Issues often accumulate content-less “+1” comments which serve only to spam the maintainers and any others subscribed to the issue. These +1s serve a valuable function in letting maintainers know how widespread an issue is, but their drawbacks are too great. We’d like issues to gain a first-class voting system, and for content-less comments like “+1” or “:+1:” or “me too” to trigger a warning and instructions on how to use the voting mechanism.
Doesn't work for datacenters. Also is implemented with round-robin DNS to nodes (1-N, check X-Forwarded-For) in each AZ, which then handle balancing.
Also worth noting that unless you turn on cross-region balancing, if an AZ doesn't have a node in it and the RR DNS points them at that AZ, they'll be turned away. Additionally, without it you need to scale by multiples of AZs you run in otherwise you'll have unbalanced traffic.
On another note, I've always been curious if they're just abstractions around HAproxy at the per-node level.
Very true. Cross-AZ load balancing works quite well. I believe Amazon has said it's RR across the servers with the least connections, but degenerates to a simple RR without many nodes per AZ.
Something like TodoMVC for backends would be nice... in essence, you create a backend for a TodoMVC front end, each using the same web-server platform and language and TodoMVC front end. The difference being the back end SQL server, with as much processing on the server, if it supports procedures, as possible. Maybe extending the example for a location, and a local date/time.
Using Node.js, and Angular for the server/front end, it would be easy enough to swap out the "todo-mvc-server-data" module... as long as each supported the same interface(s), it could be a good test...
Setup the same hardware for each backend, and then run performance tests against a node cluster for the front end. It would by no means be comprehensive, but would be a nice comparison point (like TodoMVC itself).
Not sure how one can achieve ACID on distributed system without limiting scalability. In addition, these system are limited by the CAP theorem. Therefore it is bound to have problems, would be interesting to read the high level architecture.
How many people had the slightest clue that books would be replaced by Kindles, CDs by iPODs, tablets selling as much as laptops. The comments show that people underestimated the impact bitcoin would make.
I still believe its just the tip of the iceberg when it comes to Bitcoin.
But telegraphs are virtually gone in the western world, X.25 has very few installations. Tech doesn't ever fully die out, since just when it is about to a subculture grows around keeping it alive.
The printed page is centuries if not millennia away from becoming a niche subculture.
We are talking about people who have read at least one book in the last year.
40% - That's exclusively print books.
58% of them own an e-book reader, that's including the iPad.
The charts are a bit hard to read but..
Now, only about 5% bought only e-books. This of course doesn't include borrowed books, library books, etc. MUCH much more people haven't purchased either print or e-books than have purchased only e-books! I hate to generalize but some people, many in the HN crowd, seem to think libraries are "beneath" them or "outdated" but the data suggest otherwise. Paper books are not DRMed after all.
About 38% of people bought only print books. A mix between print and e-books wins out.
Here's the important stat. About 50% said they like print books and e-books. Only 10% said they like e-books only. A FULL 90% OF READERS STILL LIKE PRINT BOOKS!!!!
It's incredibly naive to say they have been replaced. Its beyond obvious anyway, just take a commercial flight and observe people around you reading print books.
I knew all of those things as soon as those devices came out.
Not kidding. And I don't see how others can't see them either.
But that is the nature of perception. You either perceive something or you don't and it's difficult to tell someone "there is a ship out there!" when their eyesight is too crappy to discern it. They end up having to trust you and trust is problematic.
Technically true, but we still have the governments to weigh in and try to crush this. Unless they can monetize from BitCoin, they will find a way to kill it - blaming it on funding organized crime and terrorism. "Real world" currency is just too important for them to loose control of it.
Yesterday's congressional hearing made me surer than ever that bitcoin is actually an FBI/ CIA/ NSA/ whoever -made honeypot.
Think about it. Bitcoin is robust enough that it could have only been developed with the resources offered by a government, the currency records all transaction histories, and we still have absolutely no idea of who this Satoshi Nakamoto guy is, assuming he's not some government taskforce. Didn't it strike anybody as odd that the same congress and government that normally takes an extreme luddite stance was so enthusiastic about Bitcoin? They can hardly wrap their heads around things like Google Glass and net neutrality, and yet they're absolutely pandering over Bitcoin.
I think it makes perfect sense if the government types think they can apply all the existing laws to Bitcoin transactions. Money laundering, tax evasion, fraud, theft, etc. are all still illegal, even if you use Bitcoin to carry them out.
I've actually changed from thinking "financial services law will destroy this" to "this will become like the War On Drugs: ubiquitous illegality, especially evasion of exchange controls, witholding taxes, and foreign income reporting requirements".
