Re: Google Zero is coming [was Re: Wikipedia at 25: A Wake-Up Call (essay)]
Steven Walling via Wikimedia-l <[email protected]> Fri, 10 Jul 2026 10:29:03 -0700
| Newsgroups | gmane.org.wikimedia.foundation |
|---|---|
| Message-ID | <CABxVVqt0zrENXYC2FHuhzTPJqbFenJeKRL0bH11tKxzkffvwQQ@mail.gmail.com> |
--===============7294648608169688575== Content-Type: multipart/alternative; boundary="000000000000e30c33065645158a" --000000000000e30c33065645158a Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Fri, Jul 10, 2026 at 9:52=E2=80=AFAM Erik Moeller via Wikimedia-l < [email protected]> wrote: > On Fri, Jul 10, 2026 at 5:59=E2=80=AFPM James Heilman via Wikimedia-l > <[email protected]> wrote: > > > Yah a search engine that actually gives real references that supports > the statements in question would be amazing. > > Almost like .. a Knowledge Engine. ;-) > Is the WMF building an MCP server to connect Wikipedia and Wikidata directly to Gemini, Claude, and ChatGPT? This is a more lightweight, backdoor way to leverage the audience of those platforms but present structured outputs to AI chats for users based on Wikimedia knowledge. If we did so, we could present citations within the returned responses to users, and those platforms make it transparent to the user when they are calling a particular tool. Steven Walling Sadly, the only realistic path I see there would be through > acquisition, and even if that was financially feasible, you'd begin by > inheriting a lot of corporate practices that aren't really consistent > with Wikimedia values. > > But perhaps there is a middle ground where Wikimedia seeks to define > more clearly the terms of engagement that it wants with search engines > (clear attribution, clear and correct references, calls-to-edit, > etc.), and then finds and recognizes search partners who implement > those. To Luis' point, that need not be done by WMF. > > Warmly, > > Erik > _______________________________________________ > Wikimedia-l mailing list -- [email protected], guidelines > at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and > https://meta.wikimedia.org/wiki/Wikimedia-l > Public archives at > https://lists.wikimedia.org/hyperkitty/list/[email protected]= rg/message/HNDPDZBBDMILF7WGMUBVJVIAZYQ7OXOS/ > To unsubscribe send an email to wikimedia-l-leave-RusutVdil2icGmH+5r0DM0B+6BGkLq7r@public.gmane.org --000000000000e30c33065645158a Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote g= mail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">On Fri, Jul 10,= 2026 at 9:52=E2=80=AFAM Erik Moeller via Wikimedia-l <<a href=3D"mailto= :[email protected]">[email protected]</a>> w= rote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0p= x 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Fri, Ju= l 10, 2026 at 5:59=E2=80=AFPM James Heilman via Wikimedia-l<br> <<a href=3D"mailto:[email protected]" target=3D"_blank">wi= [email protected]</a>> wrote:<br> <br> > Yah a search engine that actually gives real references that supports = the statements in question would be amazing.<br> <br> Almost like .. a Knowledge Engine. ;-)<br></blockquote><div><br></div>Is th= e WMF building an MCP server to connect Wikipedia and Wikidata directly to = Gemini, Claude, and ChatGPT? This is a more lightweight, backdoor way to le= verage the audience of those platforms but present structured outputs to AI= chats for users based on Wikimedia knowledge.=C2=A0<span style=3D"backgrou= nd-color:transparent">If we did so, we could present citations within the r= eturned responses to users, and those platforms make it transparent to the = user when they are calling a particular tool.=C2=A0</span><span style=3D"ba= ckground-color:transparent">=C2=A0</span></div><div class=3D"gmail_quote gm= ail_quote_container"><span style=3D"background-color:transparent"><br></spa= n></div><div class=3D"gmail_quote gmail_quote_container"><span style=3D"bac= kground-color:transparent">Steven Walling=C2=A0</span></div><div class=3D"g= mail_quote gmail_quote_container"><br><blockquote class=3D"gmail_quote" sty= le=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);paddi= ng-left:1ex">Sadly, the only realistic path I see there would be through<br= > acquisition, and even if that was financially feasible, you'd begin by<= br> inheriting a lot of corporate practices that aren't really consistent<b= r> with Wikimedia values.<br> <br> But perhaps there is a middle ground where Wikimedia seeks to define<br> more clearly the terms of engagement that it wants with search engines<br> (clear attribution, clear and correct references, calls-to-edit,<br> etc.), and then finds and recognizes search partners who implement<br> those. To Luis' point, that need not be done by WMF.<br> <br> Warmly,<br> <br> Erik<br> _______________________________________________<br> Wikimedia-l mailing list -- <a href=3D"mailto:[email protected]= rg" target=3D"_blank">[email protected]</a>, guidelines at: <= a href=3D"https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines" rel=3D"= noreferrer" target=3D"_blank">https://meta.wikimedia.org/wiki/Mailing_lists= /Guidelines</a> and <a href=3D"https://meta.wikimedia.org/wiki/Wikimedia-l"= rel=3D"noreferrer" target=3D"_blank">https://meta.wikimedia.org/wiki/Wikim= edia-l</a><br> Public archives at <a href=3D"https://lists.wikimedia.org/hyperkitty/list/w= [email protected]/message/HNDPDZBBDMILF7WGMUBVJVIAZYQ7OXOS/" r= el=3D"noreferrer" target=3D"_blank">https://lists.wikimedia.org/hyperkitty/= list/[email protected]/message/HNDPDZBBDMILF7WGMUBVJVIAZYQ7OX= OS/</a><br> To unsubscribe send an email to <a href=3D"mailto:[email protected]= ikimedia.org" target=3D"_blank">wikimedia-l-leave-RusutVdil2icGmH+5r0DM0B+6BGkLq7r@public.gmane.org</a></= blockquote></div></div> --000000000000e30c33065645158a-- --===============7294648608169688575== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Wikimedia-l mailing list -- [email protected], guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l Public archives at https://lists.wikimedia.org/hyperkitty/list/[email protected]/message/IUDAKJRN5EQT5CCWEEYQXKDSVXCDBXR6/ To unsubscribe send an email to wikimedia-l-leave-RusutVdil2icGmH+5r0DM0B+6BGkLq7r@public.gmane.org --===============7294648608169688575==--