Sodman

Being able to add arbitrary logic to an otherwise static binary has a lot of interesting use cases. For example Envoy Proxy supports hot-loading arbitrary HTTP filters compiled to WASM into its filter-chain. This allows users to extend the feature set of Envoy, by adding custom logic to request parsing, reporting custom metrics, calling out to other services before forwarding the request, etc etc.

This allows their unique custom use-case-specific code to be managed separately to Envoy, so it can be ported across upgrades etc. It also means that if there's a very particular feature you want that's unique to your business, and wouldn't make sense to contribute to the upstream OSS project, you can write a WASM module for it.

It's somewhat niche, because anything built in this way which is generally useful to many users, will probably just be made a native feature of the host platform. Conversely, anything that does something significantly complex and custom should probably live as it's own stand-alone microservice somewhere. There's still a lot of room for value between the two extremes though!

solo.io | Software Engineers | Boston, MA (Cambridge), or fully REMOTE (NA) | Full-time

Solo is an application networking company working with open source, cloud native technologies. If you are passionate about technologies like containers, Kubernetes/k8s, Istio, Envoy Proxy, GraphQL, eBPF, serverless functions, and more, then Solo.io is the place for you! We strive to productize the bleeding edge of cloud technologies, and build solutions which scale globally. If you're interested in presenting your work at conferences and meetups, that is also encouraged and supported (although not mandatory). We are primarily a golang shop, although no prior go experience is required.

Apply here - https://boards.greenhouse.io/soloioinc/jobs/4026989005

Or see our other openings - https://www.solo.io/company/careers/

Sure, but "Pay $502 for the privilege of running code on other people's machines" doesn't seem like a big improvement?

At least to get a driver's license you need to pass a driving test, and return periodically to update the photo and pass an eye exam.

And the jQuery it pulls isn't even gzipped or otherwise encoded! Although at least it's minified.

In the interview process, ask questions like "How many days off did you take last year?" or "How many days off in a year does an average team member take?". If the answer is <= 15, or they bristle at the very mention of the question, there's your answer!

As an anecdote, I've worked at a startup with unlimited PTO where the culture was that most folks were working 45-55 hour weeks regularly, but everyone took vacation whenever they felt they needed it. I averaged ~5 weeks of vacation per year taken during my tenure there. That was probably about par for the course with that team. As long as you're performing well while you are working, give enough advanced notice, and willing to be somewhat flexible (eg don't take 2 weeks vacation right before a big deadline with 2 days notice) - it can work out quite well for all parties.

Thanks for this, I didn't even see the labels and was re-reading the text trying to see if they had specified which chart directions represented which measurements!

Unfortunately GH doesn't have a way to alternate images in a readme based on light mode vs dark mode afaik.

The company having that much influence over your day-to-day life during non-business hours every fourth week is an enormous burden. I would want at least 50% higher overall total annual comp to even consider that schedule, and even then it's still only a maybe.

I love on the checkout page that they explicitly tell you not to check the "please remember my information for next time" box, "for the sake of your own privacy" - very on brand :)

solo.io | Software Engineers | Boston, MA (Cambridge), or fully remote | Full-time

Solo is an application networking company working with open source, cloud native technologies. If you are passionate about technologies like containers, Kubernetes, Istio, Envoy Proxy, GraphQL, eBPF, serverless functions, and more, then Solo.io is the place for you! We strive to productize the bleeding edge of cloud technologies, and build solutions which scale globally. If you're interested in presenting your work at conferences and meetups, that is also encouraged and supported (although not mandatory). We are primarily a Golang shop, but no prior go experience necessary.

Apply here - https://boards.greenhouse.io/soloioinc/jobs/4026989005

Or see our other openings - https://www.solo.io/company/careers/

If you're US based, there's Code for America which is exactly that - https://codeforamerica.org . There are many local, typically city-based chapters who meet up regularly and are formed almost entirely of part time volunteers contributing whenever they have time, to whichever project match their skills and interests.

I'd love to see some more great Free-For-All (FFA) games. So much interesting emergent gameplay, game-theory musing, and player-driven engagement. I used to love Warcraft 3 custom game modes that provided this style of gameplay. It results in folks teaming up, then there are inevitable betrayals, back-door trading deals, and all sorts of player-invented fun layered on top of what is otherwise a relatively basic game.

One of the best takes on this FFA style that I've played recently (released 2015 though) is subterfuge - http://subterfuge-game.com. It's a mobile game that pits up to 10 players in an FFA game. It's more or less a "risk" style game. Expansion is rewarded with better army production rates and higher army capacity, at the cost of larger borders to defend. Orders can be issued in real time, and the game has built-in player group chats. However, the game is set up so that after attacks are launched they frequently take 8-10 hours to reach their destination, meaning a game typically lasts 1.5-2 weeks, which has some very interesting side effects on the meta side of the gameplay, as there's plenty of time for scheming with other factions in between orders.

My ideal game is a game that lasts 1-2 hours, features 6-10 players, and incentivizes each to striving for an individual win against the other 9. The actual game mechanics need to provide some kind of resources that can be traded, some kind of cost/benefit to expanding your in-game power, and a large benefit to teaming up with other players to fight a third 2v1. There also needs to be some relatively-costly way to knock out other players, which frequently incentivizes mad-dash finales, suicide runs and all kinds of other player-driven shenanigans.

I think a big contributor to this "Brilliant asshole" trope is that if you ignore some things like teamwork/communication it's a lot easier to appear to be doing 10x the work. This can manifest in so many ways...

Not doing any code reviews? Leaving tests "for somebody else to write"? Refusing to work on legacy code-bases and only working on greenfield projects? Tuning out (or just not even joining) API design meetings so you can work on your own stuff? Bonus points if you have the authority to turn around in a few weeks and make folks re-write the API because you don't like the decisions you ignored - those pesky 1x engineers, eh?

There's a bunch of things that fit into this category that can make an engineer seem like they're doing 10x the work, but if you look closer it's frequently only because they're taking shortcuts or pushing more work onto their colleagues.

To be clear - I'm not saying 10x engineers don't exist, just that folks with "teamwork issues" who appear to output 10x as many features are frequently not as productive as they seem.

If effectively enforcing the ban on impaired driving was a viable option that was truly on the table then yes, I would 100% be in favor of requiring this testing to happen privately. Breathalyzer for engine start, eye tracking to prevent phone usage (which Tesla has btw), more frequent vision/driving tests, etc would improve road safety significantly. But that's just never going to happen, it's a completely unrealistic scenario that requires a huge culture shift, and would be incredibly unpopular policy. Not to mention the public transport infrastructure that needs to be built to support all of these folks who suddenly are no longer allowed to drive but live in areas where driving is a necessity of everyday life. The demographics who'd be affected hardest by these kinds of changes are also vocal voters with high turnout.

Given the state of the public roads now, full of dangerously impaired driver's as it is, we've clearly decided as a society that it's worth the risk to allow this behavior to continue. So in that light, I think it's worth allowing these public tests of autonomous systems under some strict set of rules. I agree these rules shouldn't be exclusively self-enforced by Tesla and co., and there should be some more stringent open transparent reporting requirements around how many systems are active, miles driven, incident reports etc. But with the right controls and oversight in place I think it's a worthwhile endeavor to allow the testing to continue in the spirit of accelerating autonomous driving becoming a reality. It's the impaired drivers that _make_ it more urgent for this system to exist in the first place.

Similarly, you and I are both forced to share the roads with folks who are impaired in various ways, whether it's declining reflexes, vision, cognitive function - or just straight up drunk driving. As somebody who comes from a country with very strict anti drunk-driving rules (checkpoints randomly on commonly used roads frequently enough to deter it), it blows my mind that I have to share the roads in the US with drivers who think having "just 4 beers" before their 45 minute drive home is fine.

Since the US hasn't shown many signs for cracking down on drunk driving, and adding better infrastructure for folks who can no longer drive for various reasons is equally a non-starter at the moment - I'm all in favor of any system that speeds up the arrival of genuinely fully autonomous driving. Tesla hasn't killed or injured anyone using this "FSD beta" program to date, despite a large number of miles driven, so their existing precautions appear to be working.

Even in the interim, before we reach fully autonomous driving, "better than the average human driver" is still a huge performance milestone worth aiming for before we start obsessing over "perfect" (Not saying FSD Beta is there yet, but Autopilot on Highways might be).

Asynchronous culture and no-meetings-ever is great if you're a somebody who has the ability/experience/authority to take a high level task and run with it without needing to collaborate with or have input from anyone else, but that's not how most companies work in reality.

If you're waiting for multi-day turnaround times on agreements because there's back and forth where two parties are trying to have a debate over requirements in a document but both only check for updates every few hours, I'd argue you're not being as "productive" as you think.

It's true that many meetings could have been an e-mail, but there are a few crucial meetings that can save days by just getting everyone together in a room for 30 minutes to reach an agreement.

Conversely, other people are annoyed with "selfish" people forcing what could otherwise be a productive in-person meeting to now need to include one person dialed in over zoom, perhaps without a video feed. Hope you weren't planning on collaborating using physical media such as whiteboard or post-it notes. Obviously we can do our best to reduce unnecessary meetings, but for those remaining necessary meetings, an in-person meeting is almost always more productive than a zoom meeting in my experience.

This is why both sides are so fired up over remote vs in-person. Your decision to work either remotely or in person has negative consequences for your co-workers either way. WFH folks are mad that they are being asked to commute, in-office folks are mad that they are being forced to use clunky online tools strictly to accommodate their remote-only peers.

OP's point though is that in some companies, revealing that you have kids at home unlocks a "go home early and never be contacted outside of work hours" perk which is unavailable to those who don't have (or want to share that they have) kids.

It's a tough balance, because if you don't allow the same perks to your childless employees, they end up being "suckers", working extra hours for no extra benefit. Conversely, if you were to eg pay them more, even for working more hours and being extra contactable outside of regular hours, it could be seen as discrimination.

The 6 pro is 0.3" taller and 0.1" wider than the 6. But both are also significantly larger than the previous generation pixels. The Pixel 6 pro for example is 0.8" taller and 0.2" wider than the Pixel 5, which is a significant difference (especially considering you lose the rear fingerprint scanner that gave the ability to swipe down notifications from the unreachable top of the screen).

I hate the trend that the "pro" versions of all these flagship phones need to be an XL Phablet monstrosity. Human hands haven't gotten any bigger. I have owned most of the Nexus & Pixel phones throughout the years. After having the Nexus 6 as my daily driver for 2 years, I have always opted for the smaller option ever since.

I would totally pay flagship prices for a regular ~5.5-6" phone with flagship specs, and I'm sure I'm not the only one. Make it 50% thicker if there are space/heat concerns, but making it wider and taller just makes it super difficult to use consistently with one hand. I also say this as somebody with larger-than-average hands!

Certainly not, but if we invent a life-saving medicine that 1 in 1 million people have a fatal reaction to, we'll still consider it enough of an overall positive for humanity to celebrate it as a technological win for society.

The ratio isn't 1:1 or even 1:10, but there is a line somewhere where X deaths <caused by new tech> is acceptable because of the X^Y lives saved <by new tech>.

See also: Most modes of transport, most medicines and surgical procedures.

Conversely, if you work for a company that lists it's "values" as something like "Being customer-focused", it's easy to find ground upon which to push back on things that go against the company's supposed values.

Software engineering is largely think-work. Some of it is creatively coming up with ways to solve novel technical problems, but an often understated part is thinking about how your implementations will affect your users, and optimizing for solving your users problems.

For folks switching to iOS for the first time from Android the experience can be incredibly frustrating, if you don't know that you're supposed to do things "The Right™" way.

Examples from personal anecdotes include:

- My first Macbook pro several years ago was the cheapest one at the time, and basically got stuck in a boot loop when I opted out of creating an apple ID on setup. Needed to boot into safe mode and factory reset it to start again.

- Setting up an iPad (in the last year) for a first time iOS user, required two 2FA challenges and needed the apple ID password ended no fewer than 7 times in the first 5 minutes of operation. The App Store didn't work for a few minutes for reasons unknown, and then with no changes, started working. (I'm guessing the creation of a new apple ID takes a few minutes to propagate to the app store or something?)

I'm sure everything goes smoothly when you just log in with an apple account and import all of your existing settings from iCloud etc, but if you're jumping in for the first time and press the wrong buttons it's rough!

OP here, first, the "only for 3+ users" part is definitely strange, I didn't even realize that there was a group discount when I posted this, I just opened the e-mail they sent me and thought it seemed good - so my apologies for the misleading post title.

As for the "Why Google Fi" question... I have been a Google Fi user for about ~5 years now. First, as you said the international roaming is a big feature for me. I (used to) travel to Europe from the US quite a bit, and with Fi I never had to worry about anything. Just charged my standard plan and get a notification when my plane lands in a new country, otherwise business as usual.

But the main "feature" for me is "not having to deal with the phone companies". My experience with T-Mobile & AT&T in the past has been awful. Long wait times calling customer service for functionality that should be available via a website. Sales people trying to upsell when you try to cancel plans etc. In contrast when the pandemic started I switched my Google fi plan from "unlimited" to "flexible" by hitting a single prominent button the Fi app and I was on the cheaper plan for my next billing cycle. Now my phone bill is like $25 / month because I'm mostly on WiFi at home in the pandemic, but if my data usage creeps back up in the future I know swapping plans is literally as easy as hitting a button.

Finally, 22GB is not unlimited, true. I only use my phone on my daily commute but even if I spent 100% of my commute streaming videos on my phone I don't think I'd hit 22GB in a month. If you use more than 22GB mobile data, then Fi is probably not for you. However, unlike other other providers, at least they're completely transparent about it. It's literally in the second line-item of their plans page. Most carriers just say "Unlimited!*" and you have to dig deep into the ToS to find out what the "real" limits are. And if you find yourself somehow over 22GB and need* fast internet again, you can just pay for more if you need to get yourself out of a bind.

Not GP but usually the big reason for not buying is that you either need a large amount down (20% in my market, or 10% + additional monthly "Mortgage Insurance" payments until you hit 23% ownership). Average home price in the "Greater Area" around the city (meaning you might still not live close enough to be able to use public transit) is currently ~$665,000 according to Zillow, which means you need to come up with ideally ~$133,000 to put 20% down, or ~$66,000 and pay a premium on top of your mortgage. Either way, this huge down-payment is in addition to the ~7-15k in closing costs you'll be paying. Even then, you're bidding against cash buyers who are willing to waive inspection, so good luck.

$70k-150k up-front isn't easy to save up for even above-average earners (remember, these are average home prices, not luxury homes), so anyone earning average or below is forced into either renting forever, or moving to another town.

A lot of people are going to jump on the "he used k8s and he doesn't even work at Google scale!" part of this writeup, but I think it's a perfect demonstration of the concept of innovation tokens [1]. He admits in TFA that clickhouse was the only new piece of tech in his stack, and he was already familiar with k8s et al - so he's able to focus on actually building the products he wants. I could see somebody unfamiliar with k8s (but very familiar with all other pieces of tech in the system they want to build) being able to learn it as part of a side project, if it's the only new thing. Where the wheels come off is when you've never touched k8s, postgres, aws, rust, graphQL or vue - and you try to mash them all together in one ambitious project.

[1] https://mcfunley.com/choose-boring-technology

We talk a lot about the "You can't make me go back" crowd, but I don't think we give much thought to the "Why do I have to bend over backwards to support a few fully remote co-workers" crowd. Being fully remote works well (for some people) when the whole team is remote. My biggest fear is that hybrid environments can bring out the worst of both worlds.

I wonder how many people will quit if they're forced to work with remote-only employees on a regular basis? If you're having a meeting with 4 in-person co-workers and 2 remote co-workers, you have to make all sorts of accommodations for the remote folks that wouldn't be necessary for a meeting with the same 6 people in-person. "Can you repeat the question" / "Can you share your [other] screen?" / "Can we record this meeting?" / "Can you hold the camera up to the whiteboard?".

Similarly, if you're in the minority of remote folks on a hybrid team, you'll be at a disadvantage. You're likely forced to either miss out on some context/discussions from not being present or not being able to hear, or become "that person" asking folks to repeat themselves or specially accommodate you in various ways. Not to mention missing out on face time with the boss[es boss], which can (rightly or wrongly) often factor in to promotion/raise discussions.

For a business this is basically free. You could honestly even think of it as a net-positive for some teams. Think of all of the times you were in a group meeting and somebody has to repeat themselves because someone else was typing, living next door to construction, ambulance drove by, etc. $300 / year for a team of 5 more than pays up for itself if you think of it in terms of hourly salary. For a $120k/year salaried engineer, effectively paid $60 / hour, you only need to save them 1 hour per year for this purchase to be financially worth it for the business. Not to mention the less tangible benefits like "being less distracted/stressed during meetings".

> 80% of employees could be more productive remote, 18% could have serious issues with productivity

My experience has been almost the inverse of this. I wouldn't quite say 80% of people are less productive at home but I'd think at least above half? For various reasons mentioned elsewhere here (disruptive home environment, can't communicate effectively with co-workers remotely, etc). Many people just become very unproductive if they're not physically in the presence of other people being visibly productive.

I'd like to see some formal studies into this to get the true split.

Wow you're entirely correct, I just checked again 3 days later and there's now magically 3 in stock again :) Thanks for the tip!

This looks much more like what I'm looking for alright! Doing some background research it looks like they had some issues with their chipset supplier and had to downgrade the chip a little, but performance benchmarks put it ~around the same place as my Pixel 5's current CPU, which I'd say is much more acceptable (if not totally cutting edge flagship-tier) for a phone in 2021!

At $576 (Edit: now $621, looks like the early bird just sold out!) this also seems like a much better deal than the pro-1x at $800! I'm always worried though when the "pledge store" for these things says things like "Ships March 2021!" but their most recent update says they've just finalized their PCB and are just starting testing...