Comment by Diagon | Hacker News Reader

Comment by Diagon | Hacker News Reader

Extensive history is about to be lost. Despite being broken, many organizations still use it. Examples from that post:

A police cooperative in Washington DC that was using them as a network to communicate with their respective neighborhoods with over 17,000 members.

A phone company in the UK that assigns phone numbers using the groups and now will lose all those phone designations when it’s deleted.

A Birding group in new Delhi with 2,000 members that has collected data and research on birds for TWO DECADES.

An Adoption group in France, that has been using it for years and years to communicate and share history and photos and more.

They also would have found: Numerous support groups for people who are suicidal or depressed.

Numerous medical groups for people to communicate more effectively with their doctors.

Numerous Vet groups with 24 hr care advice for sick pets.

Numerous support and help groups for the Elderly.

Numerous Historical groups for WW2 Veterans, Vietnam Veterans, and etc.

Numerous science groups that have used them for years and have all their research there.

Numerous fan fiction groups or arts groups that have shared their work for years.

> A phone company in the UK that assigns phone numbers using the groups and now will lose all those phone designations when it’s deleted.

Wow, somebody invented a database that's even worse than an Excel file on a network share.

(Also, how are they going to assign new numbers when archive.org takes over? Is archive.org going to give them write access?)

“My understanding is that [the group] will still function as a mailing list, which is for all practical purposes, what people use this as,” https://www.theverge.com/2019/10/17/20919630/yahoo-groups-uk...

That's right, but (our main concern) is that the archives are being deleted. With no further history being recorded, it's utility for some purposes is limited. I have also come across some complaints that even as a list-serve it can be problematic. Posts, for example, are no longer coming in order.

But as a mailing list, each subscriber has the entire archive, at least from the date they joined. And any one of them can make it publicly accessible at any point in the future. In practice it will undoubtedly result in the destruction of enormous amounts of human knowledge, but at least in theory not much is getting immediately lost.

The difficulty in a lot of cases is finding someone who has a complete copy of the group. Yahoo Groups also had file, photo and database features, and archives of those are likely to be incomplete. You'd have to go through the member list (primarily early members) and find someone who still had a copy of all the messages.

The other problem is making it available - I ran a Yahoo group for many years, and have Mbox and Maildir format archives. I'm still looking for a decent web-based browser for these. HyperKitty (Mailman's archive browser) came close, but seems to require most of Mailman to be installed in order to work.

In my case, I managed to archive a bunch of groups related to amateur radio -- and I will be placing these on archive.org as soon as I have a spare moment to zip them up. A difficult-to-access archive is better than no archive at all, the important part is getting the data into a safe place.

I have the same problem. I used some old script back in 2006 to download a couple of groups in ... I think it's Mbox format. It's just not clear what to do with it.

Hmm... a standalone viewer for these formats (that exposed a webserver that could be accessed in a browser) sounds like it would pretty trivial given a parser for the email format itself. Especially maildir!

How big are these archives? Do you have any samples? Does the viewer need any special features? (threading?)

> pretty trivial given a parser for the email format itself

The problem is that there isn't any standard that defines what can and can't go inside the body of an email message. So if you want to post each email message exactly in a thread exactly as is, i.e. each with completely different typography and with all the replies attached and not sanitized in any way, then that's relatively easy. But it's also completely unreadable for more than about 30 seconds, and doesn't allow for good search functionality. These problems aren't a deal breaker if you're only trying to make sense of your own inbox, but when you're looking for specific information across millions of people's inboxes then they're a complete nonstarter.

I'm actually applying for an SBIR grant right now to work on the NLP algorithms that power fwdeveryone.com, if you have any interest in writing a letter of support. Basically it would eventually enable someone to mass export something like an MBOX archive onto the web in a cleaned up format with accessible typography. You can play around with prettyfwd.com to get an idea of the current state of the tech, it works well for 95% of (non-commercial) email threads but still needs some more work to support the rest.

I remember how great Gmane used to be with several incredible web-based views of mailing list archives. Too bad it sounds like the source code was lost and never open sourced. Another on the list of services that died without passing on enough of the torch.

Oh, neat, that's a pretty interesting read. Also good to know people will be able to keep their phone number after this.

Not to be flippant, but wouldn't one of the members of these groups have a copy of the group in their email? Given gmail and whatnot store things virtually indefinitely, couldn't the contents be recovered that way?

-EJ

Some of these groups are decades old. For them, you'd be hard pressed to find someone who was there for the whole history of the group and kept them all. Also, yahoo was often a headache, dropping emails to individuals - you'd have to go to the website to read them. And furthermore, there is a lot more than emails stored on the platform: files/images/links/calendars/databases ...

To add to all this, it's not an individual project. Most people done' have technical competence. They need someone to help. That's what the Archive Team has been trying to offer (if not for Verizon).

Reply on news.ycombinator.com