The Stack Exchange staff have asked us to remove tea (what came before meta) from the *.mathoverflow.net domain, for security reasons.

The content of tea will remain at my private hosting, under mathoverflow.tqft.net.

The SE staff have been rewriting links from meta and main to tea, to point to this domain.

If anyone has a better proposal for archiving (better than the current strategy of hoping that my website survives for while...), I'm happy to generate a dump of the data from tea.

UPDATE: There is now a static html dump of tea available at https://tqft.net/tea.mathoverflow.net.zip.

  • 2
    Right now, the site at mathoverflow.tqft.net has "tea.mathoverflow.net" as its title. This might cause confusion. Perhaps change it to something like "Previously tea.mathoverflow.net"? – Joel Reyes Noche Nov 4 at 0:40
  • 4
    It seems that all the links are broken -- they are still to tea.mathoverflow.net, rather than mathoverflow.tqft.net. -- So without manually editing the URL's in the browser, one can see only the list of topics. – Stefan Kohl Nov 4 at 9:18
  • 3
    As long as we lock tea, so nothing else is added, I think it's fine. Also, perhaps SE can provides MO with the necessary 50mb of space and little traffic as necessary, and just host it? – Asaf Karagila Nov 4 at 10:00
  • 1
    I think a permanent redirect from tea.mathoverflow.net to mathoverflow.tqft.net is in the works. I've pinged the relevant people at SE. This should fix all the broken links. – Scott Morrison Nov 4 at 10:30
  • 2
    I've fixed all the internal links on mathoverflow.tqft.net, but am still waiting for the redirect to be put in place as well. – Scott Morrison Nov 4 at 20:51
  • 1
    I'm curious: what security risks does the existence of tea.mathoverflow.net lead to? – John Pardon Nov 7 at 1:15
  • 6
    I think it's essentially cookie-theft: anyone who hacked tea.mathoverflow.net could read cookies set at mathoverflow.net, and in particular log in to user accounts. It's relatively low risk, as only someone visiting tea is actually exposed, but the mechanism is pretty straightforward. – Scott Morrison Nov 7 at 6:52
  • 3
    @ScottMorrison how about putting that static html in a git repo and hosting on github pages? Gives hosting with https for free indefinitely. – captncraig Nov 8 at 4:14
  • @captncraig, sure that’s a fine idea. As I’m already doing one version of the hosting — if we want a second backup someone else should do it so it’s actually redundant. – Scott Morrison Nov 9 at 4:45
  • That static HTML dump appears to have disappeared! – cat yesterday
  • @cat I was able to download the file from the link given in the post. (I got a zip-file, size about 30 MB.) In any case, having a copy of this file in the Wayback Machine seems like a reasonable idea - just in case. – Martin Sleziak 11 hours ago
  • @MartinSleziak thankfully, the link has been fixed, it was a 40x error – cat 9 hours ago

I would really hope we can find a more permanent solution for this content of archival value; perhaps via Archive.org using the "Archive-It" functionality ? I could even imagine using this service in order to preserve other MO content for posterity. Some relevant points from the FAQ:

What is Archive-It?

Archive-It is a subscription service that allows institutions to build and preserve collections of born digital content. Through the user-friendly web application, Archive-It partners can harvest, catalog, manage, and browse their archived collections. Collections are hosted at the Internet Archive data center and are accessible to the public with full-text search.

What are your fees?

At this time we have no fees for uploading and preserving materials. We estimate that permanent storage costs us approximately $2.00US per gigabyte. While there are no fees we always appreciate donations to offset these costs.

Who has access to the collections?

Anyone can access our collections through our website archive.org. The web archive can be searched using the Wayback Machine. The Archive makes the collections available at no cost to researchers, historians, and scholars. At present, it takes someone with a certain level of technical knowledge to access collections in a way other than our website, but there is no requirement that a user be affiliated with any particular organization.

  • 2
    Since you have mentioned also archiving of MathOverflow content, I'll just: 1. A link to previous discussion about this: Long-term archiving of MathOverflow. 2. That the content of all sites in the Stack Exchange sites is in some forem already archived: Stack Exchange Creative Commons data now hosted by the Internet Archive (at least if I correctly understood the linked post). – Martin Sleziak Nov 4 at 12:36
  • 1
    "Internet Archive" = "Archive.org", so can't we just add "tea" to this repository? – Carlo Beenakker Nov 4 at 13:03
  • 1
    That's probably question for somebody who knows more about the technical details of this. But I would hazard a guess that the format used in that data dump relies heavily on the fact that it consists of Posts, Comments and other data from Stack Exchange sites - while tea was not operated on the Stack Exchange platform. So this might be a problem if MO would want to include it into this data dump. – Martin Sleziak Nov 4 at 13:08
  • 1
    It seems the only way to get any further information about using the archive-it service is to sign up to one of their regular webinars, which sounds like a frustrating waste of time to even find out if we would be eligible for something like tea. (It does seem out of character relative to many of their other projects.) – Scott Morrison Nov 4 at 20:30
  • @ScottMorrison I think the archive-it service is a bit of a red herring, its intended more for institutions. Just the standard upload functionality is probably fine for sticking a static dump of the site. – mbrig yesterday

Any reason SE don’t want to host a static HTML version at tea.mathoverflow.net, like they already do for the former Blog Overflow? Presumably the “security reasons” are that they don’t fancy running custom forum software on a MO domain without someone actively supporting and auditing it (and they are right), but serving a static dump isn’t affected by this argument.

The current setup also breaks all former links to tea.mathoverflow.net by redirecting them to the home page of the archive, which is a Bad Thing. Link rot is a reality of the web, but that doesn’t mean we should contribute to it. (I think we’ve all been bitten by links broken by careless university website operators.)

New contributor
Alex Shpilkin is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.
  • 4
    They've said the don't want to host, even the static HTML. I hadn't realised they'd set up the redirect to just go the home page. I will ask them to fix this. – Scott Morrison Nov 9 at 19:10
  • 1
    The redirect has now been fixed, and all existing links to tea should now resolve correctly to the mathoverflow.tqft.net server. Thanks for pointing this out. – Scott Morrison Nov 12 at 0:16

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .