When discussing language syntax, people often resort to small examples using simple variables like foo or x, almost like “meta-syntactic variables”, i.e., to make clear these tokens are outside of the syntax under discussion.

One dangerous side-effect, though, is that these variables are always short and sweet. And syntax that works well with short variables doesn’t always work as well in real-world situations where they have to deal with the rest of the language.

Case the first

Recently we were discussing multiple assignment style in the Lua mailing list. Someone suggested this:

local a, b, c, d =
      e, f, g, h

…which makes the assignments “more parallel” than a single line and avoid writing lots of locals.
I think this a case where the over-simplified example is misleading.
With real-world looking variables, it would look more like

local cfg,              constraints, module_name,          initial_path =
      "default_config", {},          get_module_name(ctx), "/etc/myapp/default.config"

So yeah, It looks pretty with a, b, c but in the real world with significant names, this becomes a pain to maintain, and when we stuff too much in a single line, diffs are harder to read.

Case the second

Things always look good in tiny examples with single-letter variables. Which brings me to a gripe I have with an often-suggested Lua idiom: the famous t[#t+1] = v to append to arrays.

The reason why I think it’s so disengenious to defend t[#t+1] = v as the preferred idiom for appending to an array is because it looks good with a single-letter variable and five-line tutorial examples, but in the real world we use nested tables. In the end, table.insert(my.nested[data], v) is both more readable and avoids repetition:

• it’s one thing less to have a typo on
• one thing less to change
• and one thing less to get out of sync.

Note how it’s not even necessarily shorter: in this realistic example the variable name dominates the size of the statement:

table.insert(my.nested[data], val)
my.nested[data][#my.nested[data] + 1] = val

Do I think table.insert is too long? Yes I do, I wouldn’t mind having a shorter idiom (many were proposed in the Lua list over the years, most of them were fine, but I’m not getting into them because we risk delving into syntactic bikeshedding again, so let’s avoid that).

Do I think it’s worth it to add local tinsert = table.insert to every program? No, I think this is worse than the t[#t+1] = v idiom, because I hate having to guess which abbreviation the module author used to write a shorter table.insert in their code (I’ve even seen local append = table.insert in the wild!). And then again, the abbreviation doesn’t gain us much: being comfortable to read is more important than being comfortable to write, but being easy to maintain is just as important if not more.

And yes, it is important to ponder what are the differences between being “easy to read”, “easy to write” and “easy to maintain”. And when pondering those, watch out for misleading short variables in the examples!

Of course, some idioms are advisable specifically for when you have short variables:

local r, g, b = 0, 255, 0

Everyone can easily read what’s going on there. But note that, almost without noticing, I also used a realistic example here! Realistic examples help getting the discussion grounded, and I find that they are often lacking when discussing syntax.

PS: And before someone mentions, the performance gains for localizing such variables as local tinsert = table.insert are overstated:

• When done at the top of modules they become upvalues and not true locals;
• Most of the modules I’ve seen doing this are far from being aimed at performance-intensive tasks that would warrant this kind of micro-optimization;
• Optimization advice changes between Lua implementations; while the cached local helps for interpreted Lua, it may actually hurt for LuaJIT. So just aim for clearer code. If you have an optimization problem you can measure and the local variable does bring a benefit, do it in a small scope, close to your performance-sensitive tight loop so that whoever is reading your code can understand what is going on.

I just remembered an interesting tidbit from my childhood.

Around 7th grade in school I invented a pen-and-paper version of Street Fighter II for people to play during classes.

I don’t remember the exact details, but basically I drew a grid for the screen and then I drew stick figures in it, and passed the page around.

People would write-in their moves and then I played CPU: I’d erase the stick figures and redraw in new positions, update hit/miss, update the energy meters.

I remember trying to keep it balanced and true to the game: Dhalsim’s punch and kick could hit farther but were weaker, etc. I had all of the “sprites” with the character movements pre-determined on my notebook.

The game went on sneaking a page back and forth along players and me at the back of the class. I imagine how bored out of our minds we must have been in school to enjoy playing “Street Fighter II at 0.05 frames per second”.

Prologue

This is anecdote about roundabout ways to get stuff done. Pierre mentioned in the comments below that a proper way to solve this is to use unbuffer (though it does _not_ produce the exact same order as the terminal!). But if you want to read the improper way to do this, read on! :)

The story

Due to buffering, the terminal messes with the order of stdout and stderr of a program when redirecting to a file or another program. It prints the outputs of both descriptors in correct order relative to each other when printing straight to the terminal:

] ./my_program
stdout line 1
stdout line 2
stderr line 1
stdout line 3
stderr line 2
stderr line 3

This doesn’t change the order:

] ./my_program 2>&1
stdout line 1
stdout line 2
stderr line 1
stdout line 3
stderr line 2
stderr line 3

but it changes the order when saving to a file or redirecting to any program:

] ./my_program 2>&1 | cat
stderr line 1
stderr line 2
stderr line 3
stdout line 1
stdout line 2
stdout line 3

This behavior is the same in three shells I tested (bash, zsh, dash).

A weird “solution”

I wanted to save the log while preserving the order of events. So I ended up with this evil hack:

] strace -ewrite -o trace.txt -s 2048 ./my_program; sed 's,^[^"]*"\(.*\)"[^"]*$,\1,g;s,\\n,,g;' trace.txt > mytrace.txt
] cat mytrace.txt
stdout line 1
stdout line 2
stderr line 1
stdout line 3
stderr line 2
stderr line 3
+++ exited with 0 +++

It turns out that strace does log each write in the correct order, so I’m catching the write syscall.

Note the limitations: it truncates lines to 2048 characters (good enough for my logs) and I was simply cutting off n and not cleaning up any other escape characters. But it worked well enough so I could read my ordered logs in a text editor!

One of my least-favorite American English words is “latino”, for two reasons:

First, a linguistic reason: because it’s not inflected when used. When you’re used to the fact that in Spanish and Portuguese “latino” refers only to men and “latina” only to women, hearing “latino woman” sounds really weird (weirder than, say, “handsome woman”). Even weirder “latino women”, mixing a Spanish/Portuguese word and English grammar. “Bonito girls”? :)

Second, a sociological reason: because using a foreign loanword reinforces the otherness. Nobody calls the Italian community in America “italiano”, although that’s their name in Italian. The alternative “Hispanic” is not ideal because it doesn’t really make sense when including Brazil, which was never a Spanish colony (plus, the colonial past is something most countries want to leave behind).

I can’t change the language by myself, so I just avoid the term and use more specific ones whenever possible (Colombians, Argentines, Brazilians, South Americans, Latin Americans when referring to people from the area in general, etc.)

After writing the above, I checked Wikipedia and it seems the communites in the US agree with me:

« In a recent study, most Spanish-speakers of Spanish or Hispanic American descent do not prefer the term “Hispanic” or “Latino” when it comes to describing their identity. Instead, they prefer to be identified by their country of origin. When asked if they have a preference for either being identified as “Hispanic” or “Latino,” the Pew study finds that “half (51%) say they have no preference for either term.”[43] A majority (51%) say they most often identify themselves by their family’s country of origin, while 24% say they prefer a pan-ethnic label such as Hispanic or Latino. Among those 24% who have a preference for a pan-ethnic label, “‘Hispanic’ is preferred over ‘Latino’ by more than a two-to-one margin—33% versus 14%.” Twenty-one percent prefer to be referred to simply as “Americans.” »

I think the awkwardness in the grammar from point one actually reinforces point two, because it strikes me as something that no Spanish or Portuguese native speaker would come up with by themselves. So it sounds tacked upon.

Don’t get me wrong, I fully identify as a Brazilian, a South American and a Latin American — travellling abroad helps a lot to widen your cultural identity! — and I have no problem when people wear the term “latino” proudly, but I always pay close attention to the power of language and how it represents and propagates ideas.

I decided to stop running XSane as root. Here’s what I had to do:

* Restored the ownership of ~/.sane back to my user: sudo chown -R hisham ~/.sane
* Made /var/lock/sane writable by a regular user: sudo chmod a+rwxt /var/lock/sane
* As a quick-and-dirty fix, as suggested here, I gave permissions to my USB devices: sudo chmod a+rw -R /dev/bus/usb/*
* For a more permanent (and still untested) fix, I added a 40-libsane.rules file (found here) in /etc/udev/rules.d. This file already includes the USB id for my scanner. You may have to add yours.