« first day (3876 days earlier)      last day (1054 days later) » 

9:13 AM
What's a good choice for blocks of text in an article type class? Say an address. So something spread acros several lines, but not a regular paragraph.
Would an mbox or similar be a good choice?
 
@FaheemMitha I use every opportunity to shove something into a tcolorbox, but that might not be what you're looking for :p
 
@FaheemMitha how will it relate to the rest of text? Can be broken across pages? On a paragraph by itself embedded in a line? You can use a tcolorbox, or a tabular, or a minipage...
@FaheemMitha mbox is for horizontal material...
Travis working again! Let's see what happens with the allotted time... 'cause I am not sure about it, will the 10000 minutes be refilled every month?
 
9:33 AM
@FaheemMitha mbox is single line so not really a "block" tabular is what article and letter use for addresses
 
@Rmano Probably not broken across pages. I'm not sure what you mean by "on a paragraph by itself embedded in a line".
@Plergux Yes, I've used tcolorbox before. It's actually quite good. But maybe overkill here.
At any rate, it has quite a lot of options and functionality.
 
:58306339 Like for example:
                       box box box
text text text    box box box  more text
                       box box box
@FaheemMitha ^^^^^
But if it's like a simpler tcolorbox not to be broken between page a minipage will do.
 
@Rmano Not alongside other text, no.
@Rmano OK. minipage noted.
 
I'm starting a new project. If anybody is interested or would like to contribute: github.com/samcarter/PackageOfTheDay
 
@JosephWright is it possible to use a \PassOptionToClass to siunitix to have the same behavior as calling it with [=v2]? To quick fix a documentclass that breaks with the new version...
 
9:47 AM
@samcarter_looks_forward_TUG'21 Typo alert: As the ducks of the year soon celebrate their first birtday
birtday -> birthday
 
@FaheemMitha yeah, if you're just going to "sneakily" stick a box into text then tcolorbox is probably not the way to go. that's more for saying "oooh! hey! look at this content!" :
 
@FaheemMitha fixed, thanks!
 
@Plergux It might work, though. I'll take another look. So minipage or tabular are the less "exotic" options?
 
@FaheemMitha well, yes, they have various options for formatting so you might create one without background or border :p
 
10:01 AM
@samcarter_looks_forward_TUG'21 Also, it's publicly rather than publically.
 
@FaheemMitha thx! (note to myself: find markdown editor with spell checker)
 
@samcarter_looks_forward_TUG'21 You could use Emacs.
 
@FaheemMitha I don't want to risk @PauloCereda stop speaking to me :)
 
@samcarter_looks_forward_TUG'21 Ooh
 
10:38 AM
@Rmano I'm afraid not - it's actually not an option passed to siunitx, it's an earlier kernel-based swap. I'll raise the question with the team
 
@JosephWright basically he has to use siunitx-xyz.sty? Can't he use the replacement hooks?
 
@UlrikeFischer Quite possibly, but my point was that as this look like a package option, the 'normal' advice of \PassOptionsToPackage looks reasonable, and we should have a general idea about how to handle it
 
@JosephWright ok --- I solved it the right way (complain to the class author and just edit the .cls) ;-)
2
@JosephWright Basically the main problem was that the author is loading both siunitx and units so that siunitx complains on redefining \unit
 
11:01 AM
@Rmano Well that's been an error since v0.6 ...
 
@samcarter_looks_forward_TUG'21 why not?
@samcarter_looks_forward_TUG'21 TaBlE just hit ctan:-)
 
@DavidCarlisle I like to practise my duck language skills and it is difficult to find them in Europe after you had dinner :)
@DavidCarlisle A school friend of mine had a cat with this name :)
 
@DavidCarlisle Only taken ... er ... 25 years?
 
@DavidCarlisle Team?
 
@JosephWright ...but it works if I add [=v2]. Don't know why...
 
11:09 AM
@Rmano That is odd - there's a test that should block loading both
 
@AlanMunn @DavidCarlisle I did see the messages for memoir, haven't got any time this week or weekend. I'll look at it next week.
I just came by a PCTeX5 user. I'm wondering exactly how old his latex installation is of LaTeX says that the format is from 2001.
 
The code in the class is
 
@daleif Oh my
@daleif He should be getting the 5-year-old-format warning!
 
\RequirePackage[tight,nice]{units}   % Simple units
\RequirePackage{siunitx}[=v2]
 
@JosephWright are you stealing @PauloCereda 's line?
 
11:12 AM
@JosephWright 4 of them... ;-)
 
@Rmano Oh, right, I see - hmm, I need to pick that up
 
@JosephWright well, were there checks for that in the format bask then
 
@daleif I'd have to check the SVN I guess
 
@JosephWright don't bother. It seems to be working ok for him
 
@UlrikeFischer sorry was in work call, will join
 
11:25 AM
@JosephWright the other problems where quite trivial, an output-product that I simply removed and a \renewcommand{\SIUnitSymbolMicro} (same way)
 
yo'
Just used tweet as a unit of length of text. The world is doomed.
 
@yo' how many tweets is the TeXbook?
 
yo'
@samcarter_looks_forward_TUG'21 you have to ask @DavidCarlisle, it's in his emacs, not mine ;)
 
@Rmano OK, what's happened is I forgot that units could be loaded with v2 as I didn't define \unit, and I've changed that. That's the same as physics: I guess I need to provide a slightly better mechanism here for picking the interfaces
 
@yo' Talking about units, I recommend section 10 of mirrors.ctan.org/macros/latex/contrib/cooking-units/…
2
 
@JosephWright commented. What was output-product supposed to do?
...maybe a list of removed keys and what to do would be nice in the manual... for the lazy developers of other packages ;-)
 
yo'
@samcarter_looks_forward_TUG'21 ouch
 
@Rmano It was used when you have \num{1 x 2 x 3} as the times symbol: I've adjusted so you need \numproduct, and it's called product-symbol
@Rmano Fro programmers, there is a long list in the CHANGELOG
 
@JosephWright will have a look, thanks
 
11:55 AM
Why does a clear question with multiple nice answers gets closed just because there is no reaction from the OP? tex.stackexchange.com/questions/524621/…
Isn't the time and effort the answers have spend on this worth keeping this information around for future users?
 
@JosephWright the file is dated 1988 so 25 is rather a low guess:-)
 
@DavidCarlisle :)
 
(also it is not true that there was no reaction from the OP, although a non-answer is not the best location ...)
 
12:27 PM
tcolorbox seems to be quite popular. Is it the most popular "box" package?
 
@FaheemMitha it is at least one of the most versatile.
 
@Skillmon What is the competition?
 
@FaheemMitha the other box packages?! :) mdframed comes to mind, for example.
@FaheemMitha take a look at the CTAN list of boxing packages: ctan.org/topic/boxing
@FaheemMitha though those also include the programmatically aspect of boxing (xsavebox etc.), so maybe the list of decoration packages is a better place: ctan.org/topic/decoration
 
That's quite a list. I didn't realise there were so many.
I remember using mdframed.
@samcarter_looks_forward_TUG'21 Also gurantee -> guarantee
 
@FaheemMitha thanks (again) :)
 
12:39 PM
Sorry that I don't see these at the same time.
Also
> (package which encourage pineapple pizzas won't be included)
should probably be plural. I.e. packages.
 
@FaheemMitha If by popular you mean usage, then probably graphicx with \fbox
@FaheemMitha fixed! (thanks)
 
@samcarter_looks_forward_TUG'21 Most widely used, yes.
 
1:01 PM
@samcarter_looks_forward_TUG'21 you have to trust the author
 
@DavidCarlisle Wonder if the author is trustworthy ...
 
1:18 PM
Ok, I managed to have Travis back on circuitikz. They allotted me a number of "OSS only minutes"... I really do not know what changed lately, but it seems that they are becoming more vigilant...
 
@Rmano Same here with siunitx
@Rmano Someone commented to me that if they want us all to move to GitHub Actions, they are going the right way about it
2
 
@JosephWright I just created my quack repo to explore it... ;-)
 
@JosephWright oh my
ooh
@samcarter_looks_forward_TUG'21 ooh
@FaheemMitha ooh
 
I managed to wipe out all my gh-pages content in one click. Fortunately, it's git...
 
@Rmano oh no
@Rmano quack
 
1:23 PM
@PauloCereda Quack!
 
@samcarter_looks_forward_TUG'21 ooh quack
 
@PauloCereda github.com/Rmano/quack point 6 wiped out my gh-pages branch. It's documented (it's a flag) but quite surprising...
 
@PauloCereda counter-ooh quack!
 
@Skillmon <3
@Rmano indeed!
 
@PauloCereda <3
@PauloCereda reminds me of a package project I started a while back...
 
1:28 PM
@Skillmon rabbits are good at writing packages :)
 
@PauloCereda currently looking at the github repository. I have no memory of the stuff that's going on :)
 
2:00 PM
@JosephWright \show? Not even \tex_show:D? :-)
 
 
1 hour later…
3:09 PM
@JosephWright texfaq.org
@StefanKottwitz ^
 
@DavidCarlisle very good
 
@StefanKottwitz github kicked it for me. I think it got confused by the existing but expired certificate at the old dns settings so its automated letsencrypt provision hangs, but github support confirmed the dns was all correct and just replied in less than a day to say the certificate was issued.
 
@DavidCarlisle that was fastly
 
@StefanKottwitz terrible pun
 
@DavidCarlisle but a pun :-D
@DavidCarlisle nowadays I ask my customers, if I should do it fastly or slowly.
 
3:38 PM
Hello everyone,

May I know why I don't get the expected output of 5.7 x 10-4 in the table?
Additionally, how to disbale the alignment with respect to decimal points and flush right the numbers to the table right margin?

\documentclass[12pt]{article}

\usepackage{siunitx,xltabular,multicol,booktabs,ragged2e}
\usepackage[table]{xcolor}

\newcolumntype{Y}[1]{>{\Centering\hsize=#1\hsize\linewidth=\hsize}X} % e.g. Y{1.5}
\newcolumntype{L}{>{\RaggedRight}X}

\newcommand{\heading}[1]{\multicolumn{2}{c}{#1}}
 
3:57 PM
@Diaa well you are saying table-format = 4.3 and so siunitx formats it this way.
 
@UlrikeFischer A hyperref question: what's the best solution to "destination with the same name" warning? My university requires per chapter appendix numbers to restart every chapter which leads to these warnings, but I'd like to eliminate them at the class level if possible.
 
@AlanMunn providing an alternative name. So if e.g. \thesection is not unique across the document, define \theHsection to be e.g. blub.\thepart.\thechapter.\thesection.somesuffix
 
@DavidCarlisle Yay
 
4:13 PM
@JosephWright so I enabled force https so all four of http[s][www.]texfaq.org end up at texfaq.org (or we could make them wll go to www version if we prefer.
 
@UlrikeFischer I just realized I overlooked it. However, how to parse the numbers but disable the alignment with the decimal point?
 
@Diaa don't use an S column if you don't want alignment?
 
@DavidCarlisle Good
 
@Diaa table-alignment-mode=none,table-number-alignment = right,?
 
Thanks, I almost there.

S[
table-format = 4.3e-1,
parse-numbers = true,
table-alignment-mode = none,
table-number-alignment = right,
exponent-mode = engineering, % scientific engineering input
exponent-product = \times,
%exponent-base = 10,
%output-exponent-marker = 10,
round-mode = none,
tight-spacing = true,
]
gets me this
but I just need x 10-4 to show up when I input e-4 explicitly
 
4:25 PM
@Diaa then why do you still use table-format?
 
@UlrikeFischer It is just a six month old code I am trying to make it work again with version 3. In version 2, table-format = 4.3 was enough to give me the expected output of printing the exponent when I write e-4 and nothing when I don't.
@DavidCarlisle but S columns are helpful in parsing the numbers and printing exponents
I just found out the following gets me what I want

S[
table-format = 4.3e-1,
parse-numbers = true,
table-alignment-mode = none,
table-number-alignment = right,
exponent-mode = input,
exponent-product = \times,
round-mode = none,
tight-spacing = true,
]
 
@JosephWright and the NFC normalised form for U+2ADC (⫝̸) FORKING is... U+2ADD U+0338 ⫝ ̸ Grrrrr
 
@DavidCarlisle ???
 
@JosephWright unlike (almost every other character) that one decomposes to a base and combining slash in NFC (so I got told off by the automated W3C pubrules checker form not having the text in NFC)
 
@DavidCarlisle :)
 
4:39 PM
This is Unicode's idea of "correcting" the mistake they made whan adding the character (FORKING is a negated pitchfork symbol and NONFORKING is a pitchfork symbol)
(which Will copies in unicode math with \fork having the negation slash and \nonfork not having the slash.) so Now I hav eto work out which is easier, changing a decades old list of codepoints to be NFC or asking the webmasters for an excemption to allow a single codepoint here:-)
 
One final note about a table column for units. How should the following be fixed to make siunitx apply the new unit column s and parse it correctly?

\documentclass[12pt]{article}

\usepackage{siunitx,xltabular,multicol,booktabs,ragged2e}
\usepackage[table]{xcolor}
\usepackage{collcell}
\newcolumntype{s}{>{\collectcell\unit}c<{\endcollectcell}}

\parindent=0pt

\begin{document}
\rowcolors{2}{gray!15}{}
\setlength{\tabcolsep}{5mm}
\begin{xltabular}{0.5\linewidth}{%
X X
S[
table-format = 4.3e-1,
parse-numbers = true,
 
5:02 PM
@Diaa The s column type has been removed from siunitx version 3
 
@egreg So, for the new column s created by \newcolumntype{s}{>{\collectcell\si}c<{\endcollectcell}}, can I locally make it have these options [per-mode = symbol, bracket-unit-denominator, sticky-per]?
@egreg Also, how to left-align its units since \newcolumntype{s}{>{\RaggedRight\collectcell\si}c<{\endcollectcell}} doesn't do the trick?
 
5:28 PM
@Diaa Here's a way
\documentclass[12pt]{article}

\usepackage{siunitx,xltabular,multicol,booktabs,ragged2e}
\usepackage[table]{xcolor}
\usepackage{collcell}

\newcommand{\myscolumn}[1]{\unit[per-mode = symbol, bracket-unit-denominator, sticky-per]{#1}}
\newcolumntype{s}{%
  >{\collectcell\myscolumn}
  l
  <{\endcollectcell}
}

\parindent=0pt

\begin{document}
\begin{tabular}{
  ll
  S[table-format = 4.3e-1,
    parse-numbers = true,
    table-alignment-mode = none,
    table-number-alignment = right,
    exponent-mode = input,
 
@FaheemMitha -- But even some emacs devotees abhor spell checkers.
 
@barbarabeeton some emacs devotees never make any typos so don't need such aides
 
@DavidCarlisle -- "aides" (persons) ==> "aids" (tools)
 
@egreg Thanks for the help. If you allow my greediness, is it possible to pass the options to s when calling it so that, inside tabular, I can write s[per-mode = symbol, bracket-unit-denominator, sticky-per] wihtout having to go back and forth to see what options I wrote?
 
@barbarabeeton emacs users know that, firefox users apparently not.
 
5:37 PM
@Diaa I don't think so, at least not with \newcolumntype
 
@DavidCarlisle -- Interesting. I have both emacs and firefox "live" at the moment.
 
@egreg In Joseph's answer tex.stackexchange.com/a/600854/2288, I can't left-align as you did here. So, I wish I could combine both ways of writing the options near the column specifier while aligning the units.
 
@daleif -- Yikes! That's frightening!
 
@Diaa Replace >{\collectcell\si}c<{\endcollectcell} with >{\collectcell\si}l<{\endcollectcell}
 
@DavidCarlisle quack
 
5:48 PM
@PauloCereda di...
 
@DavidCarlisle oh no
 
@egreg Thanks for the help, I overlooked the second 'c' to replace inside his patch. Have a nice day!
 
@DavidCarlisle -- Sadly, I must report that Unicode did what they were asked. That confusing pair was defined in that seemingly backward way in the attested example submitted to STIX. Should still be in the files at AMS. (Whimper.)
 
6:36 PM
@barbarabeeton whoever, it's still nonsensical, and making forking not in nfc is a real pain, as it means you can't simply make a list of unicode code points by going through the numbers you have to special case that ine and miss it out, still theer are worse things:-)
 
@UlrikeFischer Thanks.
 
@AlanMunn did it work?
 
@UlrikeFischer I created a global counter that increments every chapter no matter what the value of the actual chapter counter is, and then used that to define \theHchapter for appendices.
@UlrikeFischer The requirements are really weird in that they require per-chapter appendices and bibliographies to be formatted like chapters, but with non-unique appendix numbers.
 
@AlanMunn typically if you have to do curious things with hyperref for the counters this indicates curious counter settings in the document ;-)
 
@UlrikeFischer Yes, indeed. And that's the source of the problem. And since this is used by students all over the university, I need to reduce user-level intervention as much as possible.
 
6:52 PM
@AlanMunn you could ask them if they think that this numbering style is accessible and how a blind user should find an appendix.
 
@UlrikeFischer I've learned the hard way not to ask. This is an office run by bureaucrats who have no understanding of anything.
 
@barbarabeeton What's wrong with spell checkers? They're great.
 
@UlrikeFischer And since LaTeX isn't yet able to produce the most accessible documents anyway, I probably shouldn't make an accessibility argument.
@FaheemMitha There grate if yew no watt your doing.
 
This one is proving to be annoying to search for. How do I include a vertical bar in a tabular cell, please?
 
@FaheemMitha You mean inside the cell or as a border to the cell?
 
6:56 PM
I'm getting a strange kind of double dash. I haven't tried an MWE, but I also don't think I'm doing anything out of the ordinary.
@AlanMunn Inside the cell. As a regular character.
Not as a separator thingy.
 
22
Q: How do I typeset vertical and horizontal lines inside a matrix?

zellyn(From the notation used in course notes for Stanford CS229 available online.) I'm trying to typeset the following matrices made up of row or column vectors: Any ideas?

 
@FaheemMitha You know how the community works regarding MWE, so...
 
@AlanMunn I like ispell.
@AlanMunn Reading. Thank you.
Ugh, those aren't great solutions. Why don't vertical bars just work? I suppose the answer is some TeX weirdness, as usual.
 
@AlanMunn well yes. But future accessibility support must be considered in class designs now. I mean if we extend the support there, then curious setups are only an obstacle. Automatic accessibility will need rather sane structures.
 
It works, but the spacing is kind awful.
 
6:59 PM
@FaheemMitha The \vrule answer is better than mine. (This was one of my first answers on the site...)
 
@AlanMunn That's the one I tried.
 
@FaheemMitha -- Spell checkers induce unreasonable reassurance that everything is okay. They (in general) don't pay attention to syntax or context. Consider this announcement I found posted on a university bulletin board: "The semester will being on September 1." Everything is spelled correctly, but it's not grammatical and it doesn't make sense. (I can cite other examples.)
 
@UlrikeFischer Of course I agree with that. And it's an easy fix to make them sequentially counted.
 
@barbarabeeton They're not perfect, of course. Nothing is.
But they're a lot better than nothing.
 
@barbarabeeton I gave one above. :)
 
7:02 PM
If natural language syntax checkers exist, I haven't met them.
If you expect spell checkers to check your grammar, you obviously think the universe is a lot nicer than it actually is.
 
@barbarabeeton ooh one of the cross beams has gone out askew on the treddle
 
@AlanMunn There is no space on either side of the vertical bar. Any suggestions on how to fix that?
 
@FaheemMitha MSWord does a pretty good job of catching lots of mistakes. I wouldn't trust its corrections, but it does find stuff. If my non-native speaker students used it instead of LaTeX I'd have a lot less correcting to do. Of course there are other reasons for using LaTeX, but grammar checking (and even spell-checking) isn't so well developed for it.
 
@AlanMunn -- Yes, and it's a good one. But most people don't reread what they write. A few days ago I came across this: "... the schools attracted are the zoo, rejecting the status quo ...", which makes sense if "the zoo" is transmogrified into "those who". I concluded (correctly) that the text had been transcribed automatically from a spoken version.
 
@AlanMunn I see. You mean it's more than a spellchecker?
 
7:08 PM
@barbarabeeton :) that's a good one.
 
@FaheemMitha Oh yes. For example in the sentence that @barbarabeeton gave you "The semester will being on September 1" it flags 'being' as incorrect.
 
So, I added \, on either side of \vrule, which works, but isn't what I'd call a nice solution.
@AlanMunn That's impressive, certainly.
 
@PauloCereda -- Probably should be "treadle".
 
But bad writing is hard enough for humans to fix, let alone a machine.
 
7:09 PM
@barbarabeeton :)
 
@barbarabeeton I'm the sort of freak who proofreads my emails.
 
> David Carlisle submitted an update to the PicTeX package.
ooh
 
It takes a lot of time, and mostly people don't properly read or answer them. A thankless job.
 
@FaheemMitha Now it doesn't "know" that the correct word should be "begin"; it suggests that the form of the verb is incorrect and should be "be", but it does flag the error.
 
@FaheemMitha -- And that's as it should be -- bravo! (I find, too often, that I haven't.)
 
7:11 PM
@AlanMunn Hmm.
@barbarabeeton I didn't always, though. But I find I get fussier as I get older. Probably a vain subconscious effort to try to impose order on a chaotic world.
 
@PauloCereda -- Code not updated, but the manual is now available (pdf only).
 
@barbarabeeton ooh
 
@PauloCereda see I can draw diagonal lines now, can tikz do that?
.
 .
  .
3
 
@DavidCarlisle ooh
 
In this answer https://tex.stackexchange.com/a/572153/2288, how should the definition of \snum be edited to make it work with siunitx version 3?

\documentclass{article}

\usepackage{siunitx,tikz}

\newcommand{\snum}[2][]{\pgfmathifthenelse{#2 >=0}{"+"}{"-"}%
\mathbin{\pgfmathresult}%
\num[#1,bracket-negative-numbers,close-bracket={},open-bracket={}]{#2}}

\begin{document}
\def\numa{5}
\def\numb{-3}

(a) $1 \snum{\numa} \snum{\numb}$

(b) $1 \ifnum\numa>0+\fi \numa \numb$

(c) $1 \pgfmathparse{ifthenelse(\numa>=0,"+",)}\pgfmathresult \numa \numb$
 
7:19 PM
@PauloCereda nicely rounds up his collection of picture mode art
 
@UlrikeFischer ooh a collection
 
@FaheemMitha why are you using OT1 encoding? it is unrelated to tables, in any sensible encoding the character | typesets as an |. In OT1 it is an emdash.
 
 
2 hours later…
9:33 PM
@AlanMunn grammarly also catch that. I, as a non native speaker, find the new grammar checker quite useful. You have to take them with a grain of salt, but they're definitely useful.
 
@Diaa Code is online
 
@Rmano I don't know it. What's the catch? It says it's a free FF addon. Is there more to it than that?
@Rmano How does it interact with LaTeX source? It sounds like it would be useful for me to recommend to students.
 
@AlanMunn it works for free for basic recommendations, and it offers a paid upgrade for style and "tone" and other things like that. My university gave me a full account for one year and I'm testing it. They have no plugins for text only editors, unfortunately, but I survive copying and pasting plain text in the web interface.
The style thing is nice but it's obsessed by convincing you to avoid passive forms. I don't know if that counts as a plus or a minus...
 
@Rmano I see. Do you know if it would work with Overleaf automatically?
 
I was using the free addon before since two years ago and I find it clearly useful.
 
9:43 PM
@Rmano Yes, Word has the same obsession.
 
No, it doesn't work on overleaf. I opened a ticket about it and they say they'll consider it (which I read as no chance in hell...)
 
@Rmano :) Maybe so. I think for students, the some level of automaticity is key, unfortunately.
@Rmano But I'll maybe recommend it and see whether things improve.
 
@AlanMunn yes. Anyway I recommend to just try it a bit...
Again, maybe for native speakers it's not so useful. But it catches my common mistakes with plurals and verb tenses (my daughter says I distribute "s"s randomly in any language) an that's valuable for me ;-)
 
 
1 hour later…
10:56 PM
@AlanMunn and @Rmano -- One criticism I've heard re Grammarly is that it tends toward the "least common denominator; if one has or is trying to develop a distinctive style, its recommendations might tend to be "middle of the road", completely grammatical and spelled correctly, but devoid of charm. I'm not thinking of fiction or ad jargon, but wonder what it would do if fed John McPhee's "Annals of the Former World", a tour de force exploring U.S, geology.
 
@barbarabeeton Very pragmatically for a student's thesis I'm completely happy with that. :)
 
@AlanMunn -- That's probably wise, but it's really nice if a complicated subject can be explained in a way that's enticing as well as accurate and thorough. (I've recently come across a book that falls into this category: Andrew Knoll, "A Brief History of Earth: Four Billion Yeats in Eight Chapters". I wish the photographic images had higher resolution, but the prose is top notch.)
 
@barbarabeeton Oh, absolutely. But that's something that takes a lot of practice to develop.
 
@AlanMunn -- True enough, but I'd hate to discourage it. If you do see any glimmers of future possibilities in that direction, do mention it and encourage it.
 

« first day (3876 days earlier)      last day (1054 days later) »