With AppEngine at Google, MongoDB at Disqus, Cassandra at Facebook and Redis at Github you can definitely say that SQL databases are one of many options available today and don't dominate like they did 5 years ago.

if i'm not mistaken, the majority of those organizations still rely heavily on relational datastores, except in the case of exceptional workloads. in addition, i believe facebook has since migrated away from cassandra to the hadoop stack for their messaging platform, though they primarily use mysql (or its successors).

SQL is being replaced in niches that strain its model. elsewhere, it remains steadfast.

I agree, ultimately it comes to everyone's own definition of "tyranny". :) I meant it as "not really having any defensible other choice". (While, of course for example CDB and BerkleyDB have "always" existed.)

I think nobody expects SQL's "market share" to fall to low levels, especially with noSQL requiring much deeper understanding of the data and it's planned use. NoSQL practically operates on a lower layer than SQL does.

Still, it's nice to see people thinking about data storage choices and not going blindly to MySQL/Oracle/etc!

After using MongoDB with Mongoid I strongly disagree with the premise that over the next 5 years SQL databases will be the default and NoSql limited to certain niches. And I know SQL better than most. Alter table ... or db:migrate no more for me.

I know that Facebook created Cassandra, but do they use it for anything substantial? I read that Cassandra was created for the Facebook inbox, but I more recently read that Facebook is now using HBase for the messaging platform.

probably worth to note that Disqus uses Redis as well, not for storing the comments, but for analytics: http://disqus.com/analytics/

AppEngine is adding an SQL/relational database to their service in the near future.

Yea, I knew the post would be vacuous when I hit "tyranny", but I guess the flourishes makes good HN bait to speak in sweeping terms like that. There's little about on-disk consistency and data loss risks, scale characteristics yet there's coverage of the wire protocol? Meh. Sure, I'm glad to have a lot of these tools available, the bad old days of rolling your stuff with Berkeley DB or NDBM seem safely behind us but the reality is that there are many classes of problems for which SQL is still and will remain the most sensible solution. Get over it.

I also don't understand the 15 years figure either, is that a reference to when MySQL was initially released? I hope the original poster understands that SQL is older than that.

No conspiracy behind the 15 years :) I estimate that that's about the time SQL bot really mainstream (as opposed to dBase files, etc).

I'm old enough to remember somebody being threatened with firing for using SQL, because of it's bad performance compared with seeking. :)

Not that they get thrown away, they are still very useful. But at least there's an alternative now.

I mean, seriously, Zabbix keeping monitoring data in MySQL? Also Piwik? That's a sick solution, IMHO. :)

(BTW, I love Zabbix and Piwik. Use them both. Only I think that having no good alternate data store at the time of their writing, their data storage is suboptimal.)

Perhaps "monopoly" is a better word here than "tyranny".