Hacker News Reader

Hacker News

Comment by miiiiiike | Hacker News Reader

Full thread

miiiiiike

Issue: Feeling bad after interaction with OSS developer at <BIG COMPANY>.

Steps to reproduce:

1) Report a bug.

2) Get called dumb by developers at <BIG COMPANY>.

3) Demonstrate that it is, in fact, a bug with examples, test cases, and offer a potential fix/PR.

4) Hear that it’s being “handled internally”.

5) Wait for the blog post re-iterating what they called dumb, talking about how great the team is that fixed it, and see your demonstration code or test cases in the linked commit.

6) Why are you still here? No, you’re not going to be credited, this is going in my promotion packet.

I’ve lost count of the times I’ve seen this over the past 20 years. Whenever I’m about to report a bug in a <BIG COMPANY> project I smile to my girlfriend and say “OK, I’m going to go get called dumb.”

View on HN

And the Windows Terminal team reeks of this. There's so many blog posts saying look, we're so smart because they're reimplementing something both Linux and macOS had for *years*, as if the whole world of CLI was waiting for their savior.

When CMD is your benchmark...

(In all fairness, PowerShell has a lot of good ideas)

Yeah it is far superior to cmd. But I mean cmd was frozen in the 90s it seems anyway.

cmd is a shell. The old terminal is called conhost I think.

I figured I'd get this from someone. ;) I'm approximating that 99% of people who use Windows use default terminals for built-in legacy shells, and thus effectively analogous terminology. You're correct though!

I get the feeling that there's a bunch of people working for, or being paid by, Microsoft who really love functional programming. Some of Microsoft's best outputs comes from those teams/folks.

It could just be a coincidence/confirmation bias, so I wonder if there's actually something to that.

If you care about programming enough to really have preferences on things like functional or OO or other kinds of "deep" topics, then you're invested.

> I get the feeling that there's a bunch of people working for, or being paid by, Microsoft who really love functional programming.

Simon Peyton Jones, Simon Marlow, Edward Kmett, Conal Elliott, and I’m sure a lot of other people with fundamental contributions to the Haskell ecosystem who I can’t remember offhand all work or previously worked at Microsoft Research, so that part is just straightforwardly true.

Are PowerShell and Windows Terminal the same team?

I can’t speak to that. I’ve pretty much Linux only since 1999, but, of the times I’ve used PowerShell, I’ve like it.

Windows Terminal is a separate project from PowerShell. It's just a terminal that you can use with different shells (including PowerShell).

I have no idea what their blog posts are saying, but at least I've always found it pretty nice.

PS has been pretty slow though. I’ve liked the fact that you can pipe objects, but it seems to slow down the startup time of the shell.

True, the slow startup is annoying (especially if you load modules in your profile), but overall I still think it's better than the alternatives. I usually just keep a terminal window always open.

PowerShell is available for Linux if you want to use it more without Windows.

The whole world of CLI has been using a poor open source reimplementation of linux subsystems on windows for decades.

> There's so many blog posts saying look, we're so smart

Have you read Casey's twitter feed? The man routinely goes into long twitter thread rants about how stupid everyone else is.

His favorite target seems to be...drumroll please...large corporation software teams.

> rants about how stupid everyone else is.

> His favorite target seems to be...drumroll please...large corporation software teams.

To be fair, that's... not unreasonable. I've thought unkind things about large corporate software teams while I was working at said large corp.

> rants about how stupid everyone else is.

> His favorite target seems to be...drumroll please...large corporation software teams.

I missed the part where he is wrong.

the part where "he" is wrong, is the part where he tells everyone that his reference software can be compared to a real terminal. You cannot compare shoes with towels, nor cats with a TV. The tiling renderer "he" wrote as part of the reference software to demo isn't that bad as a product (to some degree), but that piece of software is garbage, as it has zero use. And almost none of the techniques "he" brags about can be applied to a real terminal. I mean, I like being educated. But so far, noone of the internet fanboys "he" has, has so far ever provided a real product out of his reference software. Admitted, some people tried (just look around), but they all failed. I wonder why. Do you know it?

> the part where "he" is wrong, is the part where he tells everyone that his reference software can be compared to a real terminal

He very clearly states that his implementation is a terminal renderer.

> And almost none of the techniques "he" brags about can be applied to a real terminal

Elsewhere one of the people from the screenshots says "we implemented the same approach" because these techniques can be and have been applied. Both in Windows Terminal and, as I understand it, in every single modern terminal emulator.

Why is the word “he” in quotes? Are you implying that they’re trnsgnder or something? But you don’t put “his” in quotes, so that probably isn’t why?

It is confusing.

He's not wrong though. Large corp software teams, paradoxically, tend to run stupid even if comprised of smart people.

Thing about Casey's rants is that he's completely right though. This is obvious to pretty much anyone who isn't being disingenuous about it. And his writing takes on an admittedly aggressive tone because, frankly, we've been dealing with shitty software that keeps getting shittier written by self-aggrandizing people for a decade or so now. He tried being diplomatic in his issue report and he was shit on for it. That happens all the time.

> There's so many blog posts saying look, we're so smart because they're reimplementing something both Linux and macOS had for years

It doesn’t really read like that for me, more of a look at this cool thing we made with an undercurrent of fucking finally, and that seems entirely fair.

    s/years/multiple decades/

I've pointed out flaws to the dev team at MS they were pretty quick about acknowledging it and fixing the problem. There were some showstoppers in the media encoders early on when they were developing win10. The only thing they asked is that I prove the issue was real and as soon as I did they were on top of it.

They were also pretty quick and helpful with answering any question I had. I guess it depends on what team you get ahold of. Or, maybe how you ask the questions. I always try to be polite and patient, which is hard to do when you're in a time crunch.

They really were trying to push azure hard in a situation that didn't require it. It was odd at the time. I chalked it up to the team evangelizing a new feature that they knew was going to rake in money, so they were required to bring it up. I remember it was funny and a little annoying to have to keep saying no... now let's get back on topic.

But, they never insulted me. It was a very professional experience. Sad to hear this kind of thing is happening now.

Reading your post feels good, man. But your last sentence kind of bothers me. Why do you feel they are insulting you now? I think, if at all, someone else was, and kept doing, and still is, insulting them instead.

I'm just trying to understand here, thanks. :)

In the last paragraph I indicated that they never insulted me. The very last sentence is referring to the article and that I'm sad to hear that this is the case now for others. When I personally interacted with Microsoft that wasn't the situation I had at all. I had a good experience.

I think this goes both ways. As a developer at a big company, I've definitely been called dumb or obstinate on issue trackers for not prioritizing certain feature requests, or for not moving forward with a specific design despite explaining why that design is incompatible with other requirements.

Reply on news.ycombinator.com