How about they make a law where we repossess your wife? I mean, one failed analogy leads to another, right? It's a slippery slope down the hill of bad arguments.
I have no idea why this hasn't been done yet in the USA, but in The Netherlands it's illegal to use the SSN for any purpose, except for interactions with the state (taxes and compliance). This means the SSN is actually not that sensitive, as business are banned from relying on it.
That's how populists get voted in: their simple worldview is easier to explain.
While you're still trying to explain all the subtleties about the jobs market, they just shouted "mexicans out means more new jobs!" and "climate change is fake news!". Now you have two stories to tell while they're already on number three to seven. That' race is hard to win.
Where I work (Europe) we use [all first/middle/etc. names] + [last name] + [birth date] + [birth place] + [birth country]. That's pretty good. This is in Western Europe, however, and still not completely without false duplicates. But the few we get we can deal with manually.
The problem is that with the system of name+DOB, even if you have just 100 duplicates, only 1 is fraudulent so far. That could be an issue with 10000 duplicates or more so it's good to reduce it, but even then voter fraud is pretty much a non-issue. Unfortunately, some assholes found it convenient to use the non-issue to make it easier to de-register people. This impacts poor and uneducated people more than well-educated people and rich people, who are more prone to register in the first place.
At least in the EU there is a special addition to copyright, which is specifically aimed at databases. They are considered creative works once effort has been put into compiling them. So for at least the EU, this license makes a great deal of sense.
The most pressing problem is of course, that most open data has little documentation and is hardly any use. But hey, at least we got the legal issues fixed once a license can be slapped onto it. Who needs to actually work with it, anyway?
If it doesn't matter whether your hypothesis is true or not for the outcome of any experiment you care to design, you might as well not have the theory in the first place. A variation on Occam's razor.
Unless you can come up with a way to prove or disprove your hypothesis, there is no science in it. Feel free to call it religion though.
In this case I haven't read the court papers, but the facts of the case are pretty fishy: - He didn't have an appointment as claimed - The company he claims to work for denies knowing him - He had a video camera that he had used to video the public displays in the hall (corporate info, not that interesting, but still) - He had two laptops, a tablet, multiple external harddrives, and USB-sticks with him. He could give no reason why. - He's an IP lawyer (he claims), which means he needs to be shot at dawn no matter what.
He is very unlikely to be employed by the government (motive?). I find it much more likely he's just a freelancer who thought he'd try this too since all his friends said it was easy money to walk in on a tech company and get some data.
Must read Neuromancer again... I think Gibson described professions like this, actually.
Every experience I've ever had in working with Oracle was painful. If you need to hire a "license manager" to make sure you don't get shafted by Oracle later on, things really start to suck. A co-worker at a different company tried to switch to a cheaper option: The Oracle salesmanager just said that yes, there was that option but in their case they got a "very special price, my friend" that made the cheaper option exactly the same price as their old contract.
It's practices like that, that people consider them a bunch of crooks.
I've worked with Oracle ever since I started in IT, but I have to say: they're not just crooks when it comes to licenses, they're also lagging behind on the database. As far as I can tell they're in the cashcow phase.
Microsoft SQL Server is still playing catchup in some cases (I have a list of 5 items I'd *REALLY* want to have in SQL Server,such as their default isolation, separation between data and metadata for case sensitivity, autonomous transactions etc.), but the (mostly) hasslefree experience, great integration with good tooling and the pretty hardcore bugtesting they do (they first run the db on azure) compensate for that.
At first I'd thought it would be unpleasant to get used to SQL Server, but it really isn't and you get used to it very fast. Just try to map a number of architectural layers into your database. With Oracle it's pretty much over after the schema. Add another database, if you're lucky and your DBA likes you. After that it's all naming conventions. But database/database connections in SQL Server are painless, the schema's are actual namespaces and not logins, and if you really need naming conventions you have a very long name to use for that. That's just one little thing that helps a lot.
And it is much more consistent in its plans than Oracle ever was, in my experience. With SQL Server 2016, even more so.
I was pretty much convinced that Oracle was the best DB there was. But even SQL Server 2012 is better IMO than Oracle, in a lot of areas. With SQL 2016 and its pinnable plans, Polybase, virtual tables on Azure and R integration, the gap just widens.
I would use the new encrypted computation library by Philips.
Encrypted values go in, encrypted values go out, and only the customer will be able to decrypt the results. Good luck hacking that. The downside is it uses more computing power. But it is extremely secure.
I doubt they will get away with it. If I have a mortgage and the bank says "oh, if you don't give us an extra 400 dollar each month your house belongs to us" it's not "just a statement". Neither is this. It relates directly to the software, and as such is likely covered by the GPL. IANAL but given the ramifications if they can do this, I doubt it works like that.
Said scheme is usually countered by the local IRS by stating that "companies of type X make profit percentage y% normally, so we will levy taxes as if you do that too, since you are obviously operating more than x months now, and not bankrupt". There's also a law over here that basically says that any construct of companies used solely to evade taxes can be declared null and void for the purpose of assessing taxes. I worked for a multinational once and the tax authorities just used both rules. Tax authorities aren't stupid. Usually they employ a lot of sharpwitted financial experts because they earn more than they cost.
I actually like JavaScript a lot. It's a nice functional language that has an IDE literally on every computer on the planet. We could do a lot worse... like Java, in the browser. Applets, anyone?
I work in The Netherlands and German culture is about the same, in respectable companies. That manager would have been in quite serious trouble in almost any company I have ever worked for, and as consultant I worked at quite a few.
How about they make a law where we repossess your wife? I mean, one failed analogy leads to another, right? It's a slippery slope down the hill of bad arguments.
Some projects take that long to complete.
Do you work for Accenture by any chance?
I have no idea why this hasn't been done yet in the USA, but in The Netherlands it's illegal to use the SSN for any purpose, except for interactions with the state (taxes and compliance). This means the SSN is actually not that sensitive, as business are banned from relying on it.
Maybe your parents should have stopped inbreeding first...
To fix your lack of education on names, please read this: http://www.kalzumeus.com/2010/...
It's even on-topic.
That's how populists get voted in: their simple worldview is easier to explain.
While you're still trying to explain all the subtleties about the jobs market, they just shouted "mexicans out means more new jobs!" and "climate change is fake news!". Now you have two stories to tell while they're already on number three to seven. That' race is hard to win.
Where I work (Europe) we use [all first/middle/etc. names] + [last name] + [birth date] + [birth place] + [birth country]. That's pretty good. This is in Western Europe, however, and still not completely without false duplicates. But the few we get we can deal with manually.
The problem is that with the system of name+DOB, even if you have just 100 duplicates, only 1 is fraudulent so far. That could be an issue with 10000 duplicates or more so it's good to reduce it, but even then voter fraud is pretty much a non-issue. Unfortunately, some assholes found it convenient to use the non-issue to make it easier to de-register people. This impacts poor and uneducated people more than well-educated people and rich people, who are more prone to register in the first place.
Zing...
At least in the EU there is a special addition to copyright, which is specifically aimed at databases. They are considered creative works once effort has been put into compiling them. So for at least the EU, this license makes a great deal of sense.
The most pressing problem is of course, that most open data has little documentation and is hardly any use. But hey, at least we got the legal issues fixed once a license can be slapped onto it. Who needs to actually work with it, anyway?
If it doesn't matter whether your hypothesis is true or not for the outcome of any experiment you care to design, you might as well not have the theory in the first place. A variation on Occam's razor.
Unless you can come up with a way to prove or disprove your hypothesis, there is no science in it. Feel free to call it religion though.
get-rich-quick I think
In this case I haven't read the court papers, but the facts of the case are pretty fishy:
- He didn't have an appointment as claimed
- The company he claims to work for denies knowing him
- He had a video camera that he had used to video the public displays in the hall (corporate info, not that interesting, but still)
- He had two laptops, a tablet, multiple external harddrives, and USB-sticks with him. He could give no reason why.
- He's an IP lawyer (he claims), which means he needs to be shot at dawn no matter what.
He is very unlikely to be employed by the government (motive?). I find it much more likely he's just a freelancer who thought he'd try this too since all his friends said it was easy money to walk in on a tech company and get some data.
Must read Neuromancer again... I think Gibson described professions like this, actually.
I had the exact same feeling. Why would I want to store unstructured data in a relational database? I'd rather put the video and sound somewhere else.
Hear, hear!
Every experience I've ever had in working with Oracle was painful. If you need to hire a "license manager" to make sure you don't get shafted by Oracle later on, things really start to suck. A co-worker at a different company tried to switch to a cheaper option: The Oracle salesmanager just said that yes, there was that option but in their case they got a "very special price, my friend" that made the cheaper option exactly the same price as their old contract.
It's practices like that, that people consider them a bunch of crooks.
I've worked with Oracle ever since I started in IT, but I have to say: they're not just crooks when it comes to licenses, they're also lagging behind on the database. As far as I can tell they're in the cashcow phase.
Microsoft SQL Server is still playing catchup in some cases (I have a list of 5 items I'd *REALLY* want to have in SQL Server,such as their default isolation, separation between data and metadata for case sensitivity, autonomous transactions etc.), but the (mostly) hasslefree experience, great integration with good tooling and the pretty hardcore bugtesting they do (they first run the db on azure) compensate for that.
At first I'd thought it would be unpleasant to get used to SQL Server, but it really isn't and you get used to it very fast. Just try to map a number of architectural layers into your database. With Oracle it's pretty much over after the schema. Add another database, if you're lucky and your DBA likes you. After that it's all naming conventions. But database/database connections in SQL Server are painless, the schema's are actual namespaces and not logins, and if you really need naming conventions you have a very long name to use for that. That's just one little thing that helps a lot.
And it is much more consistent in its plans than Oracle ever was, in my experience. With SQL Server 2016, even more so.
I was pretty much convinced that Oracle was the best DB there was. But even SQL Server 2012 is better IMO than Oracle, in a lot of areas. With SQL 2016 and its pinnable plans, Polybase, virtual tables on Azure and R integration, the gap just widens.
To clarify: the input is NEVER decrypted. All computations are done on the encrypted values.
I would use the new encrypted computation library by Philips.
Encrypted values go in, encrypted values go out, and only the customer will be able to decrypt the results. Good luck hacking that. The downside is it uses more computing power. But it is extremely secure.
Ew... yeah, that would have killed any inclination of actually doing business with that company for me as well.
I guess they'd have been better off if they had just repackaged a generic no-brand headset and sold that. At least it would have worked better...
I doubt they will get away with it. If I have a mortgage and the bank says "oh, if you don't give us an extra 400 dollar each month your house belongs to us" it's not "just a statement". Neither is this. It relates directly to the software, and as such is likely covered by the GPL. IANAL but given the ramifications if they can do this, I doubt it works like that.
It seems to be :)
Said scheme is usually countered by the local IRS by stating that "companies of type X make profit percentage y% normally, so we will levy taxes as if you do that too, since you are obviously operating more than x months now, and not bankrupt". There's also a law over here that basically says that any construct of companies used solely to evade taxes can be declared null and void for the purpose of assessing taxes. I worked for a multinational once and the tax authorities just used both rules. Tax authorities aren't stupid. Usually they employ a lot of sharpwitted financial experts because they earn more than they cost.
I actually like JavaScript a lot. It's a nice functional language that has an IDE literally on every computer on the planet. We could do a lot worse... like Java, in the browser. Applets, anyone?
So... show me on the doll where the bad gay touched you.
I work in The Netherlands and German culture is about the same, in respectable companies. That manager would have been in quite serious trouble in almost any company I have ever worked for, and as consultant I worked at quite a few.