Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

MongoDB 2.0 Should Have Been 1.0

luigimontanez.com

Post reply on HN
Print view View original

11–20 of 51 posts

Re: MongoDB 2.0 Should Have Been 1.0

#11

Post by calibraxis »

People should note that yesterday's anonymous pastebin "Don't use MongoDB" article was apparently a hoax, if you look through the comments. (At least so claims the troll who posted it.) (http://news.ycombinator.com/item?id=3202081)

I used MongoDB last year and it worked fine for me. (I maintained it for about 9 months.) But of course I can't generalize that to other people's experiences, so YMMV. ;) I just used it squarely in the use-case which everyone mentions — many "well-behaved" writes which occur when no one's reading. Multiple replicas. Leveraged its indices. It wasn't an authoritative source of data, so in principle I could repopulate it. Dealt with failure modes, so missing data wasn't catastrophic.

I think articles like this are useful; when evaluating software, one thing I do is assume it's buggy and its proponents are deceivers. (Whether or not they intend to deceive. I can imagine being corrupted by wanting something to be true.) So among other things, I hunt down criticisms. Had this article been around last year, I'm sure I would've found it a useful hub in doing this due diligence.

(The actual thesis, of what the version number should be, is not so important to me... Version numbers are in some sense arbitrary.)

Re: MongoDB 2.0 Should Have Been 1.0

#12

Post by leif »

latch wrote: post #5

I'm not a big fan of having version numbers have some type of special meaning. To me, your post implies that there's a line in the sand at some magic version number with respect to due-diligence (on everyone's part). Gmail was in "beta" for years...it didn't mean anything. I understand that wiki and history disagrees with me, but it's still how I feel. Oh, and there's a chance that yesterday's drama was a hoax: http:…

The article appears to consider the version number issue symptomatic of a deeper mismanagement of expectations by MongoDB marketing.

Re: MongoDB 2.0 Should Have Been 1.0

#14

Post by dhimes »

matthewcford wrote: post #3

I've been using MongoDB for well over a year now in around 6 apps (moved on from CouchDB) and I agree prior to 1.8 it should have been made more obvious that there were still some stability issues. I have seen first hand some of the issues raised, we've had data disappear, recurring random crashes, ect. But I think the difference is 'everyone' knew that there were issues with MongoDB, you just needed to check in jira…

Why did you move from couch? I'm considering couch for a project, and am not especially knowledgeable in the space. Couch has worked fine for a low-load, minimal-functioning prototype store (no replication needs, etc.). Its scary feature to me is dealing with compacting-- how and when to schedule it so a large db won't get bogged-down.

Re: MongoDB 2.0 Should Have Been 1.0

#15

Post by rdtsc »

calibraxis wrote: post #11

People should note that yesterday's anonymous pastebin "Don't use MongoDB" article was apparently a hoax, if you look through the comments. (At least so claims the troll who posted it.) ( http://news.ycombinator.com/item?id=3202081 ) I used MongoDB last year and it worked fine for me. (I maintained it for about 9 months.) But of course I can't generalize that to other people's experiences, so YMMV. ;) I just used it…

A hoax or a double-hoax?

Original post was this by nomoremongo:

http://news.ycombinator.com/item?id=3201772

Post actually discussed was:

http://news.ycombinator.com/item?id=3202081

Very clever.

The way I understand, apparently nomoremongo wrote it but it was reposted quickly by nmongo (http://news.ycombinator.com/threads?id=nmongo) in hopes that they could then become the top post, so later on they can yell in all caps how it was a hoax and thus discredit the original post.

You know nmongo, if you are trying to help MongoDB you just did the opposite. I've said this many times here before, but the best marketing is your competitors stupid marketing. You are providing stupid marketing for MongoDB and you are hurting it.

> People should note that yesterday's anonymous pastebin "Don't use MongoDB" article was apparently a hoax,

Now another question isb do you know that it wasn't a hoax but continue in the same vein in hopes to still save the day, or you actually believe it was a hoax?

Re: MongoDB 2.0 Should Have Been 1.0

#16

Post by jimbobimbo »

"There is no doubt that MongoDB has benefitted from an aggressive marketing push. There are more MongoDB conferences held (organized by 10gen) and MongoDB books written (mainly by 10gen employees) than for the other NoSQL datastores combined."

Latest developments around MongoDB remind me the history of Oracle DB described in "The Difference Between God and Larry Ellison..." book: they basically were selling a DB that was far away from prime time. I'm not surprised at all that people may run into the issues with bleeding edge software - I figure that's the price that early adopters must pay anyway.

Re: MongoDB 2.0 Should Have Been 1.0

#17

Post by bobz »

latch wrote: post #8
kevingadd wrote: post #2

"No open source project has received more criticism in recent years than MongoDB." [citation needed] Suggesting that the 1.0 version number should have been reserved until more recently makes me think that what the author of this post is really saying is something like this: "MongoDB wasn't really production-ready until recently. People who wanted to test something bleeding-edge out in the real world should have stil…

[citation needed] indeed. My vote would be Java

"Java" is not an open source project.

As a Java programmer, I've not really heard a particular level of criticism of the OpenJDK project. Java the language sure does have its fault.

Although I suspect you were just Java bashing.

[ed] s/it's/its/g

Re: MongoDB 2.0 Should Have Been 1.0

#18

Post by justin_vanw »

feralchimp wrote: post #7

Raise your hand if you're relying on package version numbers to tell you which packages to implement in your high-volume production environment against live customer data.

Your comment would be funny, if it didn't have so much truth to it.

From now on, for all my open source code, I'm going to version with this translation table:

Alpha: Beta

Beta: increment version by random number between .1 and .2, eg: 2.0 becomes 2.1.7

Release Candidate: add the word "Enterprise Edition" to the version

Release: Add the same letter Oracle added for that version: 8 -> 8i, 9 -> 9i, 10 -> 10g. The exception is anything that happens to be a BMW model, in which case I'll just make it the same as the BMW. 3.2.8 -> 3.2.8xi, 4 -> z4, &c.

Re: MongoDB 2.0 Should Have Been 1.0

#19

Post by latch »

rdtsc wrote: post #15
calibraxis wrote: post #11

People should note that yesterday's anonymous pastebin "Don't use MongoDB" article was apparently a hoax, if you look through the comments. (At least so claims the troll who posted it.) ( http://news.ycombinator.com/item?id=3202081 ) I used MongoDB last year and it worked fine for me. (I maintained it for about 9 months.) But of course I can't generalize that to other people's experiences, so YMMV. ;) I just used it…

A hoax or a double-hoax? Original post was this by nomoremongo: http://news.ycombinator.com/item?id=3201772 Post actually discussed was: http://news.ycombinator.com/item?id=3202081 Very clever. The way I understand, apparently nomoremongo wrote it but it was reposted quickly by nmongo ( http://news.ycombinator.com/threads?id=nmongo ) in hopes that they could then become the top post, so later on they can yell in all…

There was always something suspicious about an anonymous post lacking any verifiable facts. Then, 10gen's CTO states that none of it resonates with any support issues they've had. Then, add this.

I know exactly what it will take for me to believe the pastbin story.

I'm curious, what will it take for you to not believe it?

Re: MongoDB 2.0 Should Have Been 1.0

#20

Post by cpeterso »

justin_vanw wrote: post #18
feralchimp wrote: post #7

Raise your hand if you're relying on package version numbers to tell you which packages to implement in your high-volume production environment against live customer data.

Your comment would be funny, if it didn't have so much truth to it. From now on, for all my open source code, I'm going to version with this translation table: Alpha : Beta Beta : increment version by random number between .1 and .2, eg: 2.0 becomes 2.1.7 Release Candidate : add the word "Enterprise Edition" to the version Release : Add the same letter Oracle added for that version: 8 -> 8i, 9 -> 9i, 10 -> 10g. The e…

No "Technical Preview" releases?
Post reply on HN