If only people would listen to Ryan North this problem would already be solved.
Let's drink that milkshake!
If you think that any generation has the technical skill required to effectively troubleshoot their own devices then you need to broaden the range of people you hang out with. There are very intelligent people out there that don't understand anything about computers but need to get things done on them. The answer isn't for them to take time away from their expertise so they can learn how to compile a kernel and update graphics card drivers. The answer is to make an appliance for them so they can get their work done WITHOUT HAVING TO constantly troubleshoot the device.
I am amazed that anyone who has spent any amount of time with non-geeks could think that people on average know how to do any serious technical troubleshooting.
Many of the things you say here are true and they bug me as well. But if you read the article then you realize that many things that the author is advocating are FAR more trivial than just having a large size limit or allowing arbitrary character input. The author states that no system in the world does things right and implies that programmers should start dealing with all of those situations.
The GP used hyperbole to explain the edge case. So we don't really know if that includes "my name is more than 8 characters" or "my name used to be Prince but now it is this symbol can only be stored as an image."
The key here is context. If an edge case only affects 0.0000001% of your users it's probably not worth worrying about. In all of my professional experience many of the items on the authors list have fallen into that category every time. Any sane person will do at least some sort of cursory cost benefit analysis of each constraint rather than spending a whole year writing code to handle every possible scenario when it buys you almost nothing.
It just so happens that most people are pretty sane because the author admits that no one in the world is doing what he seems to be advocating. This is not because they are stupid or have OCD. It's because doing the opposite would be a stupid, OCD, perfectionist sort of thing with vastly diminishing returns.
Citation? Seriously I would like to know one way or another. I've heard both stories. Never with a citation.
Wish I could mod this up to 6. This is not that hard to understand. Assume my options are Linux, Mac, Windows.
Linux: It's just not that easy to get everything that you want to use working. Just cause you are a geek doesn't mean that this is your setup: http://richard.stallman.usesthis.com/
Windows: cmd.exe anyone?
Mac: bash, MacPorts to install all the OSS stuff, MS Office since I don't think that responding to client emails by asking them to send it in a "non-secret" format would go over very well (http://www.gnu.org/philosophy/no-word-attachments.html). Also Flash, as much as I hate it it's what you need to watch internet video right now. From a practical standpoint it really is the best of both worlds, and the software options are already great and getting better every day.
You don't need to be open minded or even that smart to see why this is an appealing option.
If he is telling the truth then they are indeed demanding his root password. It is not clear from the original post, but if you read his other posts in this this thread he indicates that are refusing to give him back access to the box or his data until he provides them with the ORIGINAL root password.
You are right. I come from a background of always co-locating and hosting my stuff myself. I assumed he was in a similar situations and was absolutely shocked by the situation. After reading more of the thread it's obvious that he was renting the server from someone else.
Still though their response seems pretty ridiculous. He does indicate (elsewhere in the thread) that is was a dedicated, non-managed service. The strongest response that I can see being appropriate here would be to tell him that they can't guarantee uptime without access to information that they need (and still I don't know why that can't ask for logs instead of demanding root) or to tell him that he has x days to get his data off their server before they close his account.
I am asking this as a serious question not a rhetorical one: What good does it do anyone to respond as they did, instead of just doing what I have suggested?
I can understand what you are saying here but if you are renting a whole server, and not just sharing one with other customers, then shouldn't your provider limit their support to what you ask for? If you say you have a problem, and you aren't willing to give them the info then shouldn't they say, "sorry we cant' fix your problem without more info" before you do a hard shutdown on their box and start snooping around? I guess this could create some SLA issues, but it should be spelled out in the SLA that if they don't give you access then they can't guarantee uptime.
Under these circumstances your company would seriously break in and start snooping around?
In my mind this is analogous to calling your landlord with a leaky faucet but then not letting him in the house when he gets there. Your landlord keeps a key to the place, but he can't go in without your permission. If you want the faucet fixed and you don't give him permission to come in there you are out of luck on the faucet, but that doesn't give him the right to sneak in when you aren't home, lock you out of your own house, and go to town on your plumbing.
Also if this guy is telling the truth he isn't paying for any sort of management at all. (See his response to parent)
You people are both crazy. I can not even imagine any provider I have ever used even thinking that it would be appropriate to ask me for my root password, much less actually force themselves into my box after I had explicitly denied them access. Even if you don't care about the inherent security problem and the blatant illegality of it, is it not a problem for you if your hosting provider forcefully powers down your server at any time!!!???
> Yes, it's kind of lame that they are rooting boxes.
Once again, you are crazy. "kind of lame" doesn't even begin to describe how inappropriate that is even for just the downtime that it would cause alone.
> On the other hand, the questioner might be more problems than he is worth from their point of view.
Then why are they doing business with him at all? An appropriate solutions in this situation might be to say, "sorry we can't help you with this issue without access to your logs." Forcefully breaking in is way more than lame.
> If I were in the same situation, I'd just change providers and find one who will put into writing that they won't root my box (good luck with that).
Of just find any provider that isn't straight up terrible. There are so many options out there right now that it boggles my mind that any provider could get away with this. If my provider pulled a stunt like that and it got out, there would be a mass exodus of servers going out the doors.
Seriously, who are these clowns?
1. Don't EVER host with them again. I don't know what's in your contract but as far as I understand it, breaking into your server without your permission is illegal. It's possible that you could take legal action against them.
2. Figure out how they broke in. If they broke in then someone else likely could too.
I have never heard of anything like that happening with any host ever. I am amazed that a company could act like that and still expect to have any customers. It's not like there aren't options.
That is a great comparison, and contrary to some of the responses, being able to do alter table statements on an in use production system is vital to any serious database solution. It doesn't say anything about Oracle vs. Postgres though as Postgres has been able to do this for a very long time.
I'm not just trying to be contrary here, I would really like to know. What does Oracle have that puts it (20 years?) ahead of Postgres (other than RAC, there were very informative posts above about that).
Just ask yourself one question, if you don't trust your internet provider enough to do DNS correctly, should you trust them at all?
Do you mean, trust as in trust to not do shady things like violate my privacy or trust as in trust them to be competent when setting up their DNS servers?
I don't really trust any ISP to not be shady. But since I want to have the internet I haven't got much choice. I don't know if comcast or our local municipal fiber provider is better in this regard but I know which one is a lot faster.
I also know that sometimes the line just goes down for like 10 minutes with no explanation. And sometimes their DNS servers crap out for like 10 minutes. I can't do much about the line itself going down. But by switching to google DNS I can avoid issues with my ISPs DNS not working.
For me it's pretty much that simple.
The problem is for every expert there is an equal and opposite expert. In this case there are companies (not to mention the economy as a whole) that would lose enormous amounts of money if we start to crack down on carbon emissions. There are also people who are set up to profit both politically and financially from "proving" the science behind climate change.
The other problem is that the public isn't being asked to believe in this stuff because believing in it will magically fix the problem, the public is being asked to believe in it because the government is asking for our money. They are asking for us to pay higher prices for EVERYTHING because everything we do requires energy. The cheapest way to get energy right now is to burn fossil fuels. On the other hand if New York is indeed going to be buried in water in 50 years then it's going to be cheaper and better all around to do something about it now while we still can. It is very important to get this right and saying "trust me I'm an expert" just isn't good enough.
The proposed solution to this problem is of course to declare that there is a consensus among all scientists who matter. But of course as a lay person the process of deciding who gets to decide which scientists matter and which one's don't is no easier than deciding which experts are correct in the first place.
Establishing the truth about anything is simply a difficult problem and pretending that because someone is an "expert" that they can't be wrong is just stupid. Certain things are just very complicated. Fields that study complex systems evolve more rapidly and have less stable elements than those that study simpler more easily observable phenomena. For instance, if you told me that you were an expert on Newtonian physics, and you told me that you could launch a canon ball and tell me where it was going to land I would believe you. If you were an expert in quantum physics and you told me you knew exactly how, given the proper equipment, to produce a Higgs Boson, I would assume that you understood what the theories said but that in your excitement you might be a little overly confident since no one has ever done such a thing and the theories could be wrong. If you were an expert in psychology and you told me you could tell me what I was going to eat for breakfast in the morning I would call you a quack.
If you were a doctor telling me that the surgery you were recommending had an 98% chance of killing me I would get a second opinion. Even if I thought you were the best most "expert" doctor in the world and you told me that the alternative to the surgery was certain death, I would get a second opinion. And probably a 3rd, and a 4th and a 5th opinion as well. Why? Because the cost of getting the opinions of 4 doctors is way, way more acceptable than the possibility of dying on the operating table and there is some chance that my doctor made a mistake.
The impression that I get after looking into the situation with these leaked emails, and please correct me if I'm wrong because I would very much like to wrong, but the impression that I get is that no one, outside of the research teams that developed these models has ever seen this code or been able to duplicate it themselves. I don't know what happens in the peer review process but it appears that actually looking at the code and validating that it works properly is not part of it. And even if it is, why not let other scientists look at it? Why not let other "experts" look at it? I feel like you are telling me that once the first doctor has done a diagnosis that he has the right to hide all of the data related to it, not let anyone, except a self-chosen peer, look at it, and that I should believe that anyone who comes up with a different diagnosis could only be wrong and I shouldn't listen to anything they say.
Research gets published in journals for everyone to see, etc. It's not like we're keeping it a big secret
But the impression people are getting from this incident is that only part of the research gets published and that some parts are hidden and that scientists are even committing illegal acts such as deleting information that could be requested as part of a Freedom of Information Act in order to keep things hidden. When I read accounts of other climate scientists trying to duplicate the results and being unable to and then requesting information on how it was done in the first place and that request for information being denied what am I supposed to think? That the man working in secret is right because the UN published his results and that the man who appears to be working out in the open is a villain? When I read that after their models are revealed they produce the same charts regardless of whether they are fed real data or noise am I just supposed to assume that they were correct because the were created by experts?
Your right, as a lay person it is hard to understand the arguments of experts. But in a situation where determining who is correct is vital to our future prosperity there is a larger burden than just doing good science. In situations where the only thing that matters is the opinions of other scientists I would say that you are right to not waste your time arguing with idiots. But when science influences policy there is an additional responsibility to make a convincing case that those idiots can understand. Is it a lot more work. Yes. Am I asking you personally to sit down and tutor every one of them? Obviously not. If you go to http://copenhagendiagnosis.org/ you will see the following:
The report has been purposefully written with a target readership of policy-makers, stakeholders, the media and the broader public.
This is because these people are not idiots. They know that they are asking for major changes to be made with regard to how our economy functions and that the burden for those changes will be paid for by "the broader public". Hence they are willing to engage with said public. And that report in fact, I believe does a very good job of that. More is required however. The report makes sense. But if the codes and raw data sets used to come to these conclusions are secret and cannot be duplicated by others how am I to have confidence in them? If skeptics, even lay skeptics, raise concerns they should be addressed and communicated to the public in as clear a manner as possible. The burden of proof is in fact extraordinarily high. The following must be determined with a very high degree of certainty before we can be confident that we are making the right policy choices:
1. The earth is getting hotter
2. The warming is a long term trend
3. The warming is caused by humans
4. The warming is caused by human produced CO2
5. How much warming is going to occur?
6. What is the effect of the warming?
7. What is the best way to deal with the effects of the warming?
8. What is the best way to stop the warming?
The answer to all of these questions involves the analysis of very, very complex systems. If you are trying to say that a doctorate level degree equals expert equals trust everything they say without question, then you are in fact the idiot. My experience with medical doctors is that they are wrong more times than they are right when it comes to anything that falls outside of the normal situations that they regularly run into. Do I assume that there is a conspiracy amongst doctors to harm my body. No, that would be just as stupid. They are wrong because the human body is very complicated and despite the fact that we have been studying it for a long time there is still much that we don't understand. To assume that someone with a Ph.D in a much younger science such as climate science, studying a much more complicated system, the earth, of which we only have one specimen rather than billions, is automatically right to the degree that we are willing to institute programs that cost billions of dollars, is just crazy talk. I'm not saying that the science is not good only that much skepticism should be applied. Essentially:
1. The burden of proof is high. Doing a couple of peer reviews per study is not enough.
2. The entire process should all be done out in the open where all "experts" and consequently lay people have equal access to all data and code so that every expert can make his/her case.
If I go to you and demand your time to explain to me your job because I am curious than, yeah, I am the arrogant one. But if you take the results of your work and come to me and demand that I change my lifestyle and spend my money in a certain way and that I shouldn't question you or ask for you to explain yourself or show other scientists how you arrived at your conclusions then your arrogance is just stupefying.
Well I don't think anyone is suggesting that we set it up on github so every clown coding in his mothers basement can can start contributing. I don't know that the important thing here is a true "free software(tm)" or "opensource(tm)" license. The important thing is that before we start looking at this research and assuming it is all correct because a few other scientists did a peer review and then making sweeping and expensive policy changes at the highest levels we should open up what they did so that people can look for problems in their methodology.
Now I don't think that anyone will care what I think of their code but I'm guessing that there is more than one person out there with a Ph.D in climate change that could look at this stuff, if it was public, and either confirm that the work is valid or point out it's flaws. At least there could be a debate about it among scientists. It is understandable that they are worried that powerful lobbies will try to distort their work and lie about it. But there is no other option. This is science that is affecting public policy and it can not be done in the dark.
On the other hand given how poorly some of this stuff appears to be coded it seems that they could use all the coding help that they could get: http://di2.nu/200911/23a.htm. Hopefully these assessments of how sloppy their work is are not accurate, and that most of the work that has gone into the IPCC reports is less error prone than the stuff that has been leaked.