Hacker News Reader

Comment by the_doctah | Hacker News Reader

What hubris to think that your system is so complex that other engineers couldn't understand it.

Also what hubris of the Twitter engineers, who seem to be in this thread, thinking anything they built was that special.

Your video player somehow managed to be worse than Reddit, the perennial joke of web dev.

Not to mention “something went wrong” again and again, maybe by the 4th reload would a linked tweet actually show up for me.

I’m on 25 gig symmetrical fiber with <10ms to my local AWS hub, FWIW

> Not to mention “something went wrong” again and again, maybe by the 4th reload would a linked tweet actually show up for me.

That seems like a purpose-built feature to punish non-Twitter users to induce them to sign up for Twitter and/or use the Twitter app instead of the web UI. It's not a bug but a misfeature.

No, that happens on the Twitter app too.

Video player works fine for me, and I'm on 20 gig symmetrical fiber with 8ms to my local Azure hub. Maybe check your firewall settings?

Oh please, that's a ridiculous assessment of what was said. Coding giant distributed web apps is a very different domain than coding for cars. I wouldn't put much stock in Twitter engineers assessing self driving car code either.

"Tesla engineers don't just work on in-vehicle code though. The servers and systems needed to implement Tesla's telemetry and software updates would be in the same ballpark of expertise, albeit with much less scale and different requirements than Twitter."

-- hn_throwaway_99 https://news.ycombinator.com/item?id=33389128

Yes, both were me. I'll respond since you seem to be implying my responses are incongruent.

Here I was responding to a comment "What hubris to think that your system is so complex that other engineers couldn't understand it." Which is still a bullshit assessment of the argument (which the original comment was about) that you need expertise in specific domains to be able to evaluate code, which I agree with.

My comment you quoted above is that I think that there would be a subset of Tesla engineers (just not the ones that, for example, solely focus on driver systems) that would overlap with Twitter's domain.

>Oh please, that's a ridiculous assessment of what was said. Coding giant distributed web apps is a very different domain than coding for cars. I wouldn't put much stock in Twitter engineers assessing self driving car code either.

You don't need to be a part of any particular domain to effectively audit code.

This is true to an extent. At the extremities, a software system can be audited without any previous domain exposure: architecture, and implementation means (language, deployment, etc.)

It can be difficult to critique choices, specially in system architecture, when domain requirements are presented or falsely asserted as 'dictating the choice', but this can be addressed by directing conversations with stakeholders to nail down the conceptual model that has been implemented in architecture style x, and then the conversation is on more equal footing.

Auditing language usage, libraries, development process, etc. are much less domain specific.

Really the only way I've been able to make sense of it is if he brought people in to produce an audit of changes made in the last couple months. IE he knows he doesn't know how to get that information, but he knows he has people at Tesla who do.

They wouldn't have to understand any of it, just be able to identify where and when changes occurred, for later review for any last minute shenanigans.

I'm not personally convinced this whole incident even happened, but if it did, this is what I think would make sense.

What hubris to think you can wander into a system as complex as a $44bn tech company and draw any meaningful conclusions about it at all in a weekend. I mean, in my world we have some regular "tech reviews" with our Sr. Directors/VPs at a pretty abstract slides-and-diagrams level, and even that takes weeks of prep leading up like 4+ hour presentations. For something with only a glancing relationship to code level reality. I have occasionally seen my work reflected in these reviews, and the game of telephone that produces them is quite lossy.

Any competent engineer can come to understand most any part of the system, but it's going to take minimum several weeks to understand at a level where you can seriously challenge the domain experts about what's really going on in a little 0.01% corner of the engineering shop.

The hubris is thinking Tesla engineers (who are offered worse pay and wlb relative to Twitter) are coding masters of every domain. I won't even snipe at their own software here.

It takes a long time to onboard to a big mature infrastructure that is serving hundreds of millions of users for people moving from a similar company, even longer if your day job is whatever Tesla engineers are doing.

And that's onboarding, not pretending to be an expert which seems to be the role they're taking on.

>The hubris is thinking Tesla engineers (who are offered worse pay and wlb relative to Twitter) are coding masters of every domain. I won't even snipe at their own software here.

Who is thinking that?

I think you think that the GP is thinking the Tesla engineers are thinking that :D

On one hand, you can usually safely ignore 99% of code in a system that does the plumbing in a system and focus on the few key algorithms. It wouldn't be a stretch at all to take one of the self driving AI guys and have them evaluate say, an ML classifier used to flag spam/abuse content. That would be well inside their domain.

> It wouldn't be a stretch at all to take one of the self driving AI guys and have them evaluate say, an ML classifier used to flag spam/abuse content.

Facebook has a years-old bug with their ML classifier used to flag spam/abuse content in Facebook Groups. Its exact behavior has morphed over time as they flailingly attempt to fix it, but for several straight months that behavior was "posting to a Group via the Graph API fails with 'unknown error' if it includes a dollar sign". This in an API used by tens of thousands of apps and many millions of users, at one of the richest (in both money and programmers) companies on the planet.

ML is often a black box "computer says no" scenario with little meaningful ability to debug.

Key algorithms to do…what, exactly? Why not just read the documentation on that instead of having people bring you a random sample of their code on paper?

Who said they're evaluating the code? Maybe they're evaluating the people by asking them questions about the code they've just written. That's a great way to find out who's full of shit.

Why the printouts then?

So they can have the code on hand while interviewing them, ask about the code, why they made a certain choice, what other options were available, etc. Have you never had an interview with code printed out or otherwise provided for discussion?

Usually we both have copies on our computers.

> That would be well inside their domain.

Would it?

>Would it?

Would it not?

Knowing enough to call bullshit is way lower than knowing enough to contribute meaningfully to a codebase.

Even just having access to code while someone talks I can prove with much higher value questions on what they are saying… because even if I’m asking something dumb it prompts a deeper explanation.

Pay and work life benefits are generally independent of technical competence. But I agree with the rest of your comment.

You don't need to be a "coding master of every domain" to recognize shit code.

Is there any reason to suspect that there is shit code at Twitter? Or that that is in any way related to the problems that Twitter has or the "problems" that Elon said he wanted to fix?

There isn't any reason to expect that there isn't shit code, either.

If you spend $44 billion dollars on something, you're going to want to look it over extremely well.

Reply on news.ycombinator.com