HN user

gitonup

123 karma

Seattle tech since 2006.

Microsoft Exchange - ActiveSync Microsoft Word - core development Indeed - resume and sourcing agency Various startups

Posts0
Comments31
View on HN
No posts found.

The culture of Google is that you hire "the best" developers (say top 0.1%) and hamstring them with process and cumbersome tools so that you need 10x as many of them as another company would need and pay them 3x market rates, but it is OK because (1) at the scale they work at they can amortize the cost over a large user base and (2) they make monopoly products.

This describes basically every FAANG / MANGA company. Or even past that, any company that hit it big with a cash cow and now needs to come up with something new to satisfy shareholders.

In Meta's case, they have 3B MAU, they absolutely hire from the same tier of developers, and (pre-layoffs/economic downturn) they throw 10x more of them than they need at a problem. They even outcomp Google. The number of employees is more because employee growth was an indicator for company growth and only once that became a liability against the stock price it stopped.

Meta is just a newer company than Google.

Facebook on the other hand, cares about internal DX, sees it a problem when developers are stuck with cumbersome processes, and greases the skids.

I am married to a Meta engineer and have mentored folks that have gone to work there. This might be the case if you work for a product that drives their cash cow of ads, but if you are doing anything that doesn't fit within that narrow bucket ... let's just say our viewpoints diverge significantly.

The problem with React is that mediocre developers can use it to build big things that are too big for them to handle. The problem with Kubernetes is that above-average developers can use it to build small things they can't handle. And the scalability of Kube is more than almost all of the industry needs.

The problem with the Chrysler 300 is that bad drivers run over people. What does that say about the engineering culture at Chrysler?

Look, I agree that most cloud stuff is overkill, but I have a hard time indicting Google's entire engineering culture over a project they released into the wild 12 years ago that just happens to not fit your use case and that theoretical "above-average" developers wouldn't be able to tell that.

If I had a choice between React (Facebook) and Kubernetes (Google) I would pick the former anyday.

Can you elaborate on why these are at all comparable techs to use as a developer?

React seems to be the frontrunner in FE, but what do you see the BE equivalent to be?

Measurable responses to the environment lag, Moore's law has been slowing down (e: and demand has been speeding up, a lot).

From just a sustainability point, I really hope that the parent post's quote is true, because otherwise I've personally seen LLMs used over and over to complete the same task that it could have been used for once to generate a script, and I'd really like to be able to still afford to own my own hardware at home.

Disclaimer: raised Catholic, now Atheist, married to devout Catholic.

The Church as defined by the institution is a community. I do not see it as a contradiction that the head of the institution is instructing the leaders to not add more layers of abstraction between them and the community, especially when those messages are on the subject of what it means to be human.

Your boss, asking for a project to be finished early, may be an overdemanding boor – or just an Asker, who's assuming you might decline.

I don't pay for the Atlantic and thus am limited by paywall, but this ignores power dynamics.

This is the God's honest.

I worked on the MS Word core team for a little over three years from 2010-2014, and de-facto owned a significant part of implementing ODF / OOXML Strict support.

The binary format was a liability for Microsoft to begin with, because of decades of cruft lining up with actual memory alignment. During my tenure there I ran into code my GM had written as an intern and was still intact -- he had 20+ years of tenure (mostly on Word) when I joined the team.

The translation of the file format to XML involved a significant amount of performance degradation if you weren't careful. Hundreds of millions of people use the app monthly, and MS still tries to maintain backwards compatibility. Given that open APIs were a relatively late development for the app, I really don't think in the current reality of what's expected by boards of directors for the companies they oversee that _anyone_ would take years to:

a) define a spec that maintained that backwards compatibility

b) reach whatever nebulous simplicity metric today's HN article wants

c) not get whoever greenlit the project fired for taking that many engineering hours for a and b

amounts to having political control over land use decisions

This exists in America, in ways that have generally escaped the label of "Communism". The most basic example of this most will be familiar with is zoning laws, but there is significant precedent otherwise. There will always be a gradient of control, and claiming that a singular government action in expansion is therefore communist is not intellectually honest.

Detroit hasn't, downtown has.

Downtown has improved wildly since when even when I was growing up near there in the 80s/90s, but people don't realize how large Detroit is. It's nearly 140 square miles. Downtown is, charitably, 2sqmi, and that's including all the way up to Wayne State. The light rail is the People Mover loop downtown plus a trolley up Woodward, within that range.

The theft mentioned in the article occurred at 8 Mile and Wyoming, which is a 20 minute highway drive from Campus Martius in no traffic. "Clean Downtown" that happened when the city hosted the super bowl and other related efforts have very lopsidedly focused on the downtown area.

Everyone else they had interviewed had stormed out, pissed off that they had wasted their time; they didn't even realize what they did until I started asking questions after I got off the floor.

Uh, those that stormed out did it right after the salary piece, and they didn't put it together?

The point is that "escape" in this case means not having to work to American Capitalism levels, not escaping working entirely. Retirement in the FIRE sense does not imply in all cases spending every day by the pool doing nothing, and even in the AmCap sense, most who retire and tend to live longer and (self-reportedly) happier, have some work that they are doing, but more along the lines of what even a medieval serf would have enjoyed to the tune of like 15 hours a week.

ETA: often this work is volunteerism, which does not have a capital incentive in the individual case here.

I've seen this cited several times in this thread, but I've seen no one mention that (according their own FAQ):

medRxiv (pronounced "med-archive") is a free online archive and distribution server for complete but unpublished manuscripts (preprints) in the medical, clinical, and related health sciences. Preprints are preliminary reports of work that have not been certified by peer review. They should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.

The context is important, even if it turns out to be true.

Regarding taste tests, it's common for very sweet drinks to win because they give a small amount to taste. The experience of drinking the normal serving size (in the US) of 12oz is a different beast.

This behavior by recruiters is wildly frustrating to me even without NDAs involved. I can't count the number of times I've actually been looking for jobs and gotten a coy "I have a role you'd be perfect for" message from a recruiter, then when I ask who the company is they outright refuse to tell me til I get on the phone with them.

For the couple of times I've played ball with this in the past, they've told me who the client is on the phone, so why the secrecy when it's over an asynchronous messaging platform? I don't owe you synchronous time and it'd save us both a lot of headache if you'd just be upfront with who you're pitching.

As another developer in his 30s, I don't think it's quite the same.

Mystery meat navigation is purely opaque, whereas icons like GMail's are at least trying to convey meaning via their imagery. The Wikipedia page for Mystery Meat Navigation actually has an example that is completely opaque as well, as does this example video by the coiner of the term: https://www.youtube.com/watch?feature=player_detailpage&v=zZ...

I don't think it's quite fair to call bad iconography "Mystery Meat" because it's not executed well. The difference is that users can (and do) learn what the icons mean and then can find them visually, whereas in true MMN, you'd absolutely have to move your mouse over what you were trying to find.

GMail has the challenge of trying to not make their UI busy while at the same time trying to come up with iconography for "Archive", which is not as prevalent in the zeitgeist as "Bold" in a word processor.

In the context of that sentence I take independent to mean "unrelated by connection to each other" rather than "unrelated by ideology." In that context I don't see a claim that they're non-partisan.

For that matter, nor can I think of any outlet in modern days which would both a) report that information were they to find it and b) not be immediately and loudly labeled partisan by a significant portion of the internet.

I've edited my OP and hopefully what I'm saying is more clear.

If the entire ecosystem is your IDE, you're not "Turning Sublime Text into a Lightweight Python IDE".

I don't know how else to phrase this to make it clear that I'm not attacking, denigrating, claiming inferior, etc., composite workflows like vim + GDB. But if you get to an article that says XYZ is now your IDE, and it offers a piss-poor debugging experience unless you leave XYZ, then that article is not turning XYZ into an IDE.

My definition of what? I literally said there's nothing wrong with taking that approach. I quite purposely did not argue productivity.

My issue is with the acknowledgment by the author that the standard for "IDE" includes debugging capabilities and then goes on to provide a plugin that provides follow-along only via running GDB, which would need to be configured (I assume, since the author does not specify) the normal ways.

If you want to argue UNIX is the IDE I can get behind that definition, but the debugger is absolutely not integrated into Sublime (or Vim or Atom or N++ or ...) in any meaningful way.

The final requirement for a reasonable IDE experience is to be able to follow a debugging session in the file where the code exists.

There is no plugin for SublimeText that supports this. But there is a Python package you can install into the virtualenv for each of your projects that does it.

This drives me batty whenever I see "turn XYZ into an IDE for (foo)." The author of this post at least lists debugger integration as part of the IDE experience, but comparing against actual full fledged commercial / non-plugin-composite IDEs? I haven't seen one article for turning Atom/ST/Vim/etc into an "IDE" that comes close (edit: that is, close to a reasonable debugging experience, in an integrated way) capability-wise.

There's nothing wrong with taking that approach, but if you need to shim on GDB and use a plugin to just follow along in source (forget setting watches, gutter-click breakpoints, conditional breakpoints, etc., from Sublime Text), that's pretty definitively not integrated into Sublime.

Indeed | Senior Developer IC | Seattle, WA | ONSITE | Fulltime | www.indeed.jobs

Work for the world’s #1 job site.

https://www.indeed.jobs/career/JobDetail/Senior-Software-Eng...

Our Seattle office has exploded in size in the past couple of years and we need developers with strong CS backgrounds. Web focus is a plus. We run an MVC webapp-backed-by-microservices platform, with a large Hadoop slow-cycle processing backend and more and more technical challenges to work with daily at a global scale with millions of users.

We have a gorgeous office space downtown near Pioneer Square by the water. All of our openings for this office can be found at: https://www.indeed.jobs/career/JobListSeattle

We're looking for someone who Has a passion for run-time performance Thrives on simplifying complex information Loves to learn and develop new skills continuously Has a strong sense of ownership in every application they build Requirements BS in Computer Science or related area, or relevant work experience Minimum 6 years experience in software development with at least 2 years in Java programming Deep background in information retrieval (IR), knowledge representation or computational linguistics Significant experience with large scale, high performance systems. Experience building solutions for high traffic web sites a plus

You can drop our Seattle recruiter a direct line by b64decoding amdhYnJpZWwtY2UraG5AaW5kZWVkLmNvbQ==

I've seen orgs run on Excel spreadsheets: one electric utility I worked for described their load in one massive Excel sheet. Where I think this analogy falls apart is that so many spreadsheets come to depend on macros (which my experience did, via ODBC), which makes them a meld between 2 and 3. I've definitely written systems at level 2 -- at the same company -- that have had way more ROI than the spreadsheet based system that was previously employed, but the spreadsheet could have employed macros to do the same thing.

Indeed.com | Seattle, WA | Fulltime | Multiple Openings | ONSITE, VISA (can transfer H1-B) | https://www.indeed.jobs/career/JobListSeattle

Indeed in Seattle is hiring for many roles, but especially for Technical Product Managers and Product Scientists.

In a Technical Product Manager we are looking for a feature owner able to wear many hats with a strong CS background and a track record of delivery. Our Seattle office in particular is launching new teams around privacy and our pipeline of jobseeker data acquisition among others: https://www.indeed.jobs/career/JobDetail/Technical-Product-M...

In a Product Scientist, we are looking for a hybrid Data Scientist and Product Manager. We are continuously improving how we match job seekers to employers and vice versa, as well as improving our understanding of how millions of people monthly are using our site. An ideal candidate will use quantitative analysis, data mining, and machine learning techniques to understand how users on our site interact with each other, and use that information to drive new product features and improvements: https://www.indeed.jobs/career/JobDetail/Product-Scientist/3...

We have a gorgeous office space downtown near Pioneer Square by the water. All of our openings for this office can be found at: https://www.indeed.jobs/career/JobListSeattle

You can drop our Seattle recruiter a direct line by b64decoding amdhYnJpZWwtY2UraG5AaW5kZWVkLmNvbQ==