I'm thinking of getting out of the software business

[blind0ne]

I'm actually paying for "courses" to "write code" in infinite loop. It's looks like I'm in hell of squalor. Hope that the life in only one and won't repeat again.

 

[SirDice]

wasn't it more like powder keg?

The old tool they used to build packages (before PKGNG) was called 'tinderbox' (ports-mgmt/tinderbox). With that in mind the 'poudriere' name suddenly makes a lot of sense as it's the French translation of 'tinderbox'.

 

[astyle]

Has anyone seen the SmartyPants port (textproc/py-smartypants)? or even this thread:

Schedule driven GPIO events

I want to use GPIO outputs to drive an IOT Power Strip. https://www.sparkfun.com/products/14236 My question is about triggering scheduled events. For instance I want to issue a GPIO command to turn a pin on and off on a schedule. So I could use cron but would like to avoid system utilities...

forums.freebsd.org

blind0ne : Stop paying for Ruby courses, and go play with lex/YACC. It does exactly the same things as Ruby, uses the same logic, BTW Most of those "Hot new languages" are really re-inventing the wheel anyway.

 

[SirDice]

I wouldn't call Ruby "hot and new", it's almost 30 years old now. Ruby-on-Rails is quite interesting, many modern web frameworks actually mimic lots of the features that were first introduced with RoR.

 

[astyle]

wasn't it more like powder keg?

Anyways - both are rather descriptive as I have WAY more builds blowing up when trying to use poudriere than just manually running "make" (or using portmaster) inside a build jail...

But yes, I also struggled to remember that name in the beginning... some other variant of "port<something>" would be better IMHO

My builds are not exactly "blowing up", but I'm still stuck on getting Poudriere's updates to be visible to poudriere bulk ? I'm on the verge of giving up because of that.

 

[6502]

Now I understand that Ruby and Ruby on Rails are not the same. Ruby on Rails is my favorite for strange names.

 

[SirDice]

Ruby on Rails is my favorite for strange names.

Maybe you want to have a look at Brainfuck. Yes, that's an actual programming language.

 

[drhowarddrfine]

Maybe you want to have a look at Brainfuck.

The very definition of low-brow programming. There was no need to call it that.

 

[mer]

Nowadays you still see things that make sense like npm.

I think this is the nicest thing anyone has said about npm.

 

[bakul]

You guys may be interested in reading this page: https://justine.lol/sectorlisp2/ (Lisp in 436 bytes, brainfuck in 99 bytes).

 

[astyle]

The very definition of low-brow programming. There was no need to call it that.

What would you call something that is Turing-complete, and yet such a Turing Tar pit?

Brainfuck is an example of a so-called Turing tarpit: It can be used to write any program, but it is not practical to do so, because Brainfuck provides so little abstraction that the programs get very long or complicated.

Edit: Poudriere is slowly becoming a brainfuck for me... ?

 

[drhowarddrfine]

What would you call something that is Turing-complete, and yet such a Turing Tar pit?

Not what it's called.

 

[astyle]

Not what it's called.

You made it abundantly clear you don't like the name, drhowarddrfine ... do you have a constructive alternative name in mind?

 

[SirDice]

Well, trying to read that ROT13 example on the wikipedia page almost gave me a brain aneurysm. So I think the name suits it just fine.

 

[bakul]

The very definition of low-brow programming. There was no need to call it that.

It is actually more aptly named than "Hot Chocolate" for a GraphSQL server!

From its wikipedia page: "The language's name is a reference to the slang term brainfuck, which refers to things so complicated or unusual that they exceed the limits of one's understanding."
From Urban Dictionary: "1. Verb: To fuck with the mind/head, to toy with one's thoughts of reality; to cause a state of confusion and befuddlement."

 

[ralphbsz]

The very definition of low-brow programming. There was no need to call it that.

How about Intercal? It's an abbreviation of "Compiled language with no pronounceable acronym". The important thing is: Both brainfuck and intercal are jokes, not meant to be actually used.

Speaking of things that are used: A few people above lauded the logical naming of programs in the Unix and *BSD tradition. It's actually less logical than you think, and contains lots of jokes. rm makes sense, abbreviation for remove. ls is an abbreviation for list, as is cp (copy), ln (link), mv (move), ps (process status). touch, test, uniq, sort, who and echo are intuitive. But cat is already inside joke: It's named after a Lisp keyword, which in turn is named after the assembly instruction on some ancient machine. awk is not named for its functionality, but for its three authors (Aho, Weinberger, and Kernighan). dd is a joke about a statement from IBM's job control language; the way parameters for dd are stated (which is massively not unix-like) is directly taken from JCL. grep is not named for its functionality, but for the equivalent editor command that it replaces. So inside jokes have a long tradition in software engineering.

Do I like that tradition? Personally, no. I would prefer to have logical names: delete instead of rm, search instead of grep, processtext instead of awk, rawcopy instead of dd, and so on. But I don't make the rules around here.

 

[mer]

Logical names for commands/programs are good, makes it easier to figure out what you need.
But even the traditional names in Unix and *BSD are easier to remember than "which tab in which GUI thing do I need to look on for this data".

 

[hruodr]

Do I like that tradition? Personally, no. I would prefer to have logical names: delete instead of rm, search instead of grep, processtext instead of awk, rawcopy instead of dd, and so on. But I don't make the rules around here.

That names are too long. Unix two letter names have a sense.

 

[bakul]

But cat is already inside joke: It's named after a Lisp keyword, which in turn is named after the assembly instruction on some ancient machine.

cat() comes from concatenate, not from Lisp's car and cdr functions, which are named after IBM 704 assembler macros.

 

[kpedersen]

Do I like that tradition? Personally, no. I would prefer to have logical names: delete instead of rm, search instead of grep, processtext instead of awk, rawcopy instead of dd, and so on. But I don't make the rules around here.

These longer names would cause some serious slowdown to those proficient with the command line (they would all need to be aliased basically).

Also the following to me have a couple of issues:

• search - would this then need to merge the find & grep tool? Both effectively "search" things.
• processtext - Awk can do a little bit more than this. Would Python be renamed to extensibleprocesstext?

In some ways I would rather a more abstract name (i.e floopy, poofle, dudu) rather than a name that conflicts with other tasks.

I kind of feel that Powershell is what you get if you go with long logical names. I.e Get-Childitem –Path C:\ -Recurse. It just end up being a pain to use in practice. It would be very interesting though to present it to a complete beginner and see which they would choose after a couple of weeks.

At first I thought it was funny, how awkward Microsoft had made powershell. But I later realized it can't be an easy task to design a naming system of an entire CLI platform. They obviously just went with "Oh, lets make it easy; this will get us money". But in fact it ended up showing off just how impressive a job the UNIX forefathers managed as an almost first attempt.

 

[drhowarddrfine]

do you have a constructive alternative name in mind?

No. It's never on my mind.

 

[bakul]

How about Intercal? It's an abbreviation of "Compiled language with no pronounceable acronym".

From Don Wood (one of the creators of INTERCAL): "I think we actually started with the name INTERCAL. I'm not sure where it came from; probably it just sounded good. (Sort of like FORTRAN is short for "Formula Translation", INTERCAL sounds like it should be short for something like "Interblah Calculation"). I don't remember any more specific etymology. Then when we wanted to come up with an acronym, one of us thought of the paradoxical "Compiler Language With No Pronounceable Acronym."

Years ago I worked with Jim Lyon, the other creator of INTERCAL, for about 2 years though lost track of him since then. An excellent programmer.

 

[astyle]

How about Intercal? It's an abbreviation of "Compiled language with no pronounceable acronym". The important thing is: Both brainfuck and intercal are jokes, not meant to be actually used.

Speaking of things that are used: A few people above lauded the logical naming of programs in the Unix and *BSD tradition. It's actually less logical than you think, and contains lots of jokes. rm makes sense, abbreviation for remove. ls is an abbreviation for list, as is cp (copy), ln (link), mv (move), ps (process status). touch, test, uniq, sort, who and echo are intuitive. But cat is already inside joke: It's named after a Lisp keyword, which in turn is named after the assembly instruction on some ancient machine. awk is not named for its functionality, but for its three authors (Aho, Weinberger, and Kernighan). dd is a joke about a statement from IBM's job control language; the way parameters for dd are stated (which is massively not unix-like) is directly taken from JCL. grep is not named for its functionality, but for the equivalent editor command that it replaces. So inside jokes have a long tradition in software engineering.

Do I like that tradition? Personally, no. I would prefer to have logical names: delete instead of rm, search instead of grep, processtext instead of awk, rawcopy instead of dd, and so on. But I don't make the rules around here.

There's also /bin/kill - 9 <pid-of-/bin/cat /usr/libexex/more.a>

 

[ralphbsz]

cat() comes from concatenate, not from Lisp's car and cdr functions, which are named after IBM 704 assembler macros.

Sorry about that ... confused.

 

[ralphbsz]

That names are too long. Unix two letter names have a sense.

Which is why good operating system shells allow entering just the part of the name that is enough to be unambiguous, without having to tab complete. And then the people who design the shell carefully think through which commands are commonly used, and give them names are unambiguous after very few characters.

Ever run DCL on a VMS machine?