3 ms·
The beauty of Gmane v1 from a user persepctive was its simplicity. http://gmane.org/lists.php http://gmane.org/lists.php <-- a simple gzipped list of all the
by textmode 10y ago
The beauty of Gmane v1 from a user persepctive was its simplicity.
http://gmane.org/lists.php http://gmane.org/lists.php <-- a simple gzipped list of all the groups and their corresponding gmane category prefixes
ftp -4o lists.gz http://gmane.org/lists.php http://gmane.org/lists.php
zgrep and a little bit of sed was all that was needed.
http://download.gmane.org/gmane.[listname]/[start]/[end] http://download.gmane.org/gmane.[listname]/[start]/[end] <-- bulk download list messages in the range [start] to [end]; no brackets needed of course.
There was no JSON or other nonsense. No pages upon pages of "API documentation". A user could figure it out in a matter of minutes.
This was an easy way to retrieve chunks of a mailing list, from circa 2002 onwards, in a format that was easy to work with, using only basic UNIX utilities.
It was like dumping a mail spool.
Given the Sirens of gratuitous complexity that never stop singing on today's web, I fear for the future of this project.
But the point of this comment is to say thanks to the author of Gmane v1. Great project, great execution.
- alphapapa 10y agoYeah, I still don't know why Lars wouldn't let them have the existing front-end code. I understand that maybe it wasn't pretty under the hood, but who cares? It worked great for many years and was extremely useful. It could always be cleaned up and refactored a bit at a time. Now they're having to reinvent the wheel. And while I shan't presume to tell them how to do it (especially since it's at their expense), I wonder if the infrastructure they're cooking up will be as future-proof. It certainly won't be as simple to reimplement if Gmane changes hands again someday, and it won't be conducive to setting up other instances of Gmane on a smaller scale.