Hacker News new | past | comments | ask | show | jobs | submit login

What are you actually trying to say? Not everything in your post is a full truth.

MS's graphical query plan tool is actually pretty awful and hasn't been updated since forever, quite often hiding the most important information, like estimated rows vs. actual rows. It takes quite a while to learn how to actually read it unless someone shows you. Like months. And it's really just a wrapper around SHOW PLAN, which MySQL has with EXPLAIN.

I'm not convinced that SQL Server Data Tools easily manages schemas, I've only been using it a few months but it seems a massive hack to me which relies on constant schema compares.

MS also constantly push you to use the graphical tools, but if you play in the tools and then try to make something repeatable it's a right PITA. Your only option is to turn on the incredibly annoying 'Autogenerate change scripts' whereas in MySQL you can just look at the history tab in Workbench which shows every command you've run. Imagine that. Everything as SQL scripts you can modify and re-run!

Finally, SQL Server Management studio is an absolute dog performance wise, I actually still miss Query Analyser which was much better IMO.

And you can't actually use SQL Server with something like Ruby or Python without quite a massive amount of effort. I know because I've tried, there's no decent, easy to install, equivalent of ADO.Net without mucking around with ODBC connections. At least a year ago there wasn't.

As for the problems with MySQL, he mentions in the post the problem with MySQL is the lack of certain types of operations.

Also there's a bunch of gotchas, sub-selects perform terribly in MySQL, which makes the EF unusable on it.

Still, I've used both and SQL Server is much better if you're developing with C#.




I work with SQL Server and its parts (SSIS, SSAS, etc.) pretty much exclusively at my day job, and I occasionally find myself hacking up Python scripts to automate various tasks.

I don't know about a year ago, but several months ago I found pyodbc [1] and it's been wonderful. Easy to set up, works a lot like the mysql functions in PHP.

[1] https://code.google.com/p/pyodbc/


Oh this looks like a modern solution

But really, Python + MSSQL = Bag of hurt.

It's a real pity, but I see the "market" for that being very small...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: