Hacker News new | past | comments | ask | show | jobs | submit | adamrt's comments login

Can you elaborate? I’ve been using it for over 10 years and it might just be my favorite piece of software. It’s central to all development I do.

I use many features of git that I probably wouldn’t otherwise due to having to remember flags and copying around hashes. It also makes discovering git functionality very easy.


OpenBSD is also known for this. They constantly push back against adding configuration knobs or running non standard configurations.

Have you used OpenBSD? You're telling them they should be doing something, that is already basically their mission statement.


Looking at OpenSSH tells a different story. It is a massive, overly configurable behemoth. The 'WireGuard of SSH' would be 1% of the LOC. It would not provide password auth, or let you log in as root with password auth, or let you use old insecure ciphers.

Maybe OpenBSD itself is better at sticking to these principles than OpenSSH. I haven't used (experimented with) it for ~5 years but read about various updates every so often.


You seem to be confusing "OpenSSH" with "OpenSSH Portable Release". As explained here: https://www.openssh.com/portable.html

> Normal OpenSSH development produces a very small, secure, and easy to maintain version for the OpenBSD project. The OpenSSH Portability Team takes that pure version and adds portability code so that OpenSSH can run on many other operating systems.

Unless you actually run OpenBSD, what you think is "OpenSSH" is in fact "OpenSSH Portable Release". These are very different things.


I can't think of any long term, open source project that has removed and ripped out more code than OpenBSD.

They are know for doing exactly what you are suggesting.

Go ask @tedunangst. It was literally called "tedu'd" for ripping out old crusty code.


> You should target all audiences.

Here is a list of OSes[0]. Where do you draw the line on supporting these? Should every new project try to support all of these? Do you, doublerabbit, get to decide which OSes are important enough for support?

Or do you think the person who created the project and does all the work should be able to decide where to spend their free time?

[0] https://en.wikipedia.org/wiki/List_of_operating_systems


> Where do you draw the line on supporting these?

Where those are still in active development. Where those exist you should attempt at least for. It's partly why they failed in the first place.

> Should every new project try to support all of these?

As said above, attempt. My projects in Perl work most places, my TCL programs do too. C and C++ all have been universes. Heck even Python.

It's only new fangled languages like Rust and Go that make an ball ache.

> Do you, doublerabbit, get to decide which OSes are important enough for support?

Yeah why not, at least allowed to voice an opinion. I'm so sick and tired seeing the world of IT on repeat. See it get abused, capitalised and freedom sucked from it. I give Linux five more years before it will be smothered in corporate.

After working as an sysadmin from the age of 13, to 35. I wish I could call done but other opportunities are not feasible at this time. The amount of bug reports I've submitted across the board is more than a dozen. Hand crafted brittle configuration files, been there done that. This isn't just me being edgy.

For more the past twenty years we've only dominated one bloody OS. Only then do we all bitch at each other because of fanboi or whatever cliche is at the moment. Systemd comes to mind.

I am so bored of the neo-Linux crowd and I've been working with it for it since 2.x kernel.

Only when you jump off the bandwagon do you see how clunky it really is.

First HN was shouting at me how a new browser could never be made and now HN is jumping up and down because one has yet won't acknowledge that other OS exist and that I personally feel developers should catered for.

Is this like to real for everyone or something?


I sense a troll, but

> C and C++ all have been universes. Heck even Python.

> It's only new fangled languages like Rust and Go that make an ball ache.

is not even remotely accurate. The whole idea of the Rust and Go standard libraries is to abstract platform differences away, and in cases where they're unavoidable, to make them impossible to ignore. Python, by comparison, handles them badly. It does certified Bad Things like making POSIX operations silent no-ops on non-POSIX platforms. C doesn't even try. Any cross-platform C program is an #ifdef minefield, and you'll only find out whether it works on a given platform when you try to compile it and start getting obscure library header errors.


Just answering for myself, but it's because I like the programming/engine aspects. Not the creative aspects of the art/music/level-design/rules/balance/etc. I can use an existing games art/content and just focus on the part I like.

Plus nostalgia.


Click on the die a couple times and it speeds up


Thanks, I figured this out accidentally during the weekend while playing.


> But checking out branches I do on the command line

Just in case you have a longer workflow, checking out a branch should just be `b b` and then select the branch from a list with typing or up/down.


And if you use a narrowing framework like consult+vertico+order less you can select the right branch super quick.


I had the same itch for a long time. I eventually took the free Harvard CS50 course. Its the video lectures you can watch on demand. It gets into python at the end which I skipped as I already knew python.

I really enjoyed it and think the teacher is great. I don't think all of it is necessary if you are already a developer, but I didn't have a traditional education so it filled in a few gaps for me too.

https://learning.edx.org/course/course-v1:HarvardX+CS50+X/bl...


My absolute favorite podcast. Pumped every month for the new episode. I think I've gone back and listened to nearly every one.

Thanks so much for doing the show Adam. Some random ideas for guests:

- Joshua Stein (@jcs): Too many projects to list. Rad all-around hacker.

- Mike Bayer (@zzzeek): Creator of SQLAlchemy/Alembic

- Theo de Raadt: OpenBSD creator. Somewhat a controversial guy, but very interesting.


Thanks for the support and the guest recommendations!


They don’t plan ahead for features typically. They add stuff when someone wants to do the work.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: