I can't speak for all graph stores, but in addition to ADB being json-native as described below, many graph users describe the types of any value, not just floats or ints, as a relation or property to the value itself.
So you'd never have any value; object, key, value of a string, int, float, or reference, without associated meta-data typing it elsewhere in the graph, and would be unlikely to operate on that data without making reference to those properties.
More beef to be added to this until the GA release.
The benefits of distributed COLLECT will come into play for queries that can push the aggregate operations onto the shards.
Previous versions of ArangoDB shipped all documents from the database servers to the coordinator, so the coordinator would do the central aggregation of the results from all shards to produce the result.
With distributed COLLECT we now create an additional shard-local COLLECT operation that performs part of the aggregation on the shards already. This allows sending just the aggregated per-shard results to the coordinator, so the coordinator can finally perform an aggregation of the per-shard aggregates.
This will be beneficial in many cases when the per-shard aggregated result is much smaller than the non-aggregated per-shard result.
Following is a very simple example. Let's say you have a collection "test" with 5 shards and 500k simple documents that have just one numeric attribute (plus the three system attributes "_key", "_id" and "_rev"):
db._create("test", { numberOfShards: 5 });
for (i = 0; i < 500000; ++i) {
db.test.insert({ value: i });
}
Running a query that will calculate the minimum and maximum values in the "value" attribute can make use of the distributed COLLECT:
FOR doc IN test
COLLECT AGGREGATE min = MIN(doc.value), max = MAX(doc.value)
RETURN { min, max }
The database servers can compute the per-shard minimum and maximum values, so they will each only send two numeric values back to the coordinator.
Without the optimization, the database servers will either send the entire documents or a projection of each document (containing just each document's "value" attribute back) to the coordinator. But then each shard would still have to send 100k values on average.
With a local cluster that has 2 database servers and runs them on the same host as the coordinator, this simple query is sped up by a factor of 2 to 3 when the optimization is applied. In a "real" setup the speedup will be even higher because then there will be additional network roundtrips between the cluster nodes. And in reality documents tend to contains more data and collections tend to have more documents. If this is the case, then the speedup will be even higher.
I wouldn't call the stereotypical government site a "nice layout". It seems like more often than not my experience with government sites is that I have no idea how to find the data I need.
Try finding a campground in Washington in the early spring. Is it open? Do they take reservations? Are they required? Good luck.
I use the same page -- well, the same graphs -- a few times a day (usually when trying to decide whether to ride the Harley or drive the truck). I can find out everything I want to know in just a few seconds.
Any way to narrow down the location further than the zip code? My zip code covers a large area touching the ocean, there's a dramatic weather difference between the coast and 5 miles inland.
If you want to go all the way back, UNC still hosts ibiblio.org, which has links to the first website at CERN http://info.cern.ch/ and TBL's first page.
Just want to add support. One of my childhood friends has a felony from when he was an overly-rambunctious teenager that he still gets punished for - including being kicked off AirBnB - for something he stole more than 20 years ago. Despite this, he's a very successful leader in mental health services management.
So many people deserve a chance to redeem themselves from being 'branded', yet are denied the exact opportunities that would allow them to do so. This problem goes back a long, long ways.[1]
Anything you can do to help is great. Best of luck!
Unfortunately some states (eg: Texas) don't allow you to expunge a felony no matter how long ago it was. My sister-in-law's HS boyfriend was a criminal jerk and she got caught up and charged as an accessory for something stupid (theft IIRC). She fully admits she was rebelling by going after a bad-boy. She had a public defender who met with her once and told her to take the deal to get probation. Unfortunately the plea was for a felony.
It's been 20 years, she's married to a good guy and has a baby... yet that black mark still comes up on her record.
Arizona, too, and you’ll find that’s common in tough-on-crime red states. I can petition to have mine “set aside,” but it sticks. I’m fortunate enough to have a great career in the Valley, but it comes up all the time as the upthread comment pointed out.
We are looking into gubernatorial pardon (Arizona has a decent process) but not holding my breath.
This is a pretty common American attitude found in many "non red" states as well: As soon as someone commits a crime--any crime and just once--they become a criminal. It's as if their species permanently changed from human to something else. They're not a human that made a bad decision, they are an "other". Since this new thing they have become is not human, all kinds of inhumane and terrible things can be done to them and justified, including permanent removal of rights, brutalization and rape in prison, permanent loss of employability and access to normal livelihood. All of these things are seen as OK because it's a criminal we're talking about, not an actual person.
The US criminal justice system focuses on punishment and not reform, "the box" on employment applications makes this patently obvious as does the removal of voting rights for felons. All of this stems from exactly what you mentioned, once convicted you are a criminal, that label follows you and there's little you can do to get rid of it.
The first step to improving any of this is changing deeply held beliefs by our society, and many days it feels like an impossible task.
As someone who broke a whole lot of laws in his late teens and got off pretty easy for all of it, I wholeheartedly agree. They could have thrown the book at me but they didn't. Also, it helps if you don't confess to anything and hire a good attorney even if you end up taking a plea.
Let's face it, America sucks in a ton of ways and the biggest crooks are bankers and people in finance--they stole billions and got away with it over and over again. So nothing else quite compares except maybe whatever goes on with the CIA and drugs.
While I agree there are bigger crooks who go unpunished, the solution is not to let the smaller crooks go free, but to also catch the bigger ones. BTW just to be clear, I'm not advocating harsher punishment, just equal punishment.
I worked under a lot of physicians and I have a deal of respect for most of them, but yea I agree also.
I was in an administrative heavy medical lab and most attempts to streamline processes were met with a lot of backlash from the doctors. My director was told by several of them that he was "undermining and ruining medicine."
FHIR is probably the best we are going to do for a long while to come.
If you're ever interested in learning more, my company builds the national reference implementation of SMART on FHIR for the Harvard DBMI, in support of the NIH 'Sync For Science' Pilot - happy to talk all about FHIR and data exchange standards!
The participating vendors will be launching pilots this year and things are proceeding well. If we can give them some positive attention for it, maybe we'll see some progress. I sure hope so!
I've been a part of a few standards/interop efforts. Postel's Law and my experiences indicate those efforts are non-starters.
I also participated NHIN's first competitive interop whatever it was called. (We were sub to Northup Grumman. Our team won. Yay. That recognition plus $3 will get you bad cup of coffee.)
The secret to interop is to ignore the standards. No one else cares, so why should you?
For inbound, just capture whatever is sent, use web scrape style tools to extract the bits you care about. For outbound, manually create payloads that work (maybe with SOAP-UI and then tweak), turn them into templates.
If you're a monster org like Boeing or Toyota, you can rule by fiat. Everyone else has to fake it.
Where do you see things headed in terms of FHIR adoption? Personally I think SMART/FHIR is fantastic; getting vendors to actually use it is the real challenge...
I wasn't aware of this NIH 'Sync for Science' program though, very neat!
Without bickering over MU requirements, you'd probably note that MU has been almost entirely ineffective at solving actual care problems with data exchange.
Of 50 state exchanges, most have collapsed, with only occasional use of the ad-hoc implementations and networks that remain. And people dying everyday while vendors like ECW get wrist-slaps for lying to CMS about their MU compliance.
When someone asks us to integrate with a small EHR, the first thing we look at is ONC certification. We at least know that they meet some base level of functionality.
It seems to be the exact opposite philosophy that HIPAA originally took in the effort to get financial/administrative transactions to electronic formats, which was essentially that mandating interoperability would drive mainstream use.
Just-graduated MD/MPH and former EMT and ED admin here. I empathize strongly, since the first time I watched a 16-year girl who'd been in a MVC and first transported to Duke, who needed sub-specialty consults at UNC get re-scanned.
I'd read about CT dosing and realized that we had just increased this girls lifetime likelihood of ovarian or endometrial cancer by perhaps about 1/1000. For absolutely no reason other than the fact we couldn't get the images 8 miles down the road from the Duke ED, and the attending wanted to 'just be sure'.
I asked the resident why they couldn't send them digitally and they just laughed. That was 2009.
It's 8 years and a few hundred billion of national EMR spend later - and you know what? We still can't send an image between the two EDs. UNC and Duke were the first two nodes/servers on Usenet back in 1980[1], and 37 years later we can barely exchange medical data using our combined 1.3 billion dollars of Epic EMR implementations.
This problem harms people needlessly every day. Please use your voice as a provider to remind people whenever possible.
So you'd never have any value; object, key, value of a string, int, float, or reference, without associated meta-data typing it elsewhere in the graph, and would be unlikely to operate on that data without making reference to those properties.