Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
×

Comment Re:So what is the answer? (Score 1) 155

The definition of a Taxi service is definitely not "a service you can hail on the streetside".

That is basically the only significant difference between a taxi and a limo service. You need a medallion to be able to pick up random strangers hailing you. You do not to pick up someone who has registered ahead of time and called you up. While you can call a taxi on the phone ahead of time, you can also call up a chinese limo service in NYC that is basically better in every way (you can also pick one up at designated, static locations). You, however, cannot hail one in the street. Because they are not a taxi service.

There are other small differences, but that is the only relevant one here. You do not need to be a registered taxi to get hired as a driver ahead of time by someone on the phone. Cities and lobbies are getting their panties in the bunch because phone apps made ahead of time reservation even more convenient than waving on the street. They did not expect that to happen when the original rules were put in place.

Comment Re:So what is the answer? (Score 1) 155

Can you hail a Uber as you see it go through the street by waving at it?

No, you can't. Ride sharing service is not the right term for it, but neither is Taxi.

Then you have the distinction between part timers using their own car to give rides every now (eg UberX) and then vs limousine service (eg: Uber Black).

So how do you call UberX? Make up a name, it doesn't matter. But its definitely not a Taxi service. At best they should follow limousine rules, but they very well may be something else.

Comment Re:Disabled (Score 2) 155

Not the disabled in themselves, but the Taxi vs Uber (and other similar services) shows what happens once you put regulation over regulation over regulation on a system. It eventually becomes so poor, that unless you have the resources and influence of New York City, the system eventually becomes useless and expensive to the point it may as well stop existing.

Taxis were a luxury people would use to go to the airport if they couldn't find someone to drop them off, or if they were stranded drunk on a populated corner, and avoided at all cost any other time.

The disability acts in many countries have created situations of "if the disabled cannot have it, no one can", like ebooks in schools, and this.

It really sucks when shit happens, but is it really a better idea to make everyone lose out? If its we as a society needs to take better care of the weak, then its society (the government, taxes, municipalities) that should be responsible for doing it...not private entities that are trying to create new ways of doing things (beyond what they pay in taxes). You can give incentives to nudge them, sure...but don't go trying to kill services that are genuinely helping some people live a better life just because they're not helping EVERYONE live a better life.

Contrary to popular belief, your life isn't perfect the moment you're not a black female blind paraplegic.

Comment Re:Where are they? (Score 2, Insightful) 398

Are they turning them out at the same level though? Big universities discriminate like crazy, and will let weaker candidates in their pipelines in computer science if they're female or black much more easily. Some of them will do fine, but a lot will only barely squeeze through, because they were not really qualified in the first place.

Then they'll just fail all the interviews once trying to get jobs.

If leading tech firms hired at a lower rate after adjusting for universities' lower admission criteria for these people, then sure.

Google is known for having a poor hiring process, so I'll give you Google. But most tech companies lately, even the big names, don't really have the luxury of being picky when hiring. If someone is even remotely qualified, even if its a female black trans covered in tattoos going to the interview in ripped jeans and a dirty hoodie with their face covered by a hijab, they'll get hired.

That's IF they are qualified...

Comment Re:There is no skills gap, just pay/training gaps (Score 1) 391

Even if you post a job offer at 500k/year, you won't get a significant uptick in how many qualified people will apply for it (you will see an uptick, but it won't double or anything). What you will see however, is a HUGE amount of idiots applying for it, and your hiring managers will struggle to find signals in all the noise. In the end it may take longer to find qualified people.

How do I know?

Simple: we tried.

Comment Re:US does this too, but badly (Score 1) 337

The issue with people "skipping in line" is that those who do are usually burdens. like those El Salvador illegal immigrants who come to sanctuary cities, and get taken in the public schools. That wouldn't be an issue, except in many of those cities, those public schools are having a hard time with cash, and in my town at least, taxes are already extremely high. It kind of sucks that the local poor who need help are told to share it with people who shouldn't be there. We can increase taxes...but if you're going to take more money from me, I'd like it to go to the people who were already legitimately there and needed it in the first place.

If someone comes in and ponies up 1 million bucks, its a heck of a lot less likely they'll be a financial burden. They may be a cultural problem, they may be fleeding some charges..but that's true of all the immigration paths, legal or not. Not having to support them financially is one less problem to deal with.

Comment Re:Not Employees (Score 1) 88

If Uber's drivers are considered employees, Uber would simply have to get some kind of liability insurance. The cars could still be rented from a 3rd party (isn't that what ZipCar does?).

The biggest overhead would be needing an HR department and having to manage all that crap...but they could contract out for that part.

It would definitely increase their cost, because they couldnt screw over their employees as easily as they could contractors, but it would most likely be possible while staying profitable.

Comment Re:What does Oracle do well? (Score 1) 184

My experience with DB2 was mainly in large finance (Goldman and Morgan Stanley), and while they have very large datasets compared to most of the industry, they don't push these things as nearly as far as others. That said, I mainly saw DB2 scaling well for data warehousing purpose, not raw transnational loads. There's a significant difference between the two. DB2 did really well when you just dumped loads of data on it and then did reporting. But constant transactions like you would in a retail/brick and mortar + e-commerce thing (so constant insert and read of that data with complex queries in real time), not really.

Its not an optimal way of using an RDBMS, but Oracle excels at it.

Comment Re:We're ditching Oracle (Score 1) 184

10000 isn't even that much. My previous employer ran 100~ billion record tables on a 100 million user a year, about 20-30k concurrent at all time, thousand transactions a minute, on freagin MSSQL on 16 core machines.

The only thing that happens is you have to be really careful with your database design, partitioning, etc. My wife works for freagin Amazon, and they don't have that much trouble getting their e-commerce site working. I never asked, but I'd be very surprised if they ran their backend on Oracle =P

Comment Re:What does Oracle do well? (Score 1) 184

Oracle's enterprise stack is pretty strong in certain fields. Their ERP is top notch in brick and mortar retail for example (better than SAP). And once you go down that road, its Oracle stack the whole way.

Beyond that, if you're trying to scale an old school relational database to crazy scales, Oracle leaves everything else in the dust. For horizontal scaling, nothing else comes close.

The thing is, if you're trying to do that, you're very likely doing it wrong. For insane amount of data, other commercial solutions do data warehousing better (Vertica) or cheaper (Redshift), if you want to make a lot of smaller, more efficient datamarts you're better off with postgresql clusters, for big data processing, stuff like Hadoop/Spark are the norm. If you just want service for fire and forget, Aurora is better. If you just want speed, something like Dynamo or Redis may be better bets.

A giant transnational system is pretty niche now. You're almost always better having a smaller one that that pushes data to a datawarehouse, or having a queue or stream collecting data and processing it into a lot of specialized distributed systems on the fly. And if you're too small for that, Postgresql is where its at.

Comment Re:We're ditching Oracle (Score 1) 184

You also do not need 64 cores on each machine to deal with billions of rows.

disclaimer: the company I work for runs a little over a trillion (a trillion as in a thousand billion) records in _ONE_ table, which is also our most used one. That doesn't count the hundreds of other tables.

Yes, the primary source of that data is a commercial database (vertica). Its also not an RDBMS (doing the kind of stuff we do even on an Oracle DB would be absurd). But we do use open postgresql and other open source non-relational dbs for the datamarts, and they'll easily take a few billion records.

Sure, you could dump the trillion records in an RDBMS with "hundreds of gigs of ram". But that is downright retarded for anything but the most specialized scenarios. Even big finance no longer do that shit.

Comment Re:Are we talking about New Orleans or Seattle? (Score 1) 410

Yeah, and IMO its kind of a big deal. In the US, there's probably a bigger cultural difference between someone from San Francisco or NYC and an Australian, than there is between the formers and an Hispanic. Yet if you have a neighborhood full of australian, indians, french, swedish and english and not a single white...they'll still consider it non-diverse.

Seriously @.@

Slashdot Top Deals

He has not acquired a fortune; the fortune has acquired him. -- Bion

Working...