Hacker News new | past | comments | ask | show | jobs | submit | itaborai83's comments login

To be fair, some NoSQL solutions were being sold marketing wise as the be-all and end-all of data solutions. Just google "mongodb mysql migration" and look how everyone is/was so eager to jump on the non-relational bandwagon. Some backlash was to be expected, after all, we might have reached the Trough of Disillusionment


Wouldn´t it make more sense to build it in Brazil given the potential market?


yes, but in the long term.. Chile is often used as guinea pig to do this sort of things, so before go all the way and start a huge data center in Brazil, it pays off to try it in a smaller scale somewhere else in the region and learn from it, Chile is perfect for that, it has the political stability, infrastructure and just enough population to have a decent amount of trained profesionals available..

Another point is, the data center will serve all LatAm not just Chile, so what really matters is where in the region you can get the fastests and more reliable bandwidth links to the US at the lowest price..


Data centers usually can be considered a long term type of thing and I´m pretty sure that Google is not going to build a "smallish" one, but I see your point regarding bandwidth and links.

You seem to imply that Brazil may be lacking political stability, infrastructure and/or trained professionals, but I don´t know to what extent this would be true. You have to consider that, aside from the economy, Brazil has up to 34% of all latin american population with the vast majority of it living on the eastern coast line. Placing a data center on the west coast does not add up.

There might be some other underlying reason why Google decided to settle in Chile. Maybe the cooler climate can reduce their operational costs.


The creator of Lemon is Richard Hipp, the same guy behind Fossil, SQLite and (the probably now defunct) UnQL


I´m not necessarily defending the movie, but I don´t think that Janek actually "solved" it, but I do think that that´s what they want us to believe. The black goo is not an weapon IMHO, but a way to accelerate the evolutionary process in a planet, bringing forth more adaptable lifeforms (without preserving the previous ones). That´s what is shown in the beginning of the movie. The engineers could be responsible for rapid evolutionary bursts throughout the history of our and other worlds.

Perhaps we, as a species, weren´t simply done yet. Something still had to burst out our collective chests in order for us to actualize our genetic potential. Maybe the xenomorphs were the ultimate monkeys out of which the definitive mankind would evolve from.

Piss poor execution though.


I understand the necessity of writing the couchstore component in C for performance, but in what ways does it differ from the storage engine? In other words, why have 2 different things doing apparently the same thing?

I´m curious to see how this cluster aware incremental map/reduce performs in "web scale" workloads.

Looking forward to see more of this.

p.s.: Is there a roadmap?


You made me think about something. A CoffeeScript like approach to build a saner language that would sit atop SQL would be something definitely worth having. Maybe this could be the start of the "OnSQL" movement. Just my thoughts.



Well, SQLAlchemy is designed just for that. In the words of the author:

> Like "power steering" for SQL. Doesn't teach you how to drive!


Being able to parameterize table and column names in queries would be a big help.


It could have a sharding aware data definition language and some support for "on-the-fly" data migrations.



HATEOAS yes, "Hypermedia API" no -- reducing the distributed state engine of HATEOAS at the same level of RPC-style APIs doesn't adequately convey the purpose of HATEOAS


what parts? I'm interested about the future of Couch(Base|DB) sans all the Apache wankery.

I really wish you'd take more of a BDFL approach to things.


Specifically we're working to get Erlang out of the data path. So the code where throughput and latency really matter is written by hand and compiles directly to machine code, and Erlang continues to do what it does best - manage distributed systems and asynchronous processes.

In theory a great VM could out-perform C, but after you've chased down a few Erlang VM WTFs, there's something nice about being closer to the metal.


I'm curious. Would be willing to elaborate a little more?


but aren't nodes supposed to crash, albeit in a tolerable way? Even though I conceptually like Riak I haven't tried it yet, because the problems that I deal with aren't worth the trouble of setting up a cluster and managing it.

I think that the cluster is supposed to be stable. The nodes, not so much.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: