|
|
Mozy Online Backup: 2GB Free. Automatic. Secure.
Subject: Re: OpenID anonymous for LJ? - msg#00029
List: web.openid.general
Zefiro wrote:
>
> >And it'll definitely get more interesting when we have a reputation/trust
> >layer atop
> >OpenID/LID/etc/etc which will be coming, and from multiple vendors/sites, it
> >seems like.
>
> I'm not sure how I like this. Probably depends very much if this stays as
> free as OpenID is - i.e. ratings based on how you
> behave or how much you use your ID, not from which server it comes.
>
My idea here has always been something along the lines of a local rank.
> From your point of view, based on who you trust, and who they trust
(etc), what is the trust score of a given identity?
Your idea of trust, then, is based on the idea of trust from those
people you trust, which is a reasonable method in most cases. Another
similar approach is to trust those who vote similarly to you.
As long as the trust metric is different from each person's point of
view I can't see that there's a major problem with it remaining "free".
No single party is capable of manipulating everyone's relationships to
make them appear more trusted to the entire community.
Was this page helpful?
Thread at a glance:
Previous Message by Date:
click to view message preview
Re: OpenID anonymous for LJ?
Mark Rafn wrote:
>
> Agreed. Ideally, this would apply to ANY id, including other LJ users.
> Why I'd want a random just-registered LJ account to be able to post a
> comment but not an OpenID user is what I don't understand.
This is getting a little too LiveJournal-specific for this mailing list,
but okay...
"Real" LiveJournal users aren't allowed to leave comments until they've
validated an email address. While I guess that isn't really much more
difficult than getting a working OpenID identity, it does seem to reduce
the number of interlopers for some reason. Most of the undersirable
comments I see on LiveJournal are from anonymous posters in journals
where anonymous posting is allowed.
>> That said, there are a number of shortcomings in LiveJournal's OpenID
>> consumer support, but it's roughly where it'll end up.
>
> If you can add specific OpenID identities to friends lists and
> comment-allowed lists, and have the same granularity for LJ accounts,
> this seems ideal to me.
You can already add OpenID identities to your friends list as if they
were normal accounts. The only real difference between an OpenID account
and a regular account is that the journaltype field contains an I rather
than a P.
As for adding to "comment-allowed" lists, currently (for better or
worse) all trust relationships are handled by the single friends list.
If that is going to be changed, it'd likely be the long-promised
friend/watch/trust separation -- or at least something resembling it --
rather than a special case for commenting.
Next Message by Date:
click to view message preview
Re: OpenID anonymous for LJ?
On Sat, Aug 06, 2005 at 12:44:59AM +0100, Martin Atkins wrote:
> This is getting a little too LiveJournal-specific for this mailing list,
> but okay...
>
> "Real" LiveJournal users aren't allowed to leave comments until they've
> validated an email address. While I guess that isn't really much more
> difficult than getting a working OpenID identity, it does seem to reduce
> the number of interlopers for some reason. Most of the undersirable
Are you sure you're comparing to the 'openid-spam'?
> comments I see on LiveJournal are from anonymous posters in journals
> where anonymous posting is allowed.
This is indeed a bit too lj-specific, but look, the point is not giving
OpenID users more credibitily, it's just about separating them from
anonymous users. And they are definitely different, otherwise what's the
point of OpenID?
Love,
H
Previous Message by Thread:
click to view message preview
Re: OpenID anonymous for LJ?
Hi, Brad
> If you walk down the street [...]
agreed, and it was clear from the start that OpenID does not prove anything,
especially not if a user is trustworthy or
anything, except for the one thing that the same ID means it's the same user.
Which, I guess, it's what this 'disallow anonymous comments' is partly about:
to be able to have a conversation with someone you
know will be the same guy during your talk.
If it's about trustworthyness, spamming, etc I think it's equally possible with
LJ-IDs as well.
Ok, only thing I can think of right now is that creating an OpenID for spamming
is really easy, an automated LJ-account-creator
not so easy. So this might in fact be an argument pro 'internal system only'.
Still I'd like to see OpenID and anonymous not being treated the same.
> That said, there are a number of shortcomings in LiveJournal's OpenID
> consumer support, but it's roughly where it'll end up.
I think I don't understand you here. Where will it end up?
> And it'll definitely get more interesting when we have a reputation/trust
> layer atop
> OpenID/LID/etc/etc which will be coming, and from multiple vendors/sites, it
> seems like.
I'm not sure how I like this. Probably depends very much if this stays as free
as OpenID is - i.e. ratings based on how you
behave or how much you use your ID, not from which server it comes.
:Frederic:
Next Message by Thread:
click to view message preview
Re: OpenID anonymous for LJ?
On Sat, 6 Aug 2005, Zefiro wrote:
> Hi, Brad
>
> > If you walk down the street [...]
> agreed, and it was clear from the start that OpenID does not prove anything,
> especially not if a user is trustworthy or
> anything, except for the one thing that the same ID means it's the same user.
>
> Ok, only thing I can think of right now is that creating an OpenID for
> spamming is really easy, an automated LJ-account-creator not so easy. So
> this might in fact be an argument pro 'internal system only'.
Yes. LiveJournal has more checks on account creation than OpenID, so they
can be trusted by default a /little/ more, but not a hell of a lot.
> Still I'd like to see OpenID and anonymous not being treated the same.
That was just the easiest thing to bang out. Eventually we'll break 'em
into two options so users can say exactly what their preferences are.
- Brad
|
|