Was this written because of the guy claiming you should never use greater than (>) symbols?
It’s written in reaction to it. (see this comment)
It is a little cheap, making fun of an article describing an actual problem (and choosing a bit of a weird title) the original article. I mean that original blog post was illustrating how you can mess up comparison operators while checking for values in ranges. And “normalizing” the expressions to only use < and <= operators can help there. Of course how you normalize them is arbitrary, and there might be other solutions (interval objects/representations, etc.).
It’s not just that article. The level of discussion around programming is pretty shallow. Every popular article is about syntax or “community” or Some Weird Trick. We could talk about order theory all day (it’s a topic I am trying to get better at) or we can talk about why syntax means we should write “a < b” because it’s “more simple” or something.
My goal is to bring the level of discourse up by talking about things which really matter, like #noline45.
well, I just think your other writing is doing a better job at this.
but back on topic, what do you say to the fact that curl http://brianmckenna.org/blog/line45 | sed '45!d' gives
curl http://brianmckenna.org/blog/line45 | sed '45!d'
That would make sense.
I can’t really get the humor of this guy’s post if I don’t know what he’s making fun of.
We’re going to change the order of these numbers to make things interesting… 1000001… 66… 1 billion seventy-five thousand…. 1 billion and six something… 0…. 1000001… 22… 75… 11… 11…
Man, my thoughts exactly. I lost my job three times (and got rehired four times; it made for an interesting Thursday afternoon) in the Infamous 45/47 Flamewar of 2014. There are so many fucking philistines out there who can’t tell the difference between Line 45 and Line 47. They’re, like, so obviously different that I don’t know where to begin. Even Louis CK, who by my knowledge is not a programmer, has had choice words about the 45-ers out there.
There’s also no excuse for it. I know that people might occasionally write on Line 45 “by accident”, but that’s what no45 fmt is for. It’s not like programs are written on punchcards anymore.
This is why I like the NumberWang extension of emacs. Sure, it takes some getting used to, with Line 22 coming after Line 3 and all of the complex-number lines, but it’s been mathematically proven that, if you use ESNW (“emacs speaks NumberWang”) properly, you can’t end up with code on Line 45 unless it’s 15:47 UTC, and the obvious answer to this issue is “Don’t fucking write code then”.
Even script kiddies know about Line #45, but as a Real Programmer who’s written source files tens of thousands of lines long, I also need to chime in about the equally confusing #857, #1402, and (I hesitate even to name this last one) #3796. The last time I had to deal with code on Line #3796, I had to crawl into my bike shed and couldn’t come out for weeks!
What’s Peewee Herman up to? We need him to do a PSA about this. Line #45: Not Even Once.
I’ve been a pro-forty-fiver for years, but this post really turned me around. I have like a decade of files to change now, fortunately there’s sed:
sed -i -e '45i\
# This line intentionally left blank' FILENAME
This post by an Atlassian engineer does more to explain why JIRA has such crazy markup than anything else I’ve seen. <g,d&r>
Inspiring. On every line 45 I have I will now write /* this line left blank: http://brianmckenna.org/blog/line45 */. Why not just skip to 46 already? Why even have 45 as a number at all?
/* this line left blank: http://brianmckenna.org/blog/line45 */
The rest of the world has already moved on. It’s about time that the US abandoned the Imperial Numbering Scheme as well.