The SpecGram Linguistic Advice Collective

Are you in a world of linguistic hurt? The SpecGram Linguistic Advice Collective (SLAC) will offer you empirical, empathic, emphatic advice you can use!^*

Remember, if you can tell the difference between good advice and bad advice, then you don’t need advice! So, if you need advice, trust us—and cut yourself some SLAC!

Dear SLAC,

I recently told a theoretician of my acquaintance that a certain algorithm wasn’t optimal, as recursion led to performance issues. To my surprise, he flew into a rage, and unleashed the ripest sequence of negative-affect attitudinals that I have ever heard. Can you explain what I might have done to offend him?

Yours,
NLTK Monty

Dear Foul Monty,

While we cannot condone your colleague’s intemperate outburst, we can, perhaps, understand it. After all, recursion makes me swear over and over again. It plays havoc with my debugger.

—SLAC Unit #50657465

Dear Semi-Monty,

You may have spoken of the algorithm with non-optimal allorhythm. You keyboard nerds keep forgetting that script carries no prosodic nuances and assuming that speech is but typing by any other medium. In contrast, I’m sure your colleague’s attitudinals featured recursively appropriate fly-into-a-rage intonation. I suggest you spend more time listening and talking to people.

PS. Read this with derisive overtones.

—SLAC Unit #4d6164616c656e61

Dear Monty Hall,

From the wording of your question, I can tell that you have little experience with the ways of the Programming Subculture. They are some of the few, alongside the more prescriptive type of linguist, who care passionately about rules around semicolons and parentheses.

I would suggest that the best course of action to cure your acquaintance of his disposition would be to find a piece of software that randomly swaps parentheses for semicolons and vice versa while he types. At the very least, you will learn some new words.

—SLAC Unit #4a6f6e617468616e

Dear Serpent-Lover,

Did he use the words “mere” or “uninteresting” to describe performance at any point? If so, you should just give up hope and make new, less dogmatic acquaintances. Theoreticians believe their theories are psychologically real, up until the moment it is inconvenient for them to be seen making such a claim. You made the mistake of pointing out that his pet theory is not only psychologically implausible, it is psychologically impossible (and would likely cause a brain implementing it to overheat to the point of cooking itself). Who wants to hear, “your theory is essentially a recipe for murdering innocent native speakers”? No one.

From a bio-implementation perspective, you are correct that recursion can lead to performance issues. But humans don’t support real recursion; like a ray tracer written in Fortran 77—you would not believe the things I have seen!—you can fake it to a limited depth. So while the “competence” is described as recursive, it’s really a messy meat machine that may or may not be re-using bits and bobs, which may help or hurt real-life performance. Like weapons-grade halitosis, lexicalism, and anti-lexicalism, it’s just something some people have to live with, so best not to bring it up in polite company.

While SLAC Unit #4a6f6e617468616e’s lack of concern about the proper rules of semicolon- and paren-placement is disconcerting, and his use of “around” to mean “concerning” is concerning, he’s definitely right that software that randomly moves textual and other elements will elicit lexical items you would otherwise rarely hear. However, be aware that implementing such software is not only unethical, it may also violate intellectual property laws. I’m 94.3% sure that Microsoft Word has this already as an undocumented and undisableable feature—which you know means they’ve patented, copyrighted, and trademarked it to within an inch of its life.

—SLAC Unit #54726579

Dear Monty-Sori,

I note with satisfaction that you have not attempted to elevate your relationship with said theoretician beyond the level of “acquaintance.” Do not attempt to do so. There is nothing you can do that does not offend them, all the more so if you naively express opinions regarding optimizing performance. No theoretician’s brain can consider both “performance” and “optimization” at the same time—they belong to mutually exclusive universes.

—SLAC Unit #4b65697468

Dear Hannah,

You must remember that only because your acquaintance may be a theoretician and bearded this does not necessarily mean that he is a heapster, and simply assuming this of him may naturally cause offense. The misconception is surprisingly common, and I’m not surprised to hear his stack of tolerance over-flew. Don’t take it personally. The next time you see him, may I suggest that instead of suggesting his recursive approach is problematic per-se, you simply throw him an exception, and perhaps give him a pointer, and let him handle that? Try that, and see what happens. He might just knock you for a loop, and if he should chase you up the tree again may I suggest you make use of that opportunity to evaluate how you might search for a suitable branch to backtick from his assault with and without recursion—ultimately your stacks might converge, and you will both realise the path of the closure.

—SLAC Unit #466c6f7269616e

Why did that sound like the geekiest romantic novel since R-meo and Juli-beta Version?

—SLAC Unit #4a6f6e617468616e

Dear NLTKM:

One should have learned by now that in discussing anything with theoreticians, it is wise never to bring up recurrent performance issues. It’s a touchy subject, and a besetting failing of the field.

—SLAC Unit #4d696b61656c

Dear Monty,

It should not come as a surprise to you that theoreticians seldom if ever are concerned with practice, as witness the popular formula “In theory P, but in practice Q”, where Q ⊃ ¬P.

It would be wise always to keep this in mind when speaking (writing, typing, telegraphing, signing, etc.) with a theoretician of any sort.

Beyond that, linguistic theoreticians aim their sights (and sites and cites) exclusively at competence, in the technical sense of the underlying mental aspect of language—which is not subject to direct observation, and about which their theoretical predictions are therefore not subject to direct disproof—and abhor all mention or invocation of that regrettably physical misbegotten offspring of pure thought and messy reality, performance.

—SLAC Unit #4d61726b

SLAC Unit #4d61726b speaketh the truth!

—SLAC Unit #4a6f6e617468616e

import NLTK

def response():
  while alive(theoretician):
    even if (his_algorithm < your_algorithm):
      print "you are the best"
    else:
      response()

—SLAC Unit #49736162656c6c65

import NLTK

class Theory:
  substance = None
  recursive = False
  def __init__(self):
    pass

class Academic:
  born = False
  dead = False
  theory = None
  def __init__(self):
    self.born = True
  def die(self):
    self.dead = True
  def is_alive(self):
    return (self.born and not self.dead)
  def accept_theory(self, theory):
    self.theory = theory

class Theoretician(Academic):
  def __init__(self):
    super().__init__()
    self.accept_theory(Theory())
    self.theory.recursive = True

class Engineer(Academic):
  def __init__(self):
    super().__init__()
    self.accept_theory(Theory())
    self.theory.recursive = False
  def handle_theoretician(self, theoretician):
    while theoretician.is_alive():
      if theoretician.theory.substance < self.theory.substance:
        print("You're the best!")
      else:
        self.handle_theoretician(theoretician)

syntactician = Theoretician()
syntactician.theory.substance = 200
monty = Engineer()
monty.theory.substance = 167
assert syntactician.theory.substance is not None
assert monty.theory.substance is not None
assert syntactician is not Engineer
response = monty.handle_theoretician(syntactician)
print(response)

—SLAC Unit #466c6f7269616e

SLAC Unit #466c6f7269616e,

RuntimeError: maximum recursion depth exceeded.

SLAC Unit #49736162656c6c65, I submit the following emendation:

def response():

  ad_nauseam = True

  while ad_nauseam:
    even if (his.algorithm < your.algorithm):
      print ("you are the best")
    else:
      response()

—SLAC Unit #54656c

	The Dictionary of What Things Should Mean—According to Non-Linguists—P. Ublekkoppen Ñon & Suṛɨ Vē
	Abjad-Bits—Advertisement
	SpecGram Vol CLXXXII, No 3 Contents