HNHacker News
TopNewBestAskShowJobs

robmccoll

1,411 karma · joined June 5, 2014

www.robmccoll.com

[ my public key: https://keybase.io/robmccoll; my proof: https://keybase.io/robmccoll/sigs/rFoGh_YBcy3PiVvcUp7vfPqM_pDQqXXNuHrkyeB756I ]

submissionscomments
robmccoll··on Online Identity
It would be more like receiving mail from a service than sending mail to a service which isn't a problem. There isn't really a spam component.

I show up to bigsaas.com. Big SaaS asks me to identify myself. I submit that I am rob@robmccoll.com optionally with a proof of that claim.

If I have provided a proof (think attestation along the lines of "The holder of this certificate is rob@robmccoll.com requesting to use bigsaas.com at time T using signing key Kpub") it will be signed under a key chain whose root can be obtained from a specifically formatted DNS TXT record under robmccoll.com (similar to DNSSEC or DKIM - maybe it gets a custom record type). bigsaas.com can fetch said key and verify the signature.

If I haven't provided a proof, bigsaas.com looks up a DNS record under robmccoll.com that points to which server provides this authentication service for the domain and does something like an OAUTH or SAML type of exchange but without the need for point-to-point pre-configuration. It beings this exchange with a signed request "This is bigsaas.com attempting to authenticate rob@robmccoll.com at time T" which is signed under the key in its own DNS records (or at least a chain that can be verified under that key).

Either way, the service doesn't have to have some pre-seeded trust beyond trusting DNS (and this should probably insist on DNSSEC and TLS-only for whatever interactive authentication gets layered on top)

Edit: I guess this is kind of OpenID, but to be honest, OpenID looks way more complex than necessary.

robmccoll··on Online Identity
This is why we need a federated protocol. Most people can just trust a major player. More technically inclined people can host their own.
robmccoll··on Online Identity
I just want a real federated identity protocol with public key distribution and access delegation through signed certificates. Do discoverability through DNS. Tie identities to providers via their hostnames (user@provider seems to work fine for email).
robmccoll··on I Avoid Async/Await
Everytime I use a language with async / await syntax, I find myself wishing that awaiting async functions was the default behavior and that I had to specify when I didn't want that behavior. So much boilerplate, so many times I've seen bugs that are the result of not being aware of a function being asynchronous or that the function was later made asynchronous. Most of the time you want seemingly synchronous behavior within a request because it's easier to read, write, understand, and reason about and concurrent progress across requests. The places where I can overlap I/O within a request can be manually specified. I guess this is kind of how Go works (although that's parallelism and concurrency).
robmccoll··on Human brain compresses working memories into low-res ‘summaries’
Nice to know that there's an upside to this idea :-)
robmccoll··on Human brain compresses working memories into low-res ‘summaries’
Wow - just watched a talk Nick Chater gave. Sounds like an interesting model for consciousness. I'll check it out. Thanks.
robmccoll··on Human brain compresses working memories into low-res ‘summaries’
Fascinating - thanks for the article. It's strange to think that however much we think that we're completely rational and can trust our own memories, we're more like malleable rationalizing machines.
robmccoll··on Human brain compresses working memories into low-res ‘summaries’
The weird (scary?) point will be when we figure out how to subtly present adversarial information to the brain that will be coded in a way that collides with some target information to induce false recognition/ memories.
robmccoll··on In defense of complicated programming languages
A thousand times this! At the very least, I always want to have good mental model of how something probably works or mostly works even if I couldn't reproduce the implementation line-for-line. To me, it can almost be dangerous to have the power to use something without any idea of what's under the hood. You don't know the cost of using it, you don't have a good basis for knowing what the tradeoffs and reasons are for using it over something else, and the concept isn't portable if you need to work in a language or environment that doesn't have it.

If I come across a feature I like in a programming language, I usually find myself trying to figure out how to implement or emulate it in a language I already know (ideally one that won't hide too much from me). Implementing coroutines in C using switch statements, macros, and a little bit of state for example.

robmccoll··on Facebook versus The BMJ: when fact checking goes wrong
That or they would turn to a model where they charge their users to post which would result in a reduction in content on its own without Facebook having to intervene or be held responsible for it. Only those who are really willing to stand by their content would be willing to pay (even if the fee is comparatively small). Networks of utter nonsense being posted and mindlessly reshared by large networks of bots and people who plainly don't care to fact check anything themselves would likely dry up.

Largely, there is a lot of misinformation spreading on Facebook that has harmful effects. Freedom of speech is tremendously important, but typically the tradeoff is that it comes with some personal responsibility / liability depending on who the speaker, the audience, the intentions of the speaker, and the effects of the speech on the audience. The current platform setup manages to shift that responsibility away from anyone. I think most reasonable people would agree with this, but perhaps many more people than I would think believe in limitless speech without consequences.

robmccoll··on Facebook versus The BMJ: when fact checking goes wrong
Neither. Change the law to make platforms like Facebook accountable for the content published on them. Here's how you do that without damaging speech or making platform companies impossible: make platform protections dependent on revenue stream. If you are paid by the author to publish their work, you are a protected platform and it isn't your content. If you make money by selling advertising alongside the content or by selling a subscription to the content stream, you are exposed to libel lawsuits, incitement, hate speech, false advertising, etc. The rest will sort itself out.

Previously journalism cleaned itself up because it could be held accountable for outright lies. There is incentive to make outrageous claims because outrage sells. The balance is to create civil and criminal repercussions for provable lies. Lies spread on social media like wildfire because outrage sells and more eyeballs means more money for the platform, so the platform and the authors are incentivized to create and spread misinformation. There is no balance against it, so it will continue to grow.

robmccoll··on Google says iMessage is too powerful
I think I understand your point of view better now. Thanks!

I still wish Apple would have pushed to develop an open standard as the path to improved user experience and messaging rather than making the progressive enhancement path go from an open network to a proprietary one. On the other hand, walled gardens have always been a productive pursuit for them, so why change?

robmccoll··on Google says iMessage is too powerful
Not sure what to make of your "if facts were different, then things would be different" - I still think the analogy holds up ok.

You could say that things have changed as users have become more aware of the difference over the years, but honestly, if your initial expectation was the default message application on my phone does what I have known as texting (which is/was SMS/MMS) which is a part of my phone plan that I pay my provider for and then you start doing something wholly different with that communication, I would consider that some sort of hijacking. You have seized my action and intention and redirected it to serve your own purpose.

Optimally? Switch over to an RCS implementation. If existing rollouts of RCS don't support enough of a feature set to please you (looking at E2EE), push to get that done. Don't go the Google route and end up choosing "collaborating is hard, so let's just not do it".

If Apple won't do that, drop SMS from iMessages and open the protocol or at least force users to choose or acknowledge the choice. Auto-negotiating a protocol choice makes sense when you are going from one open standards-based protocol that any vendor could and should implement to a newer version of that protocol with the same expectation of general availability (think the path through the versions of SSL and TLS). It's disingenuous to automatically shift traffic and offer enhanced features through a proprietary protocol when communicating with clients that only implement and can only implement the open protocol. It makes it seem like the other clients are intentionally inferior or otherwise making user-hostile choices. It's good for business though.

With respect to Android, that's its own slew of problems and mistakes that shouldn't be the bar Apple needs to clear. iMessage and Facetime are easily better than any communications implementation Google has produced in the last 10 years. Hangouts and its various precursors were better when they were based on XMPP or had functioning XMPP gateways. XMPP wasn't an standard intended to replace or progress SMS, so that doesn't really meet my criteria either. RCS is the way forward, but there's little demand for it or business reason for it in the Apple camp. Maybe if carriers declared that they are going to start taking RCS seriously and start removing SMS from plans and charging more for it.

robmccoll··on Google says iMessage is too powerful
It's not hijacking a protocol - it's hijacking a client for the protocol. Most users were likely not aware that when they open iMessage (the default SMS application) on their brand new iPhone and send a message to another user by phone number that the application is actually choosing between routing that message via iMessage through Apple's backend or via SMS over their carrier's network.

To me this would be similar to me operating as a browser vendor and having my own CDN that implements a separate protocol from HTTP, provides additional features beyond what HTTP can do, and displays a different color of banner to indicate that the website you are using is coming over my CDN. Then I don't allow arbitrarily anyone to implement my CDN protocol or connect to my CDN, so other browser vendors can't provide a competitive experience. Later when I have built a large marketshare, I decide that I'm good with HTTP 1.1 or whatever version of HTTP at the time and stop upgrading my implementation of that protocol. This is damaging to the other browser vendors and of users of websites that only support HTTP. Obviously this is an imperfect analogy given the client-server nature of HTTP, but I figured this might resonate with this audience.

robmccoll··on Ask HN: Are most of us developers lying about how much work we do?
Yes! I bucket people into paddling the right direction, sitting in the boat, and paddling the wrong way. With enough momentum, you can tolerate the people that are sitting in the boat. Sometimes they have good reasons and it's temporary. The people always actively working against progress need to be overboard. Smaller teams may need all paddlers. I'd like to think I'm always paddling and usually the right direction.
robmccoll··on Show HN: Caffeine – Minimum viable back end for prototyping
In queries, you should use the database/sql.DB interface if possible with your database https://pkg.go.dev/database/sql#DB.Exec

It should sanitize / quote arguments for you and protect against SQL injection. Note that this doesn't mean all data sanitization is performed, just the basic '; do my stuff here; -- type of things.

robmccoll··on Show HN: Caffeine – Minimum viable back end for prototyping
Using the db Exec / Query / Query row is the same amount of code and effort the fmt.Sprintf statements. Even in quick-and-dirty mock-ups, it's a good idea to not cut corners there.
robmccoll··on Making the dislike count private across YouTube
> But that's why you have a brain, so that you can make that decision for yourself! When you censor an opinion by downvoting something, you are making an unfair decision for future viewers of that content who would rather make that decision on their own.

Would you make the same argument in favor of getting rid of upvotes?

Also, expressing disagreement isn't censorship. The availability of the content is unchanged. YouTube doesn't eliminate content from their platform based on some number of downvotes.

robmccoll··on Making the dislike count private across YouTube
I don't think this is a good idea. To me one of the biggest problems with social media that propagates disinformation to large unquestioning audiences is that there are easy ways to reinforce the message and to see how much it resonates, but it is hard to gauge when an idea is controversial or if there are many people who disagree. It is worth way more to see dislike counts and ratios on dangerous content than to protect creators from the internet being mean at all costs.
robmccoll··on The metaverse is already here, it's called the internet
Yeah as a trail runner and technologist, I enjoy VR but I'm not sure anything could replace the sights, smells, and feeling of going on a run through the woods. Similarly I think there is room for both VR-sports (and e-sports) and traditional IRL athletics. One doesn't replace the other, but the skills, strategy, physical performance and endurance are very different to watch and to experience (each great/ fascinating in their own way).
robmccoll··on The metaverse is already here, it's called the internet
The thing about the internet as we interact with it today that makes it so pervasive is that it generally enhances real world experiences. It doesn't replace them. It fills in the gaps in down time. People whip out their phones to text or take a video or catch up on social media or share something because the interaction is fast and the interfaces are complementary and the right balance of distracting enough to draw your attention, but not distracting enough that you can't convince yourself that you are loosely aware of what's around you. AR might work, but VR outside of your home seems like a stretch.
robmccoll··on Exploring PL/pgSQL: Strings, arrays, recursion, and parsing JSON
The idea of defining and executing a recursive function inside my production database seems a little scary. Might want to build in your own recursion limits to that one. Is there a way to set execution timeouts or something?
robmccoll··on Exploring PL/pgSQL: Strings, arrays, recursion, and parsing JSON
I use this https://github.com/golang-migrate/migrate in a deploy step to each environment with the upgrade / downgrade scripts committed to the repo alongside the code. The scripts can do pretty much anything you need to do in PG including defining and executing functions.
robmccoll··on After a decade as Apple’s assistant, Siri still hasn’t figured out the job
Blaming the user for the lack of simple coordination between devices when they buy into your ecosystem across multiple devices and attempt to use the advertised features doesn't seem like a great product strategy.
robmccoll··on We killed our end-to-end test suite
I don't know anything about what Nubank is up to or how things work there overall, but integration tests are absolutely worth doing. The argument against this to me reads like "coordination and testing of big systems is hard, so let's not do it."

> Waiting. Engineers had to wait more and more to get feedback from this long-running suite;

"Our tests are inefficient, not sufficiently parallelized, the setup / tear down of the test environment isn't optimized, and it isn't possible to run only the relevant subset of tests during feature development or bug triage for short feedback loops"

> Lack of confidence. Flaky tests meant that we had to re-run the suite frequently to see if something was really wrong or just a false negative;

"Our tests aren't well written (we have sleep-polling)", "we don't build-in testability into our system (we can't introspect or wait on the thing we care about in the test, so we have massive work arounds)", or possibly worst "our system is flaky and our tests reflect that".

> Expensive to maintain. Manual changes in our staging environment corrupted test data fixtures and maintaining the environment “clean” was a challenge;

"We haven't spent enough time developing our own tools for testing, so we have tests that are extremely fragile (think copy and paste of massive JSON blobs with comparisons just to check a handful of values)"

> Failures don’t point to obvious issues. Test failures were very hard to debug, specially due to our reliance on asynchronous communication that make it hard to connect the cause of failure (a message not published to a queue) with its effect (changes not made in another system);

"Our system is over-engineered and our service boundaries match our internal structure rather than clean separation in the functions of our APIs. We don't have good visibility because doing any one thing involves massive levels of coordination. We lack proper tracing and aggregation."

> Slower value delivery. Queueing of commits in the End-to-End suite resulted in less frequent deployments;

"Quality is hard and takes time. Let's not do it so we can move fast and break things."

> Not efficient. Few bugs caught in this stage. One experiment suggested that, for every 1000 runs, we had 42 failures, only 1 bug;

See above about flakiness and fragility. Also, integration tests catching bugs tend to point to really bad obvious bugs. I'd be happy about the one that was caught.

> Not effective. Bugs were still being found in production.

"We still found bugs. This means testing must be ineffective altogether?"

robmccoll··on OSHA to require employers with 100 employees vaccinate or test workforce
Fair point. I'd be curious to see it as data points on the range and variance in opinions on the site. I have less interest in which individual comments are "controversial".
robmccoll··on OSHA to require employers with 100 employees vaccinate or test workforce
Yes. Based on my traumatized friends that work in ERs in a few different states and the people I know who have died, I would say that it is much more reasonable to believe that those numbers are close to accurate than to suspect a large scale conspiracy, but I guess the world is less interesting and more depressing when you consider the likely reality over shadowy intrigue.
robmccoll··on OSHA to require employers with 100 employees vaccinate or test workforce
Society is a closed system. Mandating wearing seatbelts only saves me right? Except that someone will have to go to the effort of scraping me off the sidewalk, attempting to resuscitate me, transporting me to the hospital, resources at the hospital will be consumed that could have otherwise been preserved or applied elsewhere, and if I'm not insured, the cost of most of this will be borne by others. If I don't make it, what happens to my dependents? Friends? Family? People counting on me at work?

Communicable disease and taking reasonable precaution isn't that different. This one is interesting because its virality and fatality rates are apparently on the cusp of what society as a whole agrees are worth doing something about.

robmccoll··on OSHA to require employers with 100 employees vaccinate or test workforce
Have you considered that we took a lot of precautions that made it possible that as relatively few people died as did? Even with shutdowns, lockdowns, and extreme measures in healthcare, we lost about 0.2% of the population. If we had done nothing, we might have reached 1%. If a group was going door to door randomly killing 1 in every 100 people in your neighborhood, would you consider it reasonable to stop answering the door for a while until they left?
robmccoll··on OSHA to require employers with 100 employees vaccinate or test workforce
I really wish HN had a controversy upvote / downvote ratio - maybe multiplied by number of votes?
← PreviousPage 6 of 14Next →