I am not sure about modernizing. I am all for developing core applications like mailman, but I fail to see great value in this piece of execution.
Post voting is bad for fact oriented discussions, popularity is a poor indicator for the value of a post. The user interface is cramped, and hides too much of valuable information I would expect to see instantly. Also, the looks are from previous decade (looks like sf.net clone to me) and not very aesthetic.
Sure, this might be a step forward nevertheless, but it is a really small one.
As far as I know, those are configurable features; I'd expect any FOSS project mailing list to turn them off.
One of the goals of mailman is that the same software should be able to handle either a mailing list or a forum, so that people who want the UI of a forum can collaborate with people who want to live in mutt.
To me the huge step here is the RESTful API, assuming it is indeed RESTful. Having access to data that's structured naturally and less like a full-on layout document (HTML) means that any developer can trivially grab the underlying data and do whatever they want with it.
Can't wait for this. It currently might be released as a final after PyCon 2015, meaning it could be here this month.
We've been waiting for this to replace our ancient ecartis manager (https://launchpad.net/ecartis). While the UI doesn't look extremely pretty, compared to ecartis it will be like moving from 1999 to 2008 or so.
The first link describes it as "D news aggregator, newsgroup client, web newsreader and IRC bot." I don't see anything about mailing list. And at first, I thought the D part meant that was customized to be just for the D project's uses. But thinking about it, it sounds like it's because it was written in D?
"To answer the question that at least a few dozen of you are looking for: Mailman 3 will not, by default, send you a monthly reminder that includes your password in plain text. In fact, it can't; Mailman 3 hashes passwords before storing them."
As I understand it from the write-up (and a brief look at the source seems to confirm that), Mailman 3.0 does encrypt passwords by default and does not mail them out any longer.
It is stored in unicode because that's what passlib hash algorithms return, not because there's an underlying plaintext representation.
I've always liked Mailman a lot, and prefer mailing lists to web forums, but we've just recently made the decision to not enable it by default in Virtualmin (for the past 12+ years, we've enabled it on every install and made it easy for non-privileged users to create new lists and such). But, when we recently assessed how many of our customers were actually using it vs the resources it requires, we can't justify it.
This new version doesn't alter the reality that very few website owners want mailing lists, anymore. But, it makes me want to start some new mailing lists (I don't currently maintain any Mailman mailing lists, though in the past I was wrangling dozens of them).
The UI looks great, and the new architectural changes sound great, too. It's long overdue, though.
When I look at this screenshot [1], it seems to suffer the same issue as LWN: After around 100 characters the text should begin on a new line, not after it filled the screen. One can complain about lots of rules designers use, but this one is actually scientifically backed up. I wonder why so many websites still do this wrong, maybe web developer tools in browsers should throw warnings?
I used Mailman for more than 10 years, to run a fairly large (3,000-person) list. Before that, I had used a number of other mailing list managers.
It's true that Mailman was rock solid. But it had a number of features that made it difficult for users and administrators alike.
In particular, users who were not computer savvy complained bitterly that they couldn't easily search through the archives, couldn't understand how to subscribe and unsubscribe themselves, and that the archives (in their repeated messages to the moderators) didn't look like the mailing lists on Google and Yahoo.
In other words, it wasn't sufficient to have solid software. Mailman needed a modern UI. It took some time, but that moment seems to have come. It's a bit too late for me; for a variety of reasons (many having nothing to do with Mailman), I recently moved my mailing list to Google Groups.
I'm very happy to see this new set of functionality aimed at making Mailman relevant for a new generation of users. Kudos to the maintainers for sticking with the project, and for providing a valuable service that shouldn't need to be outsourced to third parties. And for not only pushing forward with the internals, but also on the UI. It looks quite snazzy and impressive, and should help to keep Mailman relevant for many more years.
I'm surprised about the problems they had with (un)subscribing -- the web ui has always been a controversial part of Mailman, but intended for those that think the email UI is too hard to use. FWIW I actually think it's much easier to manage subscriptions to Mailman (via the web UI) than to manage them on Google Groups.
As for search and archive -- yeah, those have been in need of a facelift for a long, long time. I've been toying with doing something about them myself, but have hold off, waiting to see what Mailman 3 would ship with.
How is it any more modern than GroupServer [0], for example? I was looking for some mailing list software, mailman looked pretty dated, now it will look newer, and, yet, not modern, but I ended up using Groups.io [1] at the end.
On Windows, http://www.mailstore.com (free home edition) will archive, index and quickly search local + cloud email, from mbox, Outlook, Thunderbird, IMAP, Yahoo, Google, etc.
Just download the MBOX files and put them in your Thunderbird profile. Thunderbird will build an index immediately and you can use Thunderbird to process, read, search, mark those posts
I hope that people turn on the post to list feature in the web viewer.
(because there really are lot of mailing lists where the desire of the core users to stick with their own mail setup is understandable but at odds with making it easy to capture more casual contributions)
I don't think mailing lists have a future. There is no new adoption of mailing lists, newer developers (those under 30) use other forums for communication (stackoverflow, github, etc) and mailing lists will eventually just die out as the generations of developers do.
I am not trying to criticize mailing lists, they definitely have a lot of advantages over some of the newer systems that have grown to replace them. Without organic growth from new developers though, mailing lists are doomed to become a relic of a distant time. To most developers, they already are.
To me a mailing list serve a purpose. Its a middle ground between IRC and not having any discussion medium at all.
IRC is suitable for chat but discussions happen on mailing lists (or here on Hackernews).
In mailing lists conversations are threaded and there's nothing wrong with carefully considering your response before the question flies off the screen as on an IRC chat. Stackoverflow is great for one-off questions, but they are not a support forum for your project.
IMO new or young projects don't yet have a need for real discussion around the project, the project is to small, the Dev team is too small and or is located geographically close together, or the project just doesn't warrant in depth discussion yet.
Edit: my biggest gripe about mailing lists is that many do not meet the one - click - unsubscribe standard, most that I've seen require two clicks. Maybe I'm being pedantic, but it does matter to me.
In my opinion, as an under-30 developer myself, and I don't see neither SO nor Github replacing mailing lists for broader discussions; SO is very specifically about the nitty-gritty, and while Github issues can accommodate them to a degree, they're not as suited, in my opinion.
For those of you who don't use mailing lists, what do you use instead for these discussions?
I used to be subscribed to a number of mailing lists maybe a decade ago, but none now. The signal:noise ratio is generally pretty terrible for anyone pressed for time.
What makes the SnR ratio so different for forums vs mailing lists? I haven't noticed, frankly.
I agree with Jeff Atwood: forum software is generally pretty terrible. The difference being that I don't find his better - the way it actively prevents people from reading offline is enough to make me extremely frustrated.
There are a number of open source discussion lists I've been subscribed to over the past decade (usenet, then Google Groups) that have recently migrated to hosted Discourse boards because of Google's lack of spam fighting tools. Who cares about what the listservs UI looks like, the entire point is to receive and send email. Lets help build good spam fighting capabilities into mailman and others so all of our knowledge isn't locked into these walled-garden communities.
* Since data is now in SQL, it's easier to interact with the DB and have simple tools delete spammers.
* Since there's an exposable API now, it's possible to develop systems that interact with mailman and help with the issue.
I find your comment quite odd. It's like saying people don't need email anymore -- which is ridiculous. So as long as people are using email, there will be the need for groups of them to coordinate over email. Mailing lists are one answer.
And what on earth does 'organic growth' of mailing lists even mean in this context?
Had you been commenting about mailman specifically, then I guess I could see your point (it's one product in this space) -- generalising to all mailing lists ever seems absurd.
I think you make a reasonable point, AND I note that there's a promotion at the top of the page.
"The following subscription-only content has been made available to you by an LWN subscriber. Thousands of subscribers depend on LWN for the best news from the Linux and free software communities. If you enjoy this article, please consider accepting the trial offer on the right. Thank you for visiting LWN.net!"
It's possible (this was posted on HN 5 hours ago) that text was added retrospectively, but working on the basis that it's boilerplate text then LWN seems to figure some subscriber links will be shared AND that's OK because it's an opportunity to promote subscription to others.
FWIW LWN posts subscriber links to HN themselves from time to time, as a promotion. I'm sure they could block access to subscriber links that saw "unusual" amount of traffic -- but I guess (whether or not this was posted by LWN or not) they usually assume that the more people that read one or two free links, the more people will subscribe (like libraries leads to increase in book sales etc).
Post voting is bad for fact oriented discussions, popularity is a poor indicator for the value of a post. The user interface is cramped, and hides too much of valuable information I would expect to see instantly. Also, the looks are from previous decade (looks like sf.net clone to me) and not very aesthetic.
Sure, this might be a step forward nevertheless, but it is a really small one.