Improving Gmail spam filter: All it needs is a barking Linus Torvalds.

Written by - 0 comments

Published on - Listed in Internet Personal


A while ago, already in December 2013, I wrote a post about Gmail's spam filter which went worse with every day (see Dear Gmail: What is going on with your spam filter?!). 

In the past few months the spam filter got to a point where 20-30% of declared spams were actually real mails (hams). Especially targeted by the spam filter seemed to be e-mails of mailing list. Suricata (OISF) and Tomcat mailing list mails seemed to be the favourites to be tagged as spam.

Now Linus Torvalds to the rescue! In his public Google Plus post on July 17th 2015, he sharply attacked the Google Mail team:

Of the roughly 1000 spam threads I've gone through so far, right now 228 threads were incorrectly marked as spam.
That's not the 0.1% false positive rate you tried to make such a big deal about last week. [...]
You dun goofed. Badly. Get your shit together, because a 20% error rate for spam detection is making your spam filter useless.

Yeah, some might think that his words are not very nice. That's a classic Linus. But guess what? It actually worked. His post was making noise, the Gmail team (in form of Google employee Sri Somanchi responded and confirmed they have improved the filters.

I waited a couple of days to make sure. And really: Since Sri claimed that the Gmail spam filters were adapted I had no ham mails in the spam folder anymore. Not even a single mailing list thread was tagged as spam. On the other side I didn't get any spams into my inbox either.

So thank you Linus for making noise, when it's necessary!

Update July 30th 2015: And we're back to the status before. Today alone at least 4 mails from the OISF mailing list were tagged as spam - the same amount as the actual amount of real spam.


Add a comment

Show form to leave a comment

Comments (newest first)

No comments yet.

RSS feed

Blog Tags:

  AWS   Android   Ansible   Apache   Apple   Atlassian   BSD   Backup   Bash   Bluecoat   CMS   Chef   Cloud   Coding   Consul   Containers   CouchDB   DB   DNS   Database   Databases   Docker   ELK   Elasticsearch   Filebeat   FreeBSD   Galera   Git   GlusterFS   Grafana   Graphics   HAProxy   HTML   Hacks   Hardware   Icinga   Influx   Internet   Java   KVM   Kibana   Kodi   Kubernetes   LVM   LXC   Linux   Logstash   Mac   Macintosh   Mail   MariaDB   Minio   MongoDB   Monitoring   Multimedia   MySQL   NFS   Nagios   Network   Nginx   OSSEC   OTRS   Office   OpenSearch   PGSQL   PHP   Perl   Personal   PostgreSQL   Postgres   PowerDNS   Proxmox   Proxy   Python   Rancher   Rant   Redis   Roundcube   SSL   Samba   Seafile   Security   Shell   SmartOS   Solaris   Surveillance   Systemd   TLS   Tomcat   Ubuntu   Unix   VMWare   VMware   Varnish   Virtualization   Windows   Wireless   Wordpress   Wyse   ZFS   Zoneminder