Although PostgreSQL is featured database enterprises still using mainly SQL Server or Oracle. http://blog.protegra.com/2012/06/22/sql-server-2012-cheaper-... What can we do for wider adoptation of PostgreSQL?
Again, there is no official Mysql Workbench for Postgres, let alone things like SequelPro. (although a member on the Indian HN - Hackerstreet - is building one at http://pgxplorer.com/)
Second - documentation as compared to other database vendors. No, this is not flamebait - by itself, it is very good and adequate. But for people used to MySQL, SQL Server - it feels...scary.
Check out http://dev.mysql.com/doc/ (sidenote: has docs in epub format as well) vs http://www.postgresql.org/docs/
What is scary about it? Genuinely curious since I started out with MySQL and then when I started using PostgreSQL thought the PostgreSQL documentation was the best ever. It probably is a matter of preference, but I am curios what is scary.
EDIT: As for your two links the MySQL page is way scarier.. So many options, so many links.
Your toplevel comparison of the index of both manuals is not very informative to me. What do you feel is missing from PostgreSQL's manual that you find in other documentation sources?
A simple example is the getting started part http://dev.mysql.com/doc/refman/5.5/en/creating-database.htm... vs http://www.postgresql.org/docs/9.2/interactive/tutorial-crea... Notice that Postgres has switched to using system commands like "createdb" while mysql sticks to SQL - there is a perfectly viable way to do this in Postgres using SQL, but perhaps the author didnt want to go directly to explaining template databases (which is simpler IMHO).
Go and ask someone who is new to both Mysql and Postgres to read these sections. Based on the distro they are on, you have completely lost them between the complexities of the OS vs the database Here is an example of the kind of confusions that result - http://superuser.com/questions/220771/how-can-i-create-a-dat...
Check this section out - http://dev.mysql.com/doc/refman/5.5/en/privilege-system.html. The reason why I am picking up client authentication example is that this where 90% of beginner postgres problems occur: they happened to me and they happen to every other first timer. Is postgres more complicated ? No, but you need to explain it better. I find this to be a much more useful tutorial to point first timers to, even if they are not on Ubuntu : https://help.ubuntu.com/10.04/serverguide/postgresql.html
http://dev.mysql.com/doc/refman/5.5/en/connecting-disconnect...
Whereby it's presumed that if you cannot get a connection you should talk to your administrator and give up. That may not be bad advice: the exhaustive covering of error cases by postgres tutorial may detract from rather than enhance the exercise.
I agree that Postgres' default authentication system one uses for development -- based on the account name -- probably deserves a different kind of explanation. It's an approach considerably better that most other software that it's probably worth keeping (especially for someone who hates more passwords), in spite of figuring out a way to explain it.
All in all, I basically don't agree with the notion that it is "abridged." It could be improved, but I think you are repeatedly using these words describing the scantness of the documentation whereas I think it is borderline over-detailed as it is. Perhaps a different kind of information presentation is required, is what I think.
Although still nascent, this lack of narrative documentation is probably why Craig started writing http://postgresguide.com/. This doesn't do a good job of explaining how to install the software, but is a lot more conversational than the Postgres documentation:
http://postgresguide.com/SQL/select.html
vs
http://www.postgresql.org/docs/9.2/static/sql-select.html
Clearly, one is more complete in some sense, but it's not going to help the beginner.
For starters, acquire the reputation for providing high-touch, enterprise-grade, enterprise-focused, enterprise-costs support, with all that entails, and preferably with a cost somewhat below what Oracle charges.
Provide enterprise-friendly DBA training, certifications, and other related materials.
Have your sales reps schmooze the C-level office, and the IT leadership.
But adoption of databases in enterprises in general? They're loathe to change their databases, and have massive investments in their existing software and hardware processes, and their staff and training. The costs involved here are staggering.
If you're looking for more organic growth (and as some of those smaller companies grow and get bigger), that'll happen. But it'll take years. Maybe some acceptance in smaller projects in enterprises, but this sort of "shadow IT" can be a guerrilla effort in an enterprise; discouraged by management.
Most of those things are basically a scam (and often also a racket).
> Have your sales reps schmooze the C-level office, and the IT leadership.
And there you got your answer: basically, bribery.
I work in enterprise and support/training is vital for maintaining 24/7 support coverage and ensuring that developers have the best knowledge from the vendor.
For costs the article really only shows one thing: Oracle is way more expensive than the rest.
For the scoring model the article just asks us to trust in them since they do not provide the results of the feature scoring method.
What's needed is more "PostgreSQL for SQL Server People" type conversion material and tools to help facilitate this process. NaviCat is not an acceptable solution here.
Years ago when I switched to MySQL it was because it was easy to get going, a lot of community support around the product. You have questions, you get answers almost immediately.
Postgres is a much more sophisticated offering and still needs to file down a lot of the sharp edges on their toolset. The psql command line tool for one is very bewildering to the uninitiated and has two conflicting syntaxes, one for the backslash commands and one for actual queries. This is not unlike other SQL shells, but MySQL, while highly non-standard, does have the advantage of a unified interface here.
There's legitimate reasons to buy from the big three DB providers: there's a shortage of quality DBAs for postgres, and of course there are the 5-10% of cases that actually need the features/performance of the big 3. Evaluations like this, however, do not make a good case.