Hm, flat files will rule the world? No more messy SQL, just iterate over the fil...

jshen · on March 30, 2009

You may not be convinced yet, nor am I, but your strawman argument makes you look ... well a little silly.

Tichy · on March 30, 2009

That is the way I understood it. What do you mean by "it creates an index for each view"? I just would like to avoid reinventing the wheel, and relational databases abstract away some of the problems of dealing with large data sets.

Edit: OK, read the CouchDB introduction now. Did not know that the "views" are server side. I have a past as a Java developer, and I have actually come across the "select all, then compute in Java" pattern several times in the real world.

jshen · on March 30, 2009

You should phrase your comments in the form of a question rather than passing judgement on things you haven't looked into.

Tichy · on March 30, 2009

I used to do that but then got negative comments from HN. Anyway, the title of the article was a bit provocative, so I hoped answering in the same spirit would not be taken as offensive. It certainly is inspiring to look into CouchDB.

jshen · on March 30, 2009

Yeah, the HN headline was stupid. Sadly stupid titles usually get a lot of up votes. By stupid I mean, intended to incite a knee-jerk reaction.

jamesbritt · on March 31, 2009

Same seems true of comments. I think I see Tichy's point; if you ask question, you are often ignored, or mocked.

If, however, you make a (willfully) ignorant assertion, people jump at the chance to tell you just how wrong you are.

Not that I want to advocate this approach, but I understand why someone would use it if, in the end, it generates more useful information.

(I think this was discussed elsewhere on HN, with "troll, don't inquire" as a disappointingly effective means of getting information.)

Tichy · on March 31, 2009

I wasn't trying to troll, I merely answered in the same tone as the submission title. Let's not blow this out of proportion.

jamesbritt · on March 31, 2009

I didn't mean to suggest you were trolling, just that some folks have found that adopting certain troll-like characteristics is (for better or worse) an effective means of getting informative replies.

intinig · on March 31, 2009

I am sorry for the headline, but you're right about stupid headlines getting reactions.

intinig · on March 30, 2009

Tichy, give it a spin. The ease of execution (and the sheer speed, even over giant amounts of data) is impressive.

dagheti · on March 30, 2009

What do you mean by impressive "ease of execution"?

How terse the code is? How easy the code is to read? How likely the code is to be correct? How flexible the data model is to new requirements? How simply you can reason about what the code does?

intinig · on March 30, 2009

The code is easy to read, and the data model is as flexible as you want it to be, since it's schemaless.

Reasoning about what the code does is simple once you shift your paradigm to document from relational.

old-gregg · on March 30, 2009

Do your research. "document-based" DBs already "ruled the world" right until relational databases were invented, which quickly obsoleted them.

The only advantage of disk-backed hash table is ease of scalability, this is why they're useful for hm.... top 0.005% of web sites, who handle thousands of updates per second.

jshen · on March 30, 2009

They are also useful for rapid app development or prototyping. I've used couchdb this way a few times and it works great.

sounddust · on March 30, 2009

I simply don't believe that this model scales "over giant amounts of data", at least in any sort of real-world usage scenario. Can you back up this claim with benchmarks?

Tichy · on March 30, 2009

Reading CouchDB documentation now and can't stop reading. It sounds like a lot of fun. Haven't understood yet how to handle the performance problems, but would like to use it just for the fun of it.

Tichy · on March 30, 2009

I can imagine some benefits, but as for speed of execution I am really doubtful. Your example suggests a "select all" followed by some analysis done in code. I can't imagine this scales well.

jshen · on March 30, 2009

It creates an index for each view essentially. It's not anything like a "select all".