OP lives in Rome, where the average driver mentality is somewhere between "need for speed" and "carmageddon".
Basically no one is ever willing to wait for anyone else, as OP's reasoning shows pretty adamantly.
Of course they're grossly incorrect.
HN user
OP lives in Rome, where the average driver mentality is somewhere between "need for speed" and "carmageddon".
Basically no one is ever willing to wait for anyone else, as OP's reasoning shows pretty adamantly.
Of course they're grossly incorrect.
Acting like what makes Rome's traffic dangerous are cyclists or moped drivers (even if those can often be quite reckless) is basically victim-blaming.
We should never take such posts too seriously.
This is indeed correct. Wikimedia overall uses less than 2000 bare-metal servers, so yes the infrastructure is tiny compared to those.
What can be interesting, I think, is that you have a completely open infrastructure that has to solve problems on a global traffic scale.
If people are interested in knowing more, I suggest you also take a peek at the wikimedia techblog, specifically to the SRE category https://techblog.wikimedia.org/category/site-reliability-eng... and the performance one https://techblog.wikimedia.org/category/performance/
It's both a waste of donor money and a starvation of resources for people actually consulting images on wikimedia commons.
Any for-profit entity hotlinking Commons is unfair. Heck, they have the right to redistribute freely the image as they see fit, instead of consuming resources that are a common good.
But this goes beyond that - it's some blind check of internet connectivity for the app, and doesn't get shown to the user. We're pretty sure of that, given that with the amount of noise that task generated, if there was an app featuring that image at least one of the ~ 90M daily "views" would've been someone reading these posts.
Now, given we want to be nice, we didn't just blindly block the traffic, although making requests without user-agent is against our UA policy https://meta.wikimedia.org/wiki/User-Agent_policy
Purges have been migrated to kafka as a mean of transport, at long last. So now if a purging daemon crashes, purge requests are not lost.
You can see per-server stats on purges happening here:
The real throwback would be going back to use newsgroups, and only have competent people posting.
The hard upper limit on a neutron star mass, the Rhoades-Ruffini limit, is at 3.2 solar masses.
This discovery is significant, because we never found a collapsed object in the 2-3 solar masses range, but it definitely doesn't need new physics to be explained.
The article is full of inaccuracies and errors. Confusing threads with eventloops is just one of them.
I can't imagine it's being upvoted.
You remember incorrectly. It was a plan from a past Executive Director that never even started.
Changing of ownership fundamentally resets the trust we all had in PIA, which was due to you having proven in court you deliver on what you declare.
And in the VPN world, trust is fundamental.
I am still surprised you didn't see this coming.
I did unsubscribe a couple weeks ago and got a canned response too, almost immediately.
When your business is privacy and trust, any acquisition is problematic. Let alone an acquisition from a company with a similar track record.
That's only partially true.
A big part of security is your relationship with the people providing you hosting.
The compromises to nordvpn and others came from a rogue hosting company, for instance.
When you people will be able to scale git to accept the number of revisions Wikipedia has, and add several new ones per second with millisecond latencies, please let me know.
Also, only a programmer could think git is a friendlier interface than a wiki.
There are things that are wrong in the wiki comunity, none of which would be solved by using github instead.
Not to point out that the forking model would lead to fragmentation and GitHub is actually a privately owned for-profit platform, of course.
Physics gave up being "deterministic" (in the sense you used the term in) about 120 years ago.
And mathematics is just a convenient tool with which we describe reality, no one would seriously think today that our math is deeply inscribed in the essence of reality as Galileo thought.
Apart from this, and the fact that the paper shows a very weak statistical significance, I agree fully with your last statement.
What a whole bunch of inaccuracies. I have no time to waste rebutting all the multiple falsehoods in this post, but I want people living in other countries know that it is filled with inaccuracies and plain falsehoods.
booking.com and hotels.com 50/50?
LOL.
And yes, failing to kill booking.com when Expedia was a clear market leader is a huge mistake indeed.
Techops at WMF here: the TL;DR is "no", but I can get into details.
MediaWiki accounts for the lion's share of the servers we run (between MW itself and the databases and the object storage for files/images, I think around 60%), but it's not really that inefficient, also because we aggressively cache content on our own CDN.
Given the amount of traffic we handle, I don't think any sgnificant change could reduce the hosting burden much - and it's already a very small footprint at 2M/year. To give you some context, we have currently less than 2000 physical nodes across all of our datacenters; it's pretty hard to reduce that number further while guaranteeing stability and performance of the sites.
I'm not saying we couldn't improve things - there is always room for it - but I doubt we could do something that reduces our hardware footprint by a significant amount. As a benchmark, moving from PHP 5.x to HHVM gave us more capacity and over two years we reduced by 15% the number of MediaWiki application servers, but that does not result in a huge change in the hosting bill.
If McDonald's provided cheap food by ignoring the labor laws, would that be fair? Or any safety regulation on how food is preserved, etc. I could go on, but I think you get my point.
I share your sentiment towards our taxi drivers, but I'd rather ask our laws to change to be fairer than to advocate Uber's right to just break the law.
The solution to that is to allow more licences, and maybe to use those new licences to compensate current drivers for the loss of value of the licence they bought.
Allowing a corporation to just ignore existing regulations while everyone else is respecting them is a completely different story.
Let me add: I have very little sympathy for the taxi drivers guild and their practices, but I don't think that's a good reason to allow Uber to operate outside of the law.
Selling at a loss when you have a lot of money will give you an unfair advantage over others that can't sell at a loss.
You do that so that you can take a stranglehold on the market and then dictate prices yourselves.
It's the texbook definition of dumping:
https://en.wikipedia.org/wiki/Dumping_(pricing_policy)
and protecting against that practice is NOT the texbook definition of fascism at all, quite the opposite; also keep in mind there are no taxi corporations here, as the taxi license must be owned by the individual driving it (with some small and irrelevant exception).
That being said, I wouldn't have a problem with Uber selling its services as long as they abide by the law, they and their drivers pay the taxes and they don't engage in anti-competitive practices.
Please beware Italy is not a common law country so legal precedents are not binding, and you do need a new law here to future regulate something.
heh, yes, just wrote this too early in the morning. Editing, thanks!
Yes. Also, license costs in Italy sunk in the last 15 years anyways, as more of them were granted. Add to that that the license is individual, and you can surely sell it, but just to another person; corporations can't own them and ask for a daily ransom from taxi drivers.
Being from Italy, let me point out a few things:
Frist of all: unfair competition can mean, for instance, a mega-multinational subsiding its riders costs to wipe out competition. That is dumping, a practice that is illegal in most advanced economies, and that Uber has been known to practice.
If you don't have a taxi license, Italian law restricts the ways you can offer transportation. If you step outside of those boundaries, you are creating an unfair competition to taxi drivers that have e.g. to pay for the license or non-taxi services that operate within the boundaries of the law.
Also: Uber was probably aware of the situation, to the point that some MPs tried to sneak in a "save Uber" amendment to an otherwise unrelated law, causing protests through the country from taxi drivers.
I see people commenting that have clearly no information or any idea of how Italian economy works, and just go around throwing approximative judgements based on hearsay and prejudice.
So let me state it out for you: Italian economy is surely burdened by anti-competitive practices and too many absurd regulations, but blocking a US multinational to repeatedly break the law and finance its services by price fixing is not an instance of that.
And finally: in Italy it's impossible for a company to own taxi licences, so they are all privately-owned by the taxi driver. So Uber is, if you remove the price jacking, practically a way to siphon money from individual drivers to a large company.
The total number of requests that get through the cache to the application layer can be seen here
https://ganglia.wikimedia.org/latest/stacked.php?m=ap_rps&c=...
which you can see is not showing any substantial increase due to the passing of Prince. The big hole of two days that ends just before the news broke is due to wikimedia switching traffic to a second datacenter for two days, see http://blog.wikimedia.org/2016/04/18/wikimedia-server-switch...
This is not just a joke; the interface is working (apart from some bugs you might expect at this stage) and while the announcement is clearly a joke, the interface is IMHO pretty cool
Wikimedia's LVS manager that does what seesaw does, although it was never "advertised" to the public:
https://github.com/wikimedia/PyBal
It actually is a bit different in that it advertises VIPs via BGP, so no floating IPs are needed, but you need to be able to configure your upstream routers, and its configuration is much more flexible.
Actually, it's a bit more complicated than that. We don't just serve .wikipedia.org from our frontends, but also e.g. .wikisource.org etc.
So yes, we are using SNI but for wildcart certs :)
Candidates naturally feel uneasy when interviewing. In the end, one hour of chatter will either get you a job you kind of desire, or not.
I think it's our responsibility, as interviewers, to make people as comfortable as possible so that they can perform at their best. It's our responsibility to prepare for the interview, smile to the candidate, downplay any hiccup they may have.
I don't believe all the BS about "taking people out of their comfort zone": 98% of people will feel uneasy at an interview and they will be well out of their comfort zone already.