to allow for pseudonyms.

Well, to an extent it is the XSFs problem if their published documents are seen as having untrustworthy data.

What I'm saying is that there is no reasonable definition of "pseudonym", so it's not something we can enforce

Untrustworthy is the wrong word here, but I hope you understand my meaning

I can legally change my name to "MattJ The Magnificent" tomorrow, and submit a XEP, and you would prefer it not to be published?

I think that it would reflect not great about you and the XSF, yes.

Don't tempt me!

It's fully your right to do so, obviously, but it doesn't look confidence inspiring to me

> Don't tempt me! I promise to buy you a hoodie with your new name if you do!

:D

There is a(nother unquantifiable?) distinction between the level of confidence inspiration between names like "MattJ The Magnificent and Fourth Bearer of His Name" and "John Livingston" though.

the former clearly being more confidence inspiring. ✏

only missing Protector of the Realm etc. etc.

My lack of recent D&D experience is showing.

(that was more of a ASoIaF reference)

Kings & Queens, Undead, Dragons - that's basically D&D.

D&D is more hi-magic though.

the only monsters, apart from the obvious undead, in ASoIaF are the humans.

'apart from the obvious undead' <-- isn't that _the_ major plot element? :D

(i've not read the books, just the subtitling on TV...)

I'll note that many projects and charities accept copyright assignments without real name, as was said so long as it's "really the real author" doing the assignment then legal intent is present. In my country it's still legal to sign documents with an X so names hardly enter into these things as often as we think. However of course I defer to the lawyers among us if there is an opinion

*shuffling sound as the lawyers scramble for the doors*

Unless the XSF is verifying identities then there is no assurance that author names are legally identifiable individuals; so the judgement is purely on the basis of 'looks convincing.' In that case "John Livingston" wouldn't have raised any suspicion had they not mentioned the possibility.

Copyright, however, needs an identifiable individual; whether that requires a legal name or a unique pseudonym would be adequate is debatable. [IANAL] I suspect a judgement would come down to whether the individual could prove their authorship under the published name/pseudonym - which again comes back to "does the XSF verify identities and, if not, there is no assurance any author's name is a legal name."

As precedent, see the author names published in XEP-0419 (the fact it's a 'humorous' XEP doesn't change the fact that false identities are used.)

The XSF is certainly not qualified to verify identities anyway

I'm not suggesting it can or should; the point is that requiring legal names is meaningless without verification, so it all comes down to whether the name 'looks legitimate' ✏

No yeah I certainly agree. And as developers we all know everything looks legitimate, as it's been linked above :)

https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-believe-about-names/

According Italian copyright act states that "The pseudonym, stage name, acronym or conventional sign, which are known to be equivalent to the real name, shall count as a name". So, you can use a nickname. Have to check how is in other jurisdictions. Tomorrow if I have the time I will do a search. ✏

In any case the person who do not want to appear with her/his real name on a public web page or other, may communicate it in private to the XSF. This may be a good solution in many jurisdictions. ✏