Re: problem with recent drafts on tools.ietf.org? {35440727}

[Margaret Cullen <[e-mail]>]

Henrik has done a lot of wonderful things for the IETF community for many years as a volunteer tools developer, as the former head of our tools development effort, and as a contractor.  He deserves to be respected and admired for those contributions.  I also understand that he has developed some intellectual property that he owns, and I think that we should respect his rights to that intellectual property.

However, pointing a subdomain of our DNS space at a service that someone is running does not result in that person “owning” that subdomain.  This is done all of the time with hosting services, and it is often done specifically to remove the possibility of lock-in to the hosting service — www.mydomain.com <http://www.mydomain.com/> can point to one web hosting service today and to a different web hosting service tomorrow.  

IMO, if the IETF community needs a specific set of URLs in our domain to point to a specific set of resources (such as tools.ietf.org/html <http://tools.ietf.org/html> URLs pointing to an up-to-date set of Internet Drafts), we should make it so.  I think that Jay and Robert are the right people to figure out the best way to make that happen.

Margaret



> On Mar 3, 2021, at 8:29 PM, Jay Daley <[e-mail]> wrote:
> 
> Hi Paul, Lou and others
> 
>> On 4/03/2021, at 1:11 PM, Paul Wouters <[e-mail]> wrote:
>> 
>> On Wed, 3 Mar 2021, Lou Berger wrote:
>> 
>>> Well this is a troubling response -- does anyone know if there's plans for tools.ietf.org? I would have thought that some thought would be given to this and
>>> maybe some announcement made to the IETF.  Did I miss it???  Are we just destined to the whims of bit rot?
>> 
>>> As you are probably aware, tools.ietf.org is owned and operated by Henrik who has, unfortunately, departed the IETF.
>> 
>> I would really like some clarification about what happened, and who owns
>> the rights of the content on tools.ietf.org, and if it is not IETF or
>> ISOC, how we ended up in a position where we didn't own what we
>> published for many years. And what the plans are with regard to ensuring
>> URLs posted there remain valid - even if only to redirect back to
>> something that _is_ under IETF / ISOC change control.
> 
> I do not know the full details and history of how we got here.  Here is the background I provided to the LLC Board for their February meeting and is now published in the minutes of that meeting [1]:
> 
>> As background, our ex-contractor and long term community member Henrik Levkowetz, controls tools.ietf.org <http://tools.ietf.org/> and operates it on servers donated by other community members over the years. tools.ietf.org <http://tools.ietf.org/> runs a set of services, some of which are maintained by the IETF such as the ​online xml2rfc​ and code repositories for ​datatracker​ and x​ml2rfc,​ and some of which are maintained by Henrik such as the HTMLized ​RFCs​ and I-Ds. Over time, some of these services have become mission critical.
>> 
>> Henrik asserts that the domain name tools.ietf.org <http://tools.ietf.org/> was delegated to him personally (by the IETF Chair in 2004) and the servers were donated to him personally not the IETF and that he should retain full control of what runs on these servers and how they are managed. Some of the servers are physically based in AMS operated datacenters.
>> 
>> We recently had an incident where one of the servers operating tools.ietf.org <http://tools.ietf.org/> failed irreparably. There were some services running exclusively on that server that therefore needed to be moved to other servers, which Henrik and another volunteer were able to do. However some performance issues remained on other services for a few days due to DNS configuration issues. To prevent a recurrence, we proposed to Henrik that the management of the DNS move to AMS so that they can make emergency changes in situations like this. Henrik’s response to this proposal was quite intense and included a threat to turn off tools.ietf.org <http://tools.ietf.org/> and all its services if we made this change.
>> 
>> The Tools Team PM (Robert Sparks) already had a partial plan in place to migrate services off tools.ietf.org <http://tools.ietf.org/> but this combination of failing servers, a single person in control and threats of service shutdown have significantly increased the risk this situation poses and consequently I am reprioritising work to assist and accelerate the migration. The good news is that Robert and Henrik have been working cooperatively to effect the migration, though Robert needs additional resources to complete this and other work (see below). 
> 
> This plan includes, in rough priority order, the following (Robert may wish to correct me if this is wrong):
> 
> 1.  Reproducing the same htmlised I-Ds and RFCs on Datatracker
> 2.  Moving the xm2rfc and ietfbd (Datatracker) trac/svn instances to GitHub
> 3.  Moving the other trac instances to a new wiki product (while these have now moved off of tools.ietf.org <http://tools.ietf.org/> they were still maintained by Henrik)
> 4.  Moving the web based tools (rfcdiff, idspell, xm2rfc, idnits) elsewhere
> 5.  Reproducing the bibxml service elsewhere
> 
> The additional resources mentioned is the new Senior Software Development Engineer that we are recruiting for [2].
> 
> Discussions are still ongoing to see if certain portions of tools.ietf.org <http://tools.ietf.org/> can be turned off as and when they either break or are reproduced or moved, and what redirections are possible and how.
> 
> Jay
> 
> [1]  https://www.ietf.org/media/documents/2021-02-18-llc-board-minutes.pdf <https://www.ietf.org/media/documents/2021-02-18-llc-board-minutes.pdf>
> [2]  https://mailarchive.ietf.org/arch/msg/ietf-announce/rCLegOyqbYcE8no8riO96F8BDdM/ <https://mailarchive.ietf.org/arch/msg/ietf-announce/rCLegOyqbYcE8no8riO96F8BDdM/>
> 
>> 
>> Paul
>> 
> 
> -- 
> Jay Daley
> IETF Executive Director
> [e-mail] <mailto:[e-mail]>