cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Lin <wool...@gmail.com>
Subject Re: Re: Dynamic Columns
Date Thu, 22 Jan 2015 01:11:26 GMT
I've written my fair share of crappy code, which became legacy. then I or
someone else was left with supporting it and something newer. Isn't that
the nature of software development.

I forget who said this quote first, but I'm gonna borrow it "only pretty
code is code that is in your head. once it's written, it becomes crap." I
tell my son this all the time. When we start a project we have no clue what
we should have known, so we make a butt load of mistakes. If we're lucky,
by the third or forth version it's not so smelly, but in the mean time we
have to keep supporting the stuff. Not because we want to, but because
we're the ones that put the users through it. Atleast that's how I see it.

having said that, at some point, the really old stuff should be deprecated
and cleaned out. It totally makes sense to remove thrift at some point. I
don't know when that is, but every piece of software eventually dies or is
abandoned. Except for Cobol. That thing will be around 200 yrs from now



On Wed, Jan 21, 2015 at 6:57 PM, Robert Coli <rcoli@eventbrite.com> wrote:

> On Wed, Jan 21, 2015 at 2:09 PM, Peter Lin <woolfel@gmail.com> wrote:
>
>> on the topic of multiple incompatible API's I recommend you look at
>> SqlServer and Sybase. Most of the legacy RDBMS have multiple incompatible
>> API. Though in some cases, it is/was unavoidable.
>>
>
> My bet is that the small development team responsible for Cassandra does
> not have anything like the number of contractual obligations that
> commercial databases from the 1980s had. In other words, I believe having
> two persistent, non-pluggable (this attribute probably excludes various
> "legacy" APIs?) APIs is far more "avoidable" in the Cassandra case than in
> the historic cases you cite. I could certainly be wrong... people who
> disagree with my assessment now have a way to make me pay for my wrongness
> by making me donate $20 to the Apache Foundation on Jan 1, 2019. [1] :D
>
> =Rob
> [1] Project committers/others with material ability (Datastax...) to
> affect outcome ineligible.
>
>

Mime
View raw message