Re: Turning off Etherpad

[Toerless Eckert <[e-mail]>]

On Wed, Aug 11, 2021 at 11:27:41AM +1000, Mark Nottingham wrote:
> My understanding was always that etherpad was a best-effort, ephemeral service; i.e., there was no guarantee (implicit or explicit) of stability or availability. That's probably the right policy, given that there able to be edited by *anyone on the Internet*.

Yes. But my comments where not about the intent of etherpad,
but if/what we would want to do now to preserve history, even
if that was not a desire when we started using it.

The main question IMHO is whether we could publish the whole content
of etherpad or not. I would like to see that question answered. If
the answer is yes, then we can discuss if and/or how to do that.

If the answer is no, then i guss a tar file could still go into the
IETF vault, and later generations could figure out if/who might
be able to take a look at it.

> We might consider a somewhat more formal and explicit policy on service stability and availability (on a service-by-service basis), so that folks aren't surprised by this in the future. E.g., the W3C has:
>   https://www.w3.org/Consortium/Persistence

Right. My suggestion for codimd was exactly around that IMHO:

- declare CodiMD content to be public.
- automatically make older CodiMD content read-only

The conflict between this goal and the design of etherpad/codiMD is that
boh seem to be based on confidentiality by obscurity, e.g.: no
public search and/or list of pages...

Cheers
    Toerless

> Cheers,
> 
> 
> > On 11 Aug 2021, at 11:24 am, Toerless Eckert <[e-mail]> wrote:
> > 
> > So, what is he plan for archiving etherpad.ietf.org ?
> > Can/will the be made read-only/no new data ?
> > 
> > Or is all that content that may still have URLs in emails and other places
> > pointing to it deemed to be lost except for the subset of pages that
> > where converted into IETF meeting minuts by WG chairs ?
> > 
> > If all this content is meant to be lost because it is too difficult to
> > change to read-only, is this the same strategy for codimd ? Aka:
> > if you do not save it, it will be lost at some point in time ?
> > 
> > If converting to read-only is too difficult, maybe the whole tree
> > can just be put up for download as a tar file for any future researchers...
> > However, i am not sure if all the content is meant to be public, or if
> > we did assume some degree of possible confidentiality because of non-searchable
> > namespace.
> > 
> > Btw: I think it would be great if codimd would solve these issues:
> > You can only create new names under whatever.ietf.org/2021/<your-string-here>
> > and after 2 years the whole years directory tree turns read-only with
> > indexing/search...
> > 
> > Cheers
> >    Toeless
> > 
> > On Tue, Aug 10, 2021 at 06:05:34PM -0700, Fred Baker wrote:
> >> People use it in recording our minutes. Losing it would be unfortunate.
> >> 
> >>> On Aug 10, 2021, at 2:01 PM, Robert Sparks <[e-mail]> wrote:
> >>> 
> >>> Chairs -
> >>> 
> >>> We are making plans to turn down etherpad.ietf.org in a few weeks. Let me know of any concerns or need for delay.
> >>> 
> >>> RjS
> >>> 
> > 
> > -- 
> > ---
> > [e-mail]
> > 
> 
> --
> Mark Nottingham   https://www.mnot.net/

-- 
---
[e-mail]