Live data from Hacker News

Au Revoir

blog.docker.com

251–260 of 305 posts

Re: Au Revoir

#251
post #231
post #222

Kinda funny that the page is down ... you'd think for a company that helps everyone scale their services this would not be a problem :-)

I'm not sure resource scaling is what's needed here. Just good ol' caching. Shameless plug: I built Cachoid[0] for caching as a service (leveraging Varnish mainly) to help with situations like this. [0] https://www.cachoid.com/

I understand that this is a shameless plug but I've met teams where the first response to any performance problem is caching

If you're reading this and have performance issues look at caches as a last resort, first figure out why things are slow

I'm assuming a traditional stack here but use mysql jet profiler, or SQL equivalent first, then your languages profiler xhprof?

If your queries are slow or impossible make sure your tables are normalised and indexed, make sure opcache is on and xdebug off

Mysql out of the box is very conservative with its resources, the limits can often be raised and the web server can often be configured to be faster, make sure you're sending the correct http cache headers

There's so many things to consider with performance Please don't denormalise your data slap it in redis/ use varnish and call it a day

No cache is free, eventually you have to pay the pied piper when you do make sure you can afford to have never used the cache at all

Some of my hardest bugs have come from out of sync caches & trying to build performant data architecture around systems that had data stuffed into a redis black hole aged for years..

Re: Au Revoir

#252
post #251
post #231

Earlier quoted context omitted.

I'm not sure resource scaling is what's needed here. Just good ol' caching. Shameless plug: I built Cachoid[0] for caching as a service (leveraging Varnish mainly) to help with situations like this. [0] https://www.cachoid.com/

I understand that this is a shameless plug but I've met teams where the first response to any performance problem is caching If you're reading this and have performance issues look at caches as a last resort, first figure out why things are slow I'm assuming a traditional stack here but use mysql jet profiler, or SQL equivalent first, then your languages profiler xhprof? If your queries are slow or impossible make su…

Also if you're doing key scans in redis, you're doing it wrong, it's possible to work around with indexes you have to manually maintain, but it's not worth it

Re: Au Revoir

#253
post #41

Earlier quoted context omitted.

There is a reason why most startups don't go big - burning vision is not even close to being the right formula for success. Running a company of 10 people is VERY different from running a company of 100, which is VERY VERY different from running a company of 1000. Just like precious few individual contributors are comfortable with and able to contribute meaningfully in this wide range of environments, it doesn't make…

> Running a company of 10 people is VERY different from running a company of 100, which is VERY VERY different from running a company of 1000 IMO thats a classist lie that keeps people in their place. Any reasonably smart CEO can learn his way from 10 - 10000. Yes, (s)he may not be capable of 10k on day one, but they can grow along the way.

To be really good at something it is necessary (but not sufficient) to be interested in it.

Running a company of 10 people is an experientially different thing than running a company of 1000, just like collaborating is an experientially different thing from doing stuff on your own and playing video games is different from using Excel. Being interested in one is no guarantee of being interested in another.

Re: Au Revoir

#254
post #41

Earlier quoted context omitted.

There is a reason why most startups don't go big - burning vision is not even close to being the right formula for success. Running a company of 10 people is VERY different from running a company of 100, which is VERY VERY different from running a company of 1000. Just like precious few individual contributors are comfortable with and able to contribute meaningfully in this wide range of environments, it doesn't make…

> Running a company of 10 people is VERY different from running a company of 100, which is VERY VERY different from running a company of 1000 IMO thats a classist lie that keeps people in their place. Any reasonably smart CEO can learn his way from 10 - 10000. Yes, (s)he may not be capable of 10k on day one, but they can grow along the way.

A classist lie? What are you talking about?

Statistics are pretty clear in the failure rate of founder-CEOs when they scale past certain sizes. Even Zuckerberg -- if it weren't for Sheryl Sandberg, Facebook wouldn't have survived as well as it did. Zuckerberg didn't know the first thing about running a large organization, I might argue that he still doesn't -- but when a company can basically print money, there's a huge margin of error afforded to Zuckerberg-type leaders. If Zuck where competing in the energy business, banking or autos, he'd have been long since buried.

Steve Jobs also wasn't particularly good (managing a large org) during his first stint at Apple -- it took a lot of maturity and learning at Next and Pixar before he was really ready to lead Apple.

Mark Benioff is another founder-CEO that did really well -- however, he was an Oracle Vice President before starting Salesforce. He was both incredibly smart but also had experience within a large organization.

> Any reasonably smart CEO can learn his way from 10 - 10000. Yes, (s)he may not be capable of 10k on day one, but they can grow along the way.

Citation needed.

The list of successful founder-CEOs who have scaled to huge orgs is a very short list.

Re: Au Revoir

#255

Earlier quoted context omitted.

No way is it a lie. The kind of people who enjoy running a company with a ragtag group of 10 employees in the early days are rarely the kind of people who will be able to effectively run a 10000 employee operation. If you’re growing fast there is no time for on-the-job training as CEO. You need to be effective immediately or else you are only hurting your company’s growth.

Oh my God, the company's growth? Then we MUST axe the CEO. He's just a human, after all. Don't want humans to hurt company's growth. Company's growth. Growth.

Yeah, the problem is those humans have nowhere to grow. Would you be happy with the same salary, job title, etc for a decade?

Re: Au Revoir

#256
post #172

Earlier quoted context omitted.

> Any reasonably smart CEO Sorry but this is extremely naive and I'm sorry that your naïveté has caused you conclude things which are simply not true. This is a very strong assumption, and a lie that is peddled by the media (Zuck is the greatest example of this) to create a novel story. I'm not saying it's impossible, it's just a dangerous assumption. For the record, I've interviewed over 75 software founders (CEO/CT…

Based on what? All of you here saying "they don't have what it takes" have yet to elicit one actual skill that a founder could not _learn_ as the company grows. There is nothing inherently special about someone who runs a small vs. big company, except for experience. I agree with the poster above: classism at its finest.

Nobody is suggesting that a large company CEO is "special." It's a different set of skills, that's just a fact. And, the idea of a CEO of a rapidly scaling company to learn through on the job training isn't always a good one. People learn by making mistakes and the marketplace isn't very forgiving -- too many mistakes and the company is toast.

There's nothing classist about it. You could train an army corporal to become a general. But you can't just take an army corporal all the way to the experience needed to become a general in a year or two. It takes generals 20+ years before they have the skills to lead thousands of soldiers. To suggest that any reasonably smart corporal can go from leading a squad to leading a division in a few years of on-the-job training is just ludicrous. If I'm wrong, then please provide some data. Actual data, not anecdotes.

Nobody is saying that a corporal can't ever become a general. The suggestion of classism is implying that a corporal can't become a general because he's not of the right class. Of course a corporal can become a general (at least in the US Army) -- which is the very opposite of classism. We're just saying that a corporal doesn't become a general through a few years of on the job training and that a corporal doesn't become a general while on the battlefield.

What I'm saying is that a good large-company CEO is gets those skills over years of experience. If a founder-CEO slowly scales, then perhaps on-the-job training would work. But when an org goes from 10 people to 1000 people in a few years -- it would take an extraordinarily lucky person to be able to handle that transition without prior experience.

Re: Au Revoir

#257
post #251
post #231

Earlier quoted context omitted.

I'm not sure resource scaling is what's needed here. Just good ol' caching. Shameless plug: I built Cachoid[0] for caching as a service (leveraging Varnish mainly) to help with situations like this. [0] https://www.cachoid.com/

I understand that this is a shameless plug but I've met teams where the first response to any performance problem is caching If you're reading this and have performance issues look at caches as a last resort, first figure out why things are slow I'm assuming a traditional stack here but use mysql jet profiler, or SQL equivalent first, then your languages profiler xhprof? If your queries are slow or impossible make su…

Lots of good common sense advice and practical experience there. I've had my fair share as a devops/sysadmin working on crazy systems (more worse than better haha). So it's unsurprising how common sense isn't always put into practice.

I do believe caching to be the proverbial Swiss Army Knife of performance. Some of its tools can be used properly and others as band-aid / stop gap. Even band-aids serve a useful purpose. But I can definitely see how a system can accrue technical debt when over leveraging caching.

But to believe in prevalent, perfectly built systems is to deny the status quo of the landscape. When you democratize development like WordPress has done with its plugins and Node has done with npm, as popular examples, you definitely have it coming.

Re: Au Revoir

#258
post #131

Earlier quoted context omitted.

What on earth makes it a “classist lie”?

The implication is that the (especially American) upper class is trenched on this belief that CEOs of large companies require a certain pedigree that may not be based on observable, empirical evidence but rather on a mythos to keep the upper class relatively "pure" and the number of rich CEOs controlled to (mostly) existing class members, while attempting to appear egalitarian ("anyone" can go to an ivy league school…

What's interesting about what you said is that if one were to look at the CEOs of the Fortune 500, this idea of classism breaks down.

CEO of IBM: Ginni Rometty -- grew up in a single parent home, with a mom that worked multiple jobs. Ginni got in to Northwestern and studied computer science and electrical engineering. She certainly didn't have any upper-class advantages. She went to public school.

Geisha Williams, CEO of PG&E (part of the Fortune 500) -- daughter of Cuban refugees -- literally arrived in the US with nothing, her father was a political prisoner in Cuba before being released. He worked multiple jobs to support his family. Geisha started her career as a residential power auditor and worked her way up to CEO. Hardly a "silver spoon" story.

Rex Tillerson (former) CEO of Exxon -- one of the largest companies in the world. Picked cotton and worked as a janitor before getting accepted to the University of Texas with a band scholarship. He joined Exxon as a production engineer and then worked his way up to CEO over the next 30 years. Hardly an "elite."

Tim Cook -- son of a dockworker and a pharmacy worker. Went to Auburn University (hardly an Ivy League, "elite" school.) He worked at IBM from 12 years working his way up to the director of North American fulfillment while earning a Duke MBA at essentially night school.

I could tell these stories all day. There certainly are some elites who do go on to be CEOs, but there are even more CEOs and self-made millionaires that aren't Boston Brahmins or cousins of the Kennedy family.

This whole "classism" idea is promoted by three types of people:

a. those who are actually ignorant about American business and its leaders. Spend some time actually learning the backgrounds of American business leaders and you'll see more "State U" than "Harvard U."

b. those from other countries where a public school girl from a broken home would never become CEO of IBM or a redneck janitor on a marching band scholarship could never become CEO of Exxon -- these are often Europeans where, in places such as France -- if you aren't a male who went to a Grande Ecole, you will never be CEO of a major French company. CEO of Groupe PSA (France's largest car maker) -- Carlos Tavares, graduate of Ecole Centrale Paris. CEO of Renault? Also a Grande Ecole graduate. Former CEO of Airbus? Louis Gallois, also from a Grande Ecole. CEO of LVMH? Bernard Arnault, another Grande Ecole graduate. The "egalitarian" French have nothing but elites leading their top companies. In the US, the most valuable company's CEO? Tim Cook -- Auburn University -- a state school. Rex Tillerson? University of Texas. CEO of Walmart (the top company on the US Fortune 500 by revenue) -- University of Arkansas and University of Tulsa. CEO of United Health? Dave Wichmannm, Illinois State University with an Accounting degree -- not even an MBA and he's the CEO of the number 6 company in the United States. CEO of United Airlines? Oscar Munoz, the oldest of 9 kids in a Mexican-American family in California and the first to graduate college in his family. Some elite. What "classism" helped Munoz? He's the CEO of the world's largest airline.

c. the type of person that is angry at their own life, their own bad decisions or their own lack of success that they want to create a political narrative that supports the "evil capitalism" trope. This is the type of person that feels like they have been wronged by life and society and rather than looking at their own failings, instead attempt to blame some boogeyman. Maybe their daddy didn't love them enough. I don't know. But these are negative people that claim conspiracy when there is none. They themselves just don't think they're good enough and they need an excuse to justify it.

Now, if we were talking about politics -- THEN I might agree that classism has some influence. When we have a brood of Bushes, a variety of Clintons and an infestation of Kennedys, then it's pretty clear classism does have a real impact. But in the C-Suite -- I would argue that the evidence suggests that classism isn't a factor at all, instead it's about hard work, tenacity, luck and intelligence.

Re: Au Revoir

#259
post #167
post #156

Earlier quoted context omitted.

Bob Young founded Lulu.com, the ebooks and print books company, before Red Hat, IIRC. That was an early pioneering ebook and self-publishing company startup, I think.

Lulu.com was founded in 2002 after Bob Young left Red Hat.

Okay, I got that the other way around, thanks. Seems there was quite a gap between the two foundings, too.

https://en.wikipedia.org/wiki/Red_Hat

https://en.wikipedia.org/wiki/Lulu.com

Post reply on HN