Re: WebID History - is also: Webid Editor/Author issue

On 1 June 2013 18:45, Henry Story <henry.story@bblfish.net> wrote:

>
> On 1 Jun 2013, at 18:36, Melvin Carvalho <melvincarvalho@gmail.com> wrote:
>
>
>
>
> On 1 June 2013 18:26, Henry Story <henry.story@bblfish.net> wrote:
>
>>
>> On 1 Jun 2013, at 17:41, Melvin Carvalho <melvincarvalho@gmail.com>
>> wrote:
>>
>>
>>
>>
>> On 1 June 2013 17:02, Henry Story <henry.story@bblfish.net> wrote:
>>
>>>
>>> On 1 Jun 2013, at 15:18, Melvin Carvalho <melvincarvalho@gmail.com>
>>> wrote:
>>>
>>>
>>>
>>>
>>> On 1 June 2013 14:53, Henry Story <henry.story@bblfish.net> wrote:
>>>
>>>>
>>>> On 1 Jun 2013, at 14:22, Melvin Carvalho <melvincarvalho@gmail.com>
>>>> wrote:
>>>>
>>>>
>>>>
>>>>
>>>> On 1 June 2013 13:32, Henry Story <henry.story@bblfish.net> wrote:
>>>>
>>>>>
>>>>> On 1 Jun 2013, at 13:20, "Andreas Kuckartz" <A.Kuckartz@ping.de>
>>>>> wrote:
>>>>>
>>>>> > Henry Story:
>>>>> >>  According to the criteria you put forward below for being author
>>>>> of a spec,
>>>>> >> in the mail to which this is a reply you wrote:
>>>>> >>
>>>>> >>> You are an editor of the document if you've contributed at least
>>>>> 25% of
>>>>> >>> the bulk of the edits to the specification (modulo obvious
>>>>> >>> search/replace/bulk copying changes).
>>>>> >>
>>>>> >>
>>>>> >> So according to your own criteria, you cannot be an author of this
>>>>> spec
>>>>> >> either.
>>>>> >
>>>>> > I do not understand this. "author" != "editor".
>>>>>
>>>>> Ah yes.
>>>>>
>>>>> The next paragraph in Manu's e-mail was the relevant one
>>>>>
>>>>> > You are an author of the document if you've contributed foundational
>>>>> > ideas, arguments, or spec text to the body of the document.
>>>>>
>>>>> And I don't believe Manu satisfies those criteria.
>>>>> The ideas and arguments had been published in academic journals and on
>>>>> the W3c site before. And as Manu wrote:
>>>>>
>>>>> > I decided to gather much of the writing across the Web and put it
>>>>> into a
>>>>> > coherent specification that would then be used to pitch a WebID
>>>>> Working
>>>>> > Group at W3C.
>>>>>
>>>>> so the spec text is probably a lot of text I wrote out there on the
>>>>> subject
>>>>> in wikis and blog posts.
>>>>>
>>>>
>>>> What it boils down to is whether manu contributed foundational
>>>> arguments, ideas or spec text to the original spec.
>>>>
>>>> Given that he WROTE the first spec, and contributed many edits, it's a
>>>> difficult position to take that he provided none of these.
>>>>
>>>>
>>>> Well that is what I disagree with, and which I have always found quite
>>>> rude in Manu's position.
>>>> He formatted  the work we had done in the prevous two years into a w3c
>>>> format, and never
>>>> mentioned the previous work, which was published on the W3C web site.
>>>>
>>>>   http://www.w3.org/2008/09/msnws/papers/foaf+ssl.html
>>>>
>>>
>>> This version certainly does mention the previous paper
>>>
>>> http://www.w3.org/2005/Incubator/webid/spec/drafts/ED-webid-20100711/
>>>
>>>
>>> yes, I added that later to keep track of the history. It should of
>>> course also be added
>>> at the end of the Change History, since parts of the spec were directly
>>> taken from that document.
>>>
>>>
>>>
>>>>
>>>> Just compare how close Manu's "foundational" text here is to the above:
>>>>
>>>>   https://dvcs.w3.org/hg/WebID/raw-file/25ba7f596f07/index-respec.html
>>>>
>>>> You will see the first two paragraphs just are copied straight out of
>>>> that paper.
>>>> Manu's work just is _not_  foundational of WebID.
>>>>
>>>> Had he joined the WebID group and continued to contribute as an editor
>>>> then the continued work over time would have been something to seriously
>>>> take into consideration.
>>>>
>>>> Anyway now that I understand the role of the author, we have very good
>>>> reason to
>>>> keep Bruno Harbulot, and others who were present in the early workings
>>>> out of WebID.
>>>> I really appreciate those days when we had tightly argued discussions
>>>> on the
>>>> mailing list, and we made huge leaps and bounds of progress.
>>>>
>>>>
>>>> If you definitively feel that he should not be there, then he's said he
>>>> wont contest it, but it does not do wonder in terms of "communications and
>>>> image" for webid.
>>>>
>>>>
>>>> What do you think it does to the image of WebID if someone whose name
>>>> is as an author,
>>>> goes around attacking webid with bad arguments?
>>>>
>>>
>>> Because it would be better to do in private, if you felt strongly about
>>> it.  Consider how this thread would appear to a newcomer to the group, or
>>> someone with a casual interest.
>>>
>>>
>>> I agree that it is better to do things in private like this, but Manu's
>>> attack on WebID on the RWW mailing list
>>> was in public, and so was the intiation of this thread.
>>>
>>
>> It was not an 'attack', Manu was asked why he did not use WebID for
>> payments and he listed good points and bad points, according to his
>> understanding, and showed the gaps his work covered.  Where his knowledge
>> was out of date, he corrected his points.  It doesnt help that the latest
>> specs are hard to find, nor linked to the main webid site.
>>
>>
>> concerning hard to find, I asked if someone would like to improve the
>> file that currently does
>> an HTTP redirect to point instead to the two files. I am very busy on the
>> LDP group and other
>> things.
>>
>
> But it's hard even to find that file, eg though a search engine like
> google.
>
>
> We need to do more. I am working on that...
>

We split webid into identity and auth specs at the start of the year.  It's
June now.

Yet : http://webid.info/spec still couples identity with TLS, and this was
a link you yourself sent out last week to explain WebID

This was exactly one of the critcisms manu raised that you claimed were out
of date.  Is there any wonder that people are getting the wrong
impression?

Note this very same objection came up from friendica on the FSW list last
week too.

I know you are busy with LDP, but if we can keep the normative spec pages
up to date, we have a better chance of making people understand what WebID
is.


>
>
>
>>
>>
>> Standards live by peer review.  You criticized his work as centralized
>> and/or insecure, without reviewing it.  I suspect X.509 would be highly
>> susceptible to brute force password attacks on a compromised machine, so
>> it's hard to claim the higher ground here without looking at the work.
>>
>>
>> I am not against criticism at all. We live by it all the time here.
>> My problem is that Manu is listed as an Author of the spec, and that this
>> creates confusion
>> when it turns out that he has not participated in the group for close to
>> 3 years, and his criticism is
>> out of date. It gives his word a disproportionate weight.
>>
>> I left Manu as an author for a long time, because I was hoping he'd come
>> back and participate.
>> But his clearly expressed feeling is that WebID went the wrong way. So I
>> don't really see why
>> he wishes to continue to be associated so strongly with this work, or why
>> he should be positioned
>> higher with more visiblity than others who put a lot more work into this.
>>
>> I think we need to move on, and get to a WG soon. The WebID identity spec
>> and the WebID
>> over TLS spec have remained very simple, and I would not be surprised
>> they can work well
>> with Manu's web keys: probably through work on the cert ontology, which
>> could be developed.
>>
>> At the same time it is easy to make things much more complicated here
>> than needed. WebID
>> is just one element in a stack that consists of LDP and WebAccessControl,
>> PingBack, and more.
>> The specification of WebID is in my view nearly finished: adding support
>> for other other crypto
>> types as Dominik  Tomaszuk has done is perhaps the only thing left to do.
>> And just doing that
>> well can be a lot of work.  Mind you we also have to build test suites
>> and other things and
>> build good demos. Then in a year or so Browser Crypto will start coming
>> out, and then some
>> of Manu's ideas will work perfectly with WebID, without requiring TLS in
>> JS.
>>
>> It is incredibly difficult to keep a spec simple. Specs like OAuth have
>> had huge trouble adding
>> layers on layers of complexity. I am happy to say that this is not what
>> hapened here.
>>
>>
>>>> I'd suggest leaving all the authors in place, for the time being, and
>>>> trying to understand the new work Manu has done with WebIDs.  We all want
>>>> secure encryption, signatures, and payments, right?  Also take time to
>>>> explain to people how WebID has improved in the last 2 years, many
>>>> (including dan brickley, inventor of FOAF) have asked for you to continue
>>>> your excellent blogging series that got people interested in WebID in the
>>>> first place.
>>>>
>>>> Why not let things cool down a bit.  And if at some point in the future
>>>> you feel that Manu's contribution was not 'foundational', talk to him in
>>>> person about it, and change the spec accordingly.
>>>>
>>>>
>>>> Look I understand Manu's role. And I am thankful for having given us
>>>> the
>>>> impetus to start a process of getting us to start using the W3C
>>>> templates
>>>> to write up a spec. But I don't think that is powerful enough to get
>>>> authorship role.
>>>>
>>>> Le me remind you of Coralie Mercier's pointer
>>>>  http://www.w3.org/Signature/Contributor.html
>>>>
>>>> Authors by their own initiative or through commitments to the Chairs
>>>> make substantive contributions that are included within the specification.
>>>> Frequently an author will make and write a proposal that is then the basis
>>>> of a section of the specification. Criteria for authorship are the
>>>> expressed interest (agreed to by the Chairs) to be listed as an author and
>>>> the substance and quality of the contributions. The Chairs look at the
>>>> consistency of participation, the willingness to take action items, and how
>>>> much "authoring" the WG member actually accomplished. This criteria is
>>>> somewhat relative in that if this role is designated, the Chairs wish to
>>>> list the top handful of people that consistently plugged away on the work
>>>> while avoid a list of names occupying the first two pages of the
>>>> specification. Where the number of authors/editors are small, the Author
>>>> and Editor role is frequently collapsed in to the Editor designation. Where
>>>> there are numerous authors, the role will be a specified subset of the
>>>> Contributor designation which is an Appendix to the specification.
>>>>
>>>>
>>>> But Manu certainly is an important contributor. Perhaps there are ways
>>>> of acknowledging
>>>> contributions more carefully.
>>>>
>>>
>>> At the end of the XG we had a conf call with Timbl.  He said that to
>>> take webid to a WG we needed to write a charter.  That was the next key
>>> action for the XG.  Timbl said that writing a charter is a substantial
>>> amount of work and that sometimes the result of a whole XG is simply to
>>> create a charter.
>>>
>>> I consider writing a spec to be also challenging in this regard.  Manu
>>> took a huge number of inputs curated and edited them and presented them in
>>> a way that partially survives today.
>>>
>>>
>>> That is indeed the case. But we started writing blog posts, then wrote
>>> academic papers, then went to
>>> W3C conferences with position papers and did a lot of work that is
>>> pretty much indistinguishable from
>>> spec writing.
>>>
>>> Although much of the document was taken from previous sources, it's hard
>>> to claim that there was no original text at all.  e.g. I dont recall seeing
>>> the term "verification agent" which is used in todays spec previously, but
>>> there are numerous other edits.
>>>
>>>
>>> Nobody is claiming that there was nothing brought to the table at all. I
>>> have not removed Manu from the
>>> contributors list either.
>>>
>>>
>>> To prove manu did not make a foundational contribution, you need to
>>> prove that ALL of the document(s) was contained no foundational input from
>>> Manu.  This is not obviously the case, but if you feel strongly about it,
>>> perhaps it would be better for the image of webid to continue that
>>> privately, in a conversation with Manu.
>>>
>>>
>>> One thing one could do is in the Change History section put:
>>>
>>>  - ...
>>>  - 2010-07-11 Manu Sporny's first draft in HTML format
>>>  - the 2008 paper presented at the W3C Workshop on the Future of Social
>>> Networking <http://www.w3.org/2008/09/msnws/>"
>>>  - perhaps a short summary of the preceeding work.
>>>
>>> The text by Coralie says that there is also an importance of consistency
>>> of participation
>>>
>>>  Criteria for authorship are the expressed interest (agreed to by the
>>> Chairs) to be listed as an author and the substance and quality of the
>>> contributions. The Chairs look at the consistency of participation, the
>>> willingness to take action items, and how much "authoring" the WG member
>>> actually accomplished.
>>>
>>>
>>> and so there is where I find it a bit topsy turvy to list Manu, when he
>>> has not been helping out
>>> over close to 3 years, and not others. It was easier to list him as a
>>> contributor.
>>>
>>
>> This *may* be a criteria for removal from the current spec, but I dont
>> see why you should remove his authorship from the original specs.  In any
>> case, I think your points would have come across more effectively if there
>> was less hyperbole.  This comes across as reactionary, and at worst petty.
>> As Tim states in his book, 'The web is more a social creation than a
>> technical one'.  The social web is about building bridges :)
>>
>>
>> I did not remove him from the current spec, I just changed his role.
>>
>
> OK, to be clear.  You just want to change his role in the current spec,
> and keep his name as author on the original and previous specs?
>
>
> The links to the version Change History won't change, though I'll be more
> specific that Manu Sporny created the version
> for 2010-07-11. I'd like to add a few dates before that so that people who
> were there right from the beginning are
> duely recognised.
>
> To change someone's role due to inactivity seems allowable under the text,
> tho I suggest consulting with them privately first.
>
>
> A good idea in the future.
>
>
>>
>>
>>
>>
>>>
>>> Now it is true we never made the round yet to decide how to rate
>>> contributions. This spec has
>>> been a huge amount of work, as you know. And I do think that attendance
>>> in the teleconfs
>>> and contributions over time should count for quite a lot too.
>>>
>>
>> Without wishing to go off-topic teleconfs typically went over time,
>> sometimes over 2 hours long.  It would be easier to attend if they could be
>> kept to the allocated time frame.
>>
>>
>> Perhaps we should have another teleconf sometime soon. I think we are
>> close to ready to come out
>> with a new version of the spec...
>>
>>
>>
>>>
>>> Henry
>>>
>>>
>>>
>>>>
>>>>
>>>>
>>>>> Now a lot of Manu's company's work was on an implementation of TLS in
>>>>> JavaScript, and that is an impressive amount of work, but it's not spec
>>>>> work.
>>>>>
>>>>> Thanks for helping me to clarify this Andreas.
>>>>>
>>>>>         Henry
>>>>>
>>>>> >
>>>>> > Cheers,
>>>>> > Andreas
>>>>>
>>>>> Social Web Architect
>>>>> http://bblfish.net/
>>>>>
>>>>>
>>>>>
>>>>
>>>>    Social Web Architect
>>>> http://bblfish.net/
>>>>
>>>>
>>>
>>>    Social Web Architect
>>> http://bblfish.net/
>>>
>>>
>>
>>    Social Web Architect
>> http://bblfish.net/
>>
>>
>
> Social Web Architect
> http://bblfish.net/
>
>