Re: [HACKERS] RustgreSQL - Mailing list pgsql-hackers

From Joshua D. Drake
Subject Re: [HACKERS] RustgreSQL
Date
Msg-id fad1a999-5555-d599-e471-5e4e0e3c1d72@commandprompt.com
Whole thread Raw
In response to Re: [HACKERS] RustgreSQL  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: [HACKERS] RustgreSQL
List pgsql-hackers
On 01/10/2017 08:12 AM, Robert Haas wrote:

> Really?  What language would you pick in a vacuum?  The Linux kernel
> is written in C, too, for pretty much the same reasons: it's the
> canonical language for system software.  I don't deny that there may
> be some newer languages out which could theoretically be used and work
> well, but do any of them really have a development community and user
> base around them that is robust enough that we'd want to be downstream
> of it?  C has its annoyances, but its sheer pervasiveness is an
> extremely appealing feature.

If we boil this down, I don't think any of this idea has to do with the 
fact that our database is written in C. I think it has to do with C is 
no longer "hip". We don't want to be hip. We are database people. Leave 
hip to MongoDB.

We want performance, stability, maturity and portability. (Not 
necessarily in that order).

There is not a single above hardware language (E.g; let's not rewrite in 
assembly) that provides those four requirements.

Rust is awesome. It is also 5 years old.
Go is awesome. It is also 8 years old.

C is awesome. It is 39 years old.

In human terms, C is the only one of these that has been around long 
enough to realize it isn't a teenager (or child really), and although 
you may still be able to do the things you could in your 20s, you are 
going to pay for them the next day.

JD

-- 
Command Prompt, Inc.                  http://the.postgres.company/                        +1-503-667-4564
PostgreSQL Centered full stack support, consulting and development.
Everyone appreciates your honesty, until you are honest with them.
Unless otherwise stated, opinions are my own.



pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: [DOCS] [HACKERS] Questionable tag usage
Next
From: Robert Haas
Date:
Subject: Re: [HACKERS] parallelize queries containing subplans