• Oliver Jones (google)

    Back in college I worked for a prof who built a machine controller from a PDP8 computer. He had the code, in octal, on little sheets of paper, stuck to a 8x4 foot hunk of plywood next to the machine.

    When the machine needed to do something different, he patched it using the front-panel switches, and scribbled something on the appropriate sheet of paper. If it was an important change, he'd get the machine to punch its program on paper tape.

    Yeh, programming with symbols like variable names is definitely for woo woo new agers. That guy was a man's man.

  • snoofle (unregistered)

    Variable names are just aliases for memory locations. Half the time they're named incorrectly anyway. What's the big deal?

  • What? I'm not giving you my name. (unregistered)

    Couldn't agree more with Octavio. I've coded using nothing more than a hexadecimal keyboard and a 8 digit display compiling my code with a pencil and a piece a paper like a real programmer. Naming variables/OOP/Virtual machines/microservices/... are for losers that like slow software.

  • Quite (unregistered)

    I am reminded of the story (probably on this site) of the secretary who was responsible for maintaining an extensive library of documentation in Word (or one of its predecessors / competitors). She was not a huge fan of renaming. Hence all the documents were called Document1, Document2, ..., Document86, Document87, ..., Document4968, Document4969, ... and so on.

    How did she know which document contained what information? Oh, that was easy, she had a little exercise book with a list of all the documents and their description and title. Worked perfectly.

    Until she left, and took the notebook with her.

  • El Dorko (unregistered)

    You kids and your modern tools, y'all are spoiled brats is all! When I was a kid I would program a cardboard box where my sisters was in, executing the programs and tapping the output as morse code using nothing but her fingers... We didn't even have all the numbers, they only went up to seven - on a good day!

  • my name is missing (unregistered)

    When I started in the 80's there were two old guys who talked about when they started in the 50's and they programmed computers with wires. Numbers are for wimps, real programmers use color coded wires.

  • Vault_Dweller (unregistered)

    Real programmers use a magnetized needle and a steady hand

    (Shamelessly stolen from xkcd)

  • Hasseman (unregistered)

    https://xkcd.com/378/

  • No Fun (unregistered)

    I wonder if "JavaBachata" is an anonymisation of JavaTango? (I hope this doesn't count as doxxing)

  • MiserableOldGit (unregistered)

    Well, it's medium insanity, I'd rather have that an implemented properly than the inconsistent broken mess of misleading misspelled and downright stupid variable and function names in front of me right now.

    To be honest though, I wouldn't want either.

    “I was doing no SQL before NoSQL was a thing,”

    Well I was done at that point because, unless the guy has been in the game longer than I've been alive, he doesn't know his own professed area of expertise.

  • Uno (unregistered) in reply to MiserableOldGit

    My favorite NoSQL system is NTFS.

  • (nodebb)

    I'm very amused at the Java developer's "shaking hands". Too much Java will do that.

  • (nodebb) in reply to snoofle

    Half the time they're named incorrectly anyway.

    Work with a severely dyslexic programmer and you'll find out just how true that is. (Nice guy and can actually code really quite well, but his code is often a bit painful for me to read because of the speelign eroors.)

  • ZZartin (unregistered) in reply to snoofle

    Names might not be useful always but using generic 1...whatever numbers is guaranteed to never be useful.

  • TheCPUWizard (unregistered) in reply to Oliver Jones

    Ahhh, the PDP-8, my all time favorite. Still have one [along with a SBC-6120 and a few emulators]. Biggest problem is I can not find a teletype repairman anywhere [I have spares and maintenance documents]

  • Nobody (unregistered) in reply to El Dorko

    Now, if you tell kids that today...they won't believe you!

  • Appalled (unregistered)

    You can't have self-commenting code with crap for names. I would assert that this is why Assemblers/Compilers were first invented, to relieve the tedium of chasing (writing) down Register/Displacement addresses. (I still remember the days of IBM Assembler and being pissed off that we only had 8 characters for labels and fields.) But these JavaBeanster banana heads know better than me, so sure, let's re-rig our current generation of languages to jump over Assembler/Compilers and fall all the way back to Machine Code. Sounds like a plan.

  • Kurbein (unregistered)

    https://en.wikipedia.org/wiki/Turing_machine

    Now THIS is real programming. Octavio probably solved the Halting problem before Turing.

  • (nodebb)

    It's soooo enterprisey. It just needs XML to make it even better.

  • MiserableOldGit (unregistered) in reply to Uno

    That's the worst ever ... except for all the others.

  • Lerch98 (unregistered)

    If you can't name a variable, you don't know what you are doing. very simple.

    Place I worked before, I was working on project and I asked one of the dev-hacks for a memory map. It looked something like this int x int xx int xxx[]

    WTF ^2

  • (nodebb)

    He really is a start programmer. He's already got one of the two hard things sorted; all he needs now is to get cache invalidation worked out and he shall truly be like unto a god.

    Addendum 2017-11-28 22:11: Star programmer, of course. Thought "start" is probably unintentionally accurate.

  • (nodebb) in reply to Appalled

    Yes, the scribble book industry wants us to learn to write down what all those memory addresses mean.

  • (nodebb) in reply to slavdude

    XML is sooooo 2000s. I heard someone invented "initialization files" that are far more human readable and at the same time easier computer parsable than XML. The naming convention for those files is, "[application name].ini", and variables and their settings are delimited by "\r\n" and each variable and its content is represented like this:

    [variable_name]=[contents]

  • (nodebb)

    Remy, where did you put the other 197 Easy Reader Versions in the HTML comments?

    Oh wait, I made just another OB1 error.

  • (nodebb)

    "and my customers have been very happy.”

    With emphasis on have been, I suppose?

  • Mike Shawaluk (google)

    What? No "Brillant OttavioBean" mention?

    Addendum 2017-11-29 11:38: (um, OctavioBean, I can't spell)

  • I Am A Robot (unregistered) in reply to MiserableOldGit

    "“I was doing no SQL before NoSQL was a thing,”" "Well I was done at that point because, unless the guy has been in the game longer than I've been alive, he doesn't know his own professed area of expertise."

    Could just mean that for him, like a lot of people, he was using a database different to "SQL" before the NoSQL term became popular.

  • Zenith (unregistered)

    I would've been tempted to walk out the second I heard the organization had a framework. Every one of those I've been saddled with has been so poorly designed as to be practically useless.

  • not me (unregistered) in reply to P. Wolff

    Do you normally go to the HTML source hunting for Easter eggs?

  • (nodebb) in reply to not me

    For Remy's articles? Absolutely. Some of us even use Greasemonkey userscripts (such as https://github.com/gcochard/tdwtf-uncommenter) to make the comments visible on the page.

  • MiserableOldGit (unregistered) in reply to I Am A Robot
    Could just mean that for him, like a lot of people, he was using a database different to "SQL" before the NoSQL term became popular.

    Well yes, that would apply to me too, although I'd say all the people I know in that category start spitting profanities/laughing/stomp off in disgust when faced with know-it-all semi-techs banging on about noSQL. They tend to be people who couldn't get the hang of normalisation or SQL and so start advocating shoving any old trash data in a bucket and fishing results out with a bent stick, then have to dream up ways to tell the client why all the unstructured, inconsistent, unreliable slow to retrieve garbage is a good thing.

    There are, of course, people who know what these tools are for and how and when to use them effectively, but the coding described in this WTF would suggest this guy is not one of them.

  • Zork (unregistered) in reply to Quite

    I remember this too. It was posted on Rinkwork's Computer Stupidities.

  • cowbert (unregistered) in reply to P. Wolff

    XML is sooooo 2000s. I heard someone invented "initialization files" that are far more human readable and at the same time easier computer parsable than XML. The naming convention for those files is, "[application name].ini", and variables and their settings are delimited by "\r\n" and each variable and its content is represented like this:

    [variable_name]=[contents]

    I wrote a data integrator using .ini files for configuration, since it's a lot easier for the end user it's intended to be used by to edit an .ini file than YAML, JSON or XML (otherwise they would need a GUI to write the config rather than just a CLI tool). The integrator is written in Python, which ships with the configparser class in the standard library that specifically parses .ini files for you (it even supports %()s style interpolation for power users!)

Leave a comment on “A Handful of Beans”

Log In or post as a guest

Replying to comment #490709:

« Return to Article