Very inventive, congratulations on your project! I forwarded your website to my wife as she's very much a DYI person and she was impressed. Keep up the good work, would like to see more projects like this.
the youtube placeholder for the stream has some description[1]: "SpaceX Founder, CEO, and Lead Designer Elon Musk will discuss the long-term technical challenges that need to be solved to support the creation of a permanent, self-sustaining human presence on Mars. The technical presentation will focus on potential architectures for sustaining humans on the Red Planet that industry, government and the scientific community can collaborate on in the years ahead."
Good to see another sysadmin input on this thread. Any bits that you would advise us younger (not that much, 30 yrs old) sysadmins about what to expect from the future?
So my biggest bit of advice is to get out of the tech sector as much as possible. One of the great things about ops is that there literally is not an industry that doesn't employ us. I've worked for non-profits, education, farms (you'd be amazed some of the data needs agribusiness has), and now aerospace. I've even gigged for food service (a local brewery paid me in beer...)
The advantage of this is that gets you out of some of the myopia that the tech sector has a tendency to instill, as well as broadening your geographical opportunities beyond the few tech hot spots. It also teaches you the three most important words a sysadmin has to learn to say: "I don't know". A sysadmin who can't say that immediately when it's true is going to have problems (after all ops is about 85% reading documentation), and figuring out how to get data from a barn to a datacenter will teach you to say it quickly.
Another piece of advice: if you haven't already, work in a datacenter, at least once, for at least a year, before you quit in disgust (everybody does). No datacenter sysadmin is happy, but it's a huge resume buff and it teaches you a lot.
After finishing the article I still don't understand what were the reasons to scorn Tux. Maybe OP should have given an example of a good mascot for the readers to get the point.
I'm still quite happy with Tux though, it's a sober mascot that's very simple/minimal, and is easily recognizable.
Also, wipe clean the existing server after the Postmortem. Even if fairly sure that the server can be cleaned it's never safe to re-use a box which was compromised.
No, using a towel could lead to contamination. Use a disposable paper towel. Also, Security Experts (TM) have not proven that Windex helps, but I'd use it just in case.
This appears to be correct based on everyone of my age sharing the same problems, but the article doesn't give any reasons on why it happens. Would be good to get some more information on causes as this is a rather very interesting topic. Any HN readers have good sources?
Profits from labor's vastly increased productivity over the 40 years have been captured almost entirely by employers/owners/shareholders.
Nothing ever trickled down. What power we had to recapture the gains you and your parents worked hard for has been whittled away by years of anti-labor legislation and rhetoric.
Negotiating for increased wage for the average laborer in a non-unicorn market is difficult. Without union representation, it is a solitary worker with little funds versus firms and industries made up thousands of well-moneyed and well-lawyered people who share the same interest in wanting to make sure you never see more money than they absolutely have to give you. Without solidarity, you can be fired and replaced without repercussions for your employer from your fellow workers, all of whom share the interest in seeing you get paid fairly so that one day they might, too.
Globalization has made it easy to avoid pesky things like fair compensation, hours, working conditions, benefits and taxes that we've come to take for granted in the US.
The world economy slowed in 1973. No one really knows why. Some books focus on the effects in the USA, some focus elsewhere. In the West this has been "The Great Stagnation" whereas in several countries in Asia this has been "The Great Bounce Back" as formerly wealthy civilizations, such as China, recovered some of the ground they lost in the 1800s.
Sad to say, I'm not aware of any books that tell the whole international story.
For the USA, there have been an abundance of books and articles showing the decline of the middle class:
Some writers focus on long-term factors that have little to do with politics. Robert Gordon got good reviews for his historical analysis of the specialness of the 2nd Industrial Revolution 1860-1940, and why the current era is not as robust:
1973 was the first oil crisis. Oil prices went up by 400% and there was a huge spike in inflation.(There's a book waiting to be written about the curiously close relationships between the Saudi and the US oil establishments, and the Saudi influence on US foreign policy since the 70s.)
The oil crisis gave the Right a convenient pretext to blame wage increases for inflation.
This became a standard element of economic orthodoxy, and ever since then increases in wages and salaries - which were a driver of post-war prosperity - have been labelled "inflationary', even though inflation is primarily caused by other factors.[1]
By a remarkable and unexpected coincidence, this was the moment when labour share of GDP and middle class prosperity both started to decline. See e.g.
[1] The Bank of England always plays on this. About a year ago there was talk of increasing interest rates to "head off inflationary pressure" just in case wages might have started rising in the future. Meanwhile asset inflation in the form of rising property prices is ignored, and more obvious drivers of increasing costs, like consumer energy prices, are downplayed.
No one knows why 1973 was the beginning of a long era of economic sluggishness. Oil prices crashed in the 1980s and again in 1990s and again after 2008, but none of these crashes restarted the post-war boom. Adjusting for inflation, oil achieved new lows in the 1990s, and again in recent years, but the very low prices that we see are not enough to bring back the years when the economy grew at 4% on a regular basis. Nothing has revived productivity growth to the 3% a year seen for most of the era 1945-1973. If the spike in oil prices in 1973 was enough to start an era of sluggish growth, then why didn't growth restart once the oil prices collapsed? There are many theories about this, and many brilliant people have written long books trying to explain the events of the last 40 years, but no one knows for sure, and there is nothing like a consensus on the issue, which is why I wrote "No one really knows why."
Consider reading this literature review from an MIT professor of 21 different texts that tackle the Great Recession and its causes if you're looking for more info on the subject:
Lo thinks that the Reinhart/Rogoff This Time is Different is among the best-researched texts with the most quantitative support and analysis while still maintaining its readability.
Seems obvious - economic turmoil and globalization has ravaged the job market, educational requirements have crept upwards, educational costs have climbed too, and housing prices went skyward.
Where a highschool-educated man once could get s factory job and buy a house on that salary, a modern young person gets a part-time job with an expensive degree and uses it to pay off a pile of student debt and while a house is completely out of reach.
Different countries have different problems which the article barely touches though the graph hints at.
The UK has a skills shortage, it's vacancy to applicants ratio is back to pre-recession levels despite huge levels of immigration from the EU. London is the 6th largest French city in terms of population, tens of thousands of millennials from Southern and Eastern Europe have migrated to the UK. This is brilliant because they are often well educated (by someone else) and motivated - we have several recently arrived European devs in our office.
However at the same time it takes people longer to get into jobs. Since the 90s the concept of a degree being a pre-requisite for professional jobs has been pushed and pushed. I remember at my grammar school it being assumed you were going to university. Combine this with the rise of the "gap year" and people are emerging into the job market years later than they used to, with different expectations about the kind of work they will do. The article (it is the Guardian after all) highlighted a charity sector employee - not a sector you traditionally associate with pay rises keeping up with the commercial sector. I know far too many acquaintances who emerged from university unwilling to work for a profit making enterprise.
Many European countries, especially in the south, have completely different issues - huge problems with youth unemployment which is driving emigration. A labour market which has two tiers of employees due to the difficulty of laying off permanent employees causing young people to disproportionately end up on casual contracts.
Someone much better placed will no doubt be able to explain the issues in the US and Canada, indeed I'd be unsurprised if there weren't different issues within the US.
It's clearly for the job market overall, there will be variation within sectors. Given over 2 million private sector jobs have been created over the past few years while the public sector has shed jobs I doubt workfare is involved.
I don't believe there's a skill shortage in the UK. Maybe for industries like banking and programming, but in general, it's really hard for a skilled (STEM, except programmer) graduates to get jobs paying more than 30k (in London).
Not even programming. CS has one of the worst employment rates of all degrees.
Admittedly a lot of arts grads don't get jobs as artists, where most CS grads are only really interested in a programming job.
But even so - the skills shortage is somewhat mythical.
As you say - the remuneration shortage is more obvious than the skill shortage.
Housing costs in London are in the same ballpark as SF, but £40k is considered a generous salary for permanent developers. £60k might be possible for seniors.
It isn't unusual to find web dev jobs offering <£20k.
Work in the City offers much better pay, but the culture can be pretty horrific for anyone who isn't suited to it.
CS is a crazy field. We pay insane salaries to people in their 20s (150K in the Bay area sees normal) but people in their 40s have a hard time finding any paid work. I'm trying to understand why this happens have a hypothesis:
Tech is a winner-take-all industry. You get nothing for being second place. This requires long-working hours (it needed to ship yesterday!) and people you can exploit (younger folks who don't know any better). Moreover, the ever changing nature of languages and tools makes it that on the surface, a young person in their mid to late 20s seems as much experience as an older person. When you combine this, you get shitty software that has shipped fast.
Can we do anything about it? Unionizing software or having some kind of guild membership is not going to happen in my lifetime. (we're all too libertarian). We can slow down the churn of tools (this is also hard and requires discipline ... if we all understood the damage caused by the new new thing, maybe we can do it but I doubt it - can you make a promise to NOT learn any new languages this year? I don't know if I can). We can do our own startups (this requires access to funds that not everyone has, and perhaps a particular lifestyle situation; also ... odds of me getting into something like YC are near 0 I feel). Or, we can change fields. As someone who loves tech and not much else, I have severe doubts I'll be able to find something I have as much passion for as tech. But hey ... who knows. Life is a new adventure everyday :)
The baby boomers and the hippie generation have sold out technology and thus jobs to overseas countries while holding and controlling real estate themselves.
[Disclaimer: somewhat of a hippie myself, but there's a lot of hypocrisy in that area.]
Deficit spending by governments creates a generational obligation wherein the future generations must pay for profligate spending in the past (which they cannot go back in time and vote against). The primary means of paying down this debt is through currency devaluation; note how the person in the article claims her pay raises have at best tracked her cost of living increases. This transfer of wealth across generations also hurts the poor the hardest; if inflation weren't so bad for the poor we wouldn't have to raise the minimum wage periodically.
I suggest reading "the great wave: price revolutions and the rhythm of history". I found it to be a surprisingly easy read.
> Update 2: Earlier this afternoon, Chris Vickery confirmed that the three IP addresses that were disclosing user information have been secured. The issue wasn't a hack, but a misconfigured MongoDB installation.
> The source of the configuration error isn't clear, as neither the ISP nor Sanrio has answered questions on the matter.
A MongoDB database open to the outside world on a public IP address?
> At the moment, there are at least 35,000 publicly available, unauthenticated instances of MongoDB running on the Internet. [...] all of the exposed databases combined account for 684.8 TB of data.
These things happen to the best: In 2001, the World Economic Forum had their MS-SQLServer with personal data of the participants connected to the Internet. With the standard account enabled (user: "sa", password: empty). It then got "hacked" and the data leaked.
Why is everybody immediately making these types of diversion responses regarding relational datastores to mongodb breaches? Mongo has popularity and 2 fixable issues that make the exposure numbers outweigh those of every other datastore right now.
1. There is no password prompt during installation. This was the same problem many relational databases had back in the day with default user credentials. Today, I don't know of a single relational datastore that comes with default credentials during installation. (ignoring things that shouldn't go to production like homebrew) Individually, they all said it was a "security threat" and did something about it.
2. Things like MMS or other offering from MongoDB itself expect you expose your nodes externally so they can manage things for you. This can easily be made secure... but its contradictory in nature. It suggests to somebody who probably doesn't know what they're doing on the IT or DevOps fronts to do something more easily dangerous than not... just so they can sell their vendor services to you. Its a recipe for disaster but its also easily fixable with a few design changes on their side.
I guess you could read my post as "others aren't better than MongoDB", but that wasn't my intention. I merely remembered the WEF anecdote and thought I'd mention it here (although it's unfortunate I didn't find an English source for it).
MongoDB is repeating the same mistakes others made 15 years ago.
Apologize, wasn't really directed at you. I'm sure you've seen all the other comments related to this type of news.
Sounds like we both agree that MongoDB is repeating the same mistakes others had made 15 years ago. The only difference is, back then, there was a less dramatic excuse for how it got that way.
This is only true after the Mongo people were lambasted on the web for having such a terribly insecure out of the box product. Which was fairly recently. For years the product would bind to all IP addresses. Which is insane for a default install.
Feb 2015:
Discovered 40,000 vulnerable MongoDB databases on the Internet
Just spent some time sending my scripts to this site for it to analyse and see what it does.
I can see that while it wasn't be able to tell me of more efficient code to achieve my goal (wasn't really hopping for that), it did spot 1) one liners where some commands are not needed, 2) variables which are not used, 3) where I should use double quotes to prevent word splitting and 4) lines where my ssh was eating up my stdin.
What a great sanity check for the days when I'm writing something on my own without a second pair of eyes to proof-read it.
Plus, it seems like puppet is working on a plugin to make use of OneGet. It'll make work much more interesting in the future that I'm actually looking forwarding this happening.