HNHacker News
TopNewBestAskShowJobs

peterwwillis

1 karma · joined March 4, 2021

submissionscomments
peterwwillis··on US plans to produce ‘Iberian’ ham in Texas and Georgia
It's already nearly impossible to find jamón ibérico, much less de bellota, in even the expensive U.S. grocery stores. This will hurt Spanish sales much more in the developing markets where it's more in demand. America will just keep eating its crappy knockoff products like it always has, and you'll still be able to special order the real thing for a small fortune.
peterwwillis··on I Tried to Live Without the Tech Giants. It Was Impossible
> When you install Linux, where is the OS downloaded from? Where do the software packages and updates reside?

About 1000 mirrors distributed around the world by independent institutions using donated storage and bandwidth. Some distros may have "corporate-funded" primary mirrors on providers like AWS, but independent mirrors of those still pop up. The majority of the mirrors are universities and ISPs.

> What about the source code?

Most distros have "source mirrors", as most package managers will build a "source package" which includes the source code used to build the package, in addition to vendor patches and build instructions. If they don't provide a source mirror, you go to wherever the dev kept their source, which used to default to either SourceForge or a personal website, and is now mostly GitHub.

By default most distros have a list of mirrors that they try, and they also allow you to specify a specific list of mirrors if the defaults don't work. If your country shuts off internet access to the rest of the world, and there's no mirror in your country, someone can stand up a mirror at a local business and you can use that. And you can install Linux distros from a DVD, CD, floppy disk, USB stick, and local LAN. Even if you only have one copy of the install media for one release of one distro, you can make a mirror out of it.

peterwwillis··on I Tried to Live Without the Tech Giants. It Was Impossible
The author's major complaint is that they can't have their cake and eat it too.

You can easily live without tech giants! Stop using a smart phone, use a Linux PC/Laptop, pay for your few online services like e-mail, stop using the internet for shopping and everything else. You know, the way the entire world worked 15 years ago.

> Yes, there are alternatives for products and services offered by the tech giants, but they are harder to find and to use.

If life is somewhat difficult without giant tech corporations, this is inherently unacceptable? No - this is called living in the real world. You do very often have to choose between an immoral yet easy life, and a moral yet difficult life. This has been our constant choice for literally millennia. Every society has had to grapple with it. Entire religions are based on it. It is the completely unavoidable part of life that everyone is forced to participate in.

> If I were still blocking the tech giants today, I wouldn’t have been able to watch this week’s antitrust hearing online. C-SPAN streamed it live via YouTube, which Google owns.

So get a TV! I don't know if you know this, but C-SPAN is a TV channel! You can watch it without the internet! You can even make comments on it without the internet by calling a live show and speaking your mind! Amazing! (But, probably don't do this, as the average caller seems to have about the same political insight as the average YouTube commenter)

This whole premise - that life is impossible without tech giants - concedes that life is impossible without the internet. I assure you that it is not. It's up to you to decide whether that convenience and lack of expense is preferable to the alternative, which is doing things "old school".

peterwwillis··on Mozilla lays off 250 employees while it refocuses on commercial products
Most people on HN (nee the world) don't think critically when they see a clickbait title / submission. I've worked for many companies that have been flogged by the users of public forums, without those users having the slightest idea what they are talking about. But being angry sure does feel right. (I am guilty of the same...)
peterwwillis··on We Can Do Better Than SQL
If you want people to take your replacement language seriously, lead with an RFC that's been reviewed by industry peers. We all know the status quo sucks, but it's the status quo for a reason.
peterwwillis··on I want to have an AWS region where everything breaks with high frequency
> Basically I want an API where I can torture an EC2 instance to see what happens to it, for science!

  # cat > dropme.sh <<'EOFILE'
  #!/bin/sh
  set -eu
  read -r SLEEP
  tmp=`mktemp` ; tmp6=`mktemp`
  iptables-save > $tmp
  ip6tables-save > $tmp6
  for t in iptables ip6tables ; do 
    for c in INPUT OUTPUT FORWARD ; do $t -P $c DROP ; done
    $t -t nat -F ; $t -t mangle -F ; $t -F ; $t -X
  done
  sleep "$SLEEP"
  iptables-restore < $tmp
  ip6tables-restore < $tmp6
  rm -f $tmp $tmp6
  EOFILE
  # chmod 755 dropme.sh
  # ncat -k -l -c ./dropme.sh $(ifconfig eth0 | grep 'inet ' | awk '{print $2}') 12345 &
  # echo "60" | ncat -v $(ifconfig eth0 | grep 'inet ' | awk '{print $2}') 12345
If you're lucky the existing connections won't even die, but the box will be offline for 60 seconds.
peterwwillis··on Fire your bad customers
Unless they're cheapskate big dysfunctional companies.
peterwwillis··on IAM is hard – Thoughts on $80M fine from the Capital One Breach
Make IAM part of the design of the application. If you need to use AWS API calls, then you need your app's design architecture docs (you have those right?) to list the IAM permissions needed to do each thing, and stuff that info into ADRs, and link to some IaC that you used to stand up your dev/test environment. All of this creates 1) formal IaC used to apply the permissions, 2) formal documentation of what functions need what permissions. As a final step during development, write tests that verify the IAM permissions are as expected.

You will of course need some way to scale this later, but as long as an artifact of your build pipeline is auto-generated IAM policy jsons, an org-wide security team can analyze them with automated tools and remediate as needed.

peterwwillis··on Wheat yield potential in controlled-environment vertical farms
I think what you're asking is, couldn't a Beowulf Cluster of wheat be as efficient as a Mainframe of wheat? And I think that it could be, but there's so many "ifs" involved there it's crazy to speculate at the moment. We might get flying cars before we get massively parallelized vertical underground farms.
peterwwillis··on Why shaving dulls even the sharpest of razors
I have used straight razors, safety razors, electric razors, and various cartridges. A good multi-blade cartridge is vastly superior to anything the layman can do with the first three. They shave all kinds of growth in one pass, don't cut you, and get very close, without needing any foam. They really are the pinnacle of shaving technology. It's amazing how people ignore them for older tech which can only match them with a lot of extra work.
peterwwillis··on The creeping scourge of tooling config files in project root directories
If the organization of files in your repo is your biggest problem, you've obviously got excellent code and no tech debt.

In which case you probably have enough free time to build a tool that just moves files around at runtime. Organize your repo however you want, then run your tool as a wrapper for any other command, and it'll move all files into the places other tools expect them. Now you've solved the problem, which was that humans like aesthetically pleasing things that serve no useful purpose.

peterwwillis··on One year of automatic DB migrations from Git
Some DBs have inherent limitations and you need to design something to get around them. Sometimes you have many apps and you don't want to reinvent the wheel to support what are otherwise standard accesses of a database. Sometimes it's a "special" database which makes this more convoluted, or perhaps you need to implement things like a gradual back-off retry to prevent swamping the backend. Sometimes you don't want to have to patch every app that uses the database if a database change is breaking. Sometimes you want to prevent developers from abusing your database, provide more fine-grained access control, prevent poorly researched queries which can kill performance, etc. Sometimes you're talking to more than one db.

Personally, I would start writing the abstraction as soon as I plan to have more than two apps.

peterwwillis··on One year of automatic DB migrations from Git
> your app being quite tightly coupled with the database

I mean, this is the case any time your app talks directly to a database

peterwwillis··on Why is America stuck with bad headlights?
Sounds like too much tongue weight. Assuming you aren't exceeding your vehicle's tow capacity, the tongue weight should be about 10% of the loaded trailer weight (check your owner's manual) which the suspension should be designed to handle and thus not tilt your vehicle under load. Or you may have a worn suspension...
peterwwillis··on Why is America stuck with bad headlights?
The American market has had "adaptive headlights" for nearly 20 years. They tilt up and down, side to side, depending on the orientation of the car. My 2004 Lexus IS300 had them. And they were HIDs, so they were plenty bright. (My base model 2012 Toyota's headlights completely suck in comparison...)

But our headlight market and rules are pretty ridiculous. Most "fancy" non-HID bulbs are a ripoff and actually make visibility worse, and it's also illegal to convert a halogen setup to HID. Combine this with the fact that some designs are just crap no matter what the bulb type, and we have a general market failure to understand the real performance of headlights. People think they can generalize headlight types, but it doesn't bear out in practice. https://www.consumerreports.org/headlights/are-hid-and-led-h...

peterwwillis··on Not an Amazon Problem
This focus on business and capitalism is just one part of a larger jigsaw puzzle, which is an economy based on American exceptionalism and the American dream. The former states that we must be unique and the best, and the latter states that everyone is able to have a castle, a good job, leisure time, retirement, etc. But they don't explicitly state how we get those things.

At no point in the American story does it say that wealth is to be distributed fairly. The past century has shown that as popular revolts upended the old ideas of aristocracy and monarchy, the US elite became extremely fearful of the idea of a more equal society. Communism was the worst specter of this potential loss of personal wealth for the elites, but its modern forms are also increased taxation, universal health care, increased public education/transporation/housing, etc. The more you distribute wealth to the rabble, the less wealth you can have as an individual, and Americanism is all about personal wealth (personal freedom comes largely from personal wealth; you can't even be the king of a puddle if you don't own the land it sits on).

It's similar to ancient Rome. The citizens can live quite comfortably as long as they have soldiers, slaves, and foreign riches. The big corporations of today are a modern-day Roman army: roving the world to capture and bring home cheap goods. The generals, senators, and merchants get the majority of the loot, and the citizens get fresh bread and garum, fruit, and the occasional exotic spice.

And much like the circuses of Empire, as long as the rabble are kept in bread and games, they'll put up with anything. No health care, no education, 3 jobs? Keep them in bread and games and they'll be satisfied. Netflix provides the circus, Wal-Mart provides the bread. Why would the rabble want them broken up when they're getting exactly what they want?

So this notion of fighting the status quo is really anti-American. You can't get far with the argument without butting heads against the fact that people would have to give up their cheap t-shirts, and that some would always have to work hard and live poor their whole lives. There's no way around it: somebody has to pay for our wealth, surplus, and convenience. If it's not us, it'll have to be someone else.

peterwwillis··on Cloudflare Workers Announces Broad Language Support
I appreciate their support for customization, but what I as a customer really want is pre-built, tested, supported, maintained, features.

For example, CloudFront has features that let me create redirects for different content to go to different origins and prefixes, but Cloudflare does not have this option. You can write a .js edge worker to do it, but I'd rather not trust the integrity of my site to some crappy javascript that I threw together, and I'd rather not have to maintain my custom code indefinitely. (Actually Cloudflare's support technicians will offer to write the custom code, but it's still not something I would like to have sitting out there outside of a VCS and SDLC, with bit rot, tech debt and other nasties lurking in the future)

peterwwillis··on Small mail server best current practices
> The big players are able to deal with the spam using specialised teams and I suspect advanced algos including machine learning.

I remember when I imagined big companies might use their considerable resources to put together something like that. Then I found out it's just a handful of random engineers cobbling together crap and hiding it behind a fancy domain name.

It is probable that Office365's entire spam filtering system is a very large and ugly spreadsheet that gets passed around between a few teams using a ticketing system.

peterwwillis··on More than 1k people at Twitter had ability to aid hack of accounts
It's actually fine to have multiple people get access to production databases, if you have an automated change management system.

Change Management authorize sactions which could be problematic by vetting them through a process. The end result of the process is more confidence in the change: we know who's making it, when, where, why, what the impact will be, what we'll do if it goes wrong, etc. If the change looks stupid, dangerous, or uncertain, it gets rejected. If it looks good, it gets approved. The appropriate access is doled out temporarily to make the change. In this way, Larry the Janitor can deploy changes to your prod db, and you can be just as certain (maybe more so?) about the outcome than if your 10x developer were doing it.

Just one example of this is Terraform applies. By loading your changes into a .plan file and requiring approval of the .plan, you know for certain what actions are going to be taken and approve only those. (That doesn't stop Terraform from totally hosing your system due to all the ways it can't predict the outcome of its actions, but at least you know what the intention was)

peterwwillis··on Ask HN: Is all of FAANG like this?
Don't think about whether this is normal. Just ask yourself what you would like to be doing instead. Can you do that while barely doing any work at FAANG and collecting a paycheck? Or would you need to change jobs? I would personally wait until the recession has hit bottom before changing jobs. But then again you could get hit by a truck tomorrow.

I can tell you that lots of huge companies have this, but not consistently. Some teams move fast while some move glacially. Greenfield work can move the fastest, unless you get stuck waiting on someone to build environments.

Definitely don't stay there too long if you're unhappy. Worst case you try something new and it stinks, in which case you try again.

peterwwillis··on Consider getting a “DIY” degree online instead of a traditional CS major
> No college life: you will be missing on the college experience. This one is a big one.

This is the biggest one.

After dropping out of high school, I never attended another school. The amount of socialization and long-lasting human connections that I missed out on is incalculable. Not to mention exposure to different subject matter. Macintosh had great fonts because Jobs took a calligraphy course on a whim. You might discover a hidden passion for entomological forensics. You might join a friend for a gap year trip, meet an amazing person, get married, move to Spain, get divorced 3 years later, become an accountant. Or experience the rush of unity and purpose from joining your classmates at a protest march. Or attend your dorm mate's band in some dinky basement and fall in love with beatboxing to electro swing. Screw education; go to college to wade hip deep into new experiences.

$40K to be totally immersed in innumerabile possibilities that will effect the next 80 years of your life? Compared to ~$15K for an economy car, or ~$250K for a house? Sounds worth it to me.

...that said, if you have economic hardships, self-study is completely feasible, and you can have a great career with no degree. It will still take you years to really get going, but it can work.

peterwwillis··on Show HN: Dockerfiler: declarative management of images built from Dockerfiles
It's your company. I've seen it a lot. They want to go whole-hog on some new technology, but they never learned how. So then they adopted more new technology to solve not knowing how to use the previous technology.
peterwwillis··on Cloudflare was down
Or just stop using the internet. The majority of tech problems stem from people using tech. Don't rely on it alone and you don't have problems.
peterwwillis··on Cloudflare was down
Yesterday CloudFlare took down some of our products because they (not us) misconfigured some DNS thing. Kind of funny to see it happen again a day later.
peterwwillis··on Twitter hack has profound national security implications
Twitter as a platform is engineered to be pithy, immediate, and constantly engaging with as little forethought as possible. It's a machine for viral stupidity. You could call it a mis-communication tool.

I get why people like to use it, but people also like slowing down to stare at car crashes on the highway. Maybe it's not a good thing.

peterwwillis··on How to write technical posts so people will read them (2019)
# Tips for Writing Documents

## About

This comment explains what I know about writing docs to make them easier to understand. A succinct paragraph at the top should state what the rest of the document will cover. Don't create a document with random "stuff" in it and not explain what that "stuff" is at the beginning.

## Header Sections

Break up your document into discrete sections. They can be categories, a series of steps, etc. These headers break up the page into logical chunks of information. The header title pre-loads your brain to receive the following information, and makes it easier to skim. These header sections become a Table of Contents later.

## Grouping Information

A leading sentence groups a set of correlated bullet points:

  - bullet points are useful visual indicators
  - they break up important points
  - they are very easy to absorb individually
Call out the major steps needed to perform an action.

  Step 1. Figure out the major parts of your instructions
  Step 2. Break up instruction into discrete steps
          - Provide sub-grouping for complex steps
  Step 3. If additional steps are required but not included here, call out 
          that they are necessary and where to find further information.
## Adding Emphasis

Using specific emphasis can help to underscore __unique information__ and make it easier to find specific information ON THE PAGE. It also helps break up concepts visually, making it Easier To Read.

Emphasis makes skimming easier. Skimmability makes it more likely people will absorb the information you want them to.

## Be descriptive

Using extra, otherwise unnecessary words can "de-compress" a concept so that it takes less brain power to understand what the words mean. It may not be beautiful prose, but it will be easier to understand to more people.

Don't be afraid to break a paragraph up to make it easier to read. One idea spread over many paragraphs is a lot easier to understand than trying to pack it all into one.

## Don't make assumptions

Probably more than one person will be reading your document. We all have different life experiences and will interpret it slightly differently. Ask yourself if all your readers know everything you know. Probably not, or they wouldn't be reading your doc!

Before you dig into something complicated, think about what information is necessary to understand it. List the pre-requisite knowledge needed to grok the next information. Provide background information for concepts you haven't covered yet, and refer to other documents when possible.

## Use Graphics

Complex concepts can be understood easier when presented visually. If it would take you two pages to explain something that a simple diagram or flow chart can express, start with the visual. Use the text to explain details that the diagram doesn't show.

# Edit, edit, edit

Drafts are important. Constantly revise your document as you write it. Review it and see if re-writing or re-organizing sections makes it easier to read. Read it from the beginning and see whether your edits have stopped making sense, and edit again.

Documents get better when you revisit them later and solicit feedback. A fresh perspective is always good.

peterwwillis··on Everything you need to know about OAuth 2.0
This introduction is pretty simple and concise to just understand what it is: https://www.cloudflare.com/learning/access-management/what-i...

Follow that up with this SAML and OAuth comparison: https://www.ubisecure.com/uncategorized/difference-between-s...

Then see how OpenID Connect fits in: https://www.okta.com/identity-101/whats-the-difference-betwe... https://www.gluu.org/blog/oauth-vs-saml-vs-openid-connect/

And this page shows examples of a web app using OAuth2: https://connect2id.com/learn/oauth-2

peterwwillis··on Etcd, or, why modern software makes me sad
> If you are running a truly enormous system and want to have off-the-shelf orchestration for it, Kubernetes may be the tool for you.

Not even then. If you're an enterprise using EKS, you still have to work around the pain points, missing features, incompatibilities, and do a metric shit-ton of custom integration. And that's before even touching CI/CD, and ignoring how you build the infrastructure.

It's like a Formula 1 engine. Really complicated, you still have to assemble the rest of the vehicle, and you need a team of engineers to understand what it's doing when you drive it.

I think the author is right. gRPC is bullshit that you only use if you're dedicating your whole organization to one protocol. They actually re-added that wacky backwards-REST API-gateway because people really needed a REST API. On the other hand, Consul not only has a REST API since the beginning, but it incorporates more useful (some would say necessary) features so you don't have to write them from scratch. So you can go with Consul and get more compatibility and features, or go with Etcd for the opposite.

But at the same time, a "lightweight database written around Raft consensus" is like saying "clogs for runners". Most people shouldn't use it, and those that do won't be happy with the consequences later.

peterwwillis··on ELF Statifier, self-contained dynamically linked executables
Why? Every sandbox that has ever existed has been exploited/broken out of, including VMs. Only explicit security controls like RBACs (ex. SELinux) can create a secure runtime environment, if you configure them right.

It's much, much easier to just download a binary and compare its cryptographic hash with the origin's before running it. That's how all Linux distributions ship apps, and Windows has a slightly more modern version of that.

peterwwillis··on Tutorial sites treating FreeBSD like a Linux distro
I really miss BSD (I'm a Slacker so at least I get to be nostalgic about it, even if Slack and Linux in general is annoyingly "modern").

I would like to use it again, but I'm a afraid I'd miss out on the latest developments that happen only in Linux-land (I'm still pissed that containers are such a Linux-centric piece of tech when they in no way need to be).

← PreviousPage 7 of 34Next →