On Thu, February 5, 2009 9:38 am, Lawrence Sica wrote:
>
> On Feb 5, 2009, at 12:24 PM, Patrick Coskren wrote:
>
>> On Feb 5, 2009, at 12:21 PM, Lawrence Sica wrote:
>>
>>> MySQL is, to be frank, shittastic.  I never understood
>>> the joy and awesomeness that some people think that it is made up of.
>>
>> Short learning curve and very quick for tasks with high-frequency
>> reads, low-frequency writes, and no strong need for data integrity.
>>
>
> That short learning curve is part of the problem.  You have people
> managing databases with no actual understanding of what that means as
> a result.  Not that something needs to be hard, but it needs to not be
> too simple.

Well, I'm not sure if you mean me.. I'll readily admit I'm not a DBA. But
I deal in data, and have built a number of database-backed applications
over the years - typically internal applications, supporting production
environments and clients. I know my limits, and in my current environment
I'm happy to have far more skilled people than I am to do the real hard
work.

But in between, there's a set of needs - for instance, right now I'm
working on cleaning up a complete wreck of a database - the schema was
clearly designed by a committee of unstable, sadistic morons (a commercial
product, not one of ours). The data itself is a disaster of epic
proportions. I've got to simultaneously clean up live data - normalizing
columns with often 5 or 6 completely different types of data stored within
- maintain existing schema and data, and prepare for a migration to a sane
application. And I'm uniquely in a position to do this, as I have the best
sense for the data as it is, and as it should be.

I'd love not to be touching the database at all, but I don't know of any
reasonable alternatives. So, I'm here bitching about the relatively steep
approach SQL Server requires for what are otherwise undemanding needs - I
have a good grasp of SQL fundamentals, and am not trying to be an
architect, just a technical business user. Also, I'm reminded frequently
that SQL wasn't intended just for DBAs - it was meant as a general purpose
query language for people like me... and I tend to have no problem with
the language itself, with grokking complex databases. It's just the tools
that occasionally get in the way.

>
>> There are tasks for which that's a good set of tradeoffs.  That
>> describes what a lot of web developers think they need.  (I believe a
>> lot of web developers underestimate their need for data integrity
>> until it bites them in the ass.)
>
> I've found not just web developers underestimate this.  It does amaze
> me how data-centric the web is and how much people seem to not really
> think about the value of it.
>
> --Larry
> _______________________________________________
> OSX-Nutters mailing list | [email protected]
> http://lists.tit-wank.com/mailman/listinfo/osx-nutters
> List hosted at http://cat5.org/
>


_______________________________________________
OSX-Nutters mailing list | [email protected]
http://lists.tit-wank.com/mailman/listinfo/osx-nutters
List hosted at http://cat5.org/

Reply via email to