Uhh, er, erm

Wed, 01 Apr 2026 00:00:00 +0000

For the past few weeks I’ve been working on a personal project. I came up with the idea as a joke response to a post (since deleted) written by a friend elsewhere. But what is this idea?

There already exists a programming language Erlang. I have designed a language unrelated to this but with the name uhhlang. It is a stack-based language and the ONLY valid tokens (aside from comments) are hesitation words in various languages - so uh, erm, euh, äää, and so on. I’ve taken words from English, Danish, French, German, Hindi, Mandarin and more.

A tradition in the programming world is for an early (usually the first) program that a new user creates to display the message “Hello, world!”.

In uhhlang, a Hello World program looks like the following

ə ə Say hello
errr errrr ymmm errr uhhh äh
y
err хмм errr ym errr errrr errr ym ymmm ymmm
uhh errrr хмм ymmm errrr ymmmm
uhh errrr errrr ym errr ymmmm ymmm
uhh
uhh ää err ymmm äää errrr 呃
errr errrr ym хмм
uhh
errrr errr errr ym ym ymmm uhhh

errr errr errrr ymmm ym er er ymmmm ymmm
errrr errr ym errr ym ym err ymmmm
uhh errr errrr äh ymmmm
uhh errrr ymmm
errr errrr ym хмм
uhh errr errrr äh ymmmm
хм err ymmmm хмм err ymmm
.y.
uhhh uhhhh
אֶההה
ehh

I still have some more features to add (variables, associative arrays, perhaps file I/O) but it is minimally usable. I make no statement about useful.

This is the first language I’ve designed, and in doing so I’ve learned a plenty about language design. I’ve also learned lots about Go, a language that I’ve only used in small amounts. My apologies to seasoned Go-developers who might recoil in horror. But to them I ask, have you noticed the launch date of this language?

The repo for the language is located at https://codeberg.org/mattp/uhhlang/src/branch/main

Z is for zh

Tue, 30 Apr 2024 00:00:00 +0000

I’ve mention in several posts in this series that I enjoy language - and languages. The language I started studying most recently is Mandarin, sometimes called Standard Chinese or Standard Mandarin Chinese.

But why zh? That’s the ISO 639-1 code for the language, coming from the Chinese name of the language 中文. One of the most popular ways of writing Chinese characters in the Latin alphabet is called pinyin. Pinyin is typically used to make it easier for non-speakers of the language to sound out the characters but it’s also one of the techniques used to type Chinese characters one a qwerty keyboard.

The pinyin representation of 中文 is Zhōng wén (occasionally zhong1 wen2), and it’s this zh from which the ISO 639-1 code is taken.

For a couple of years I was having 2 lessons per week though I took a break a couple of months ago. I learned not only to speak, but also to read and write a little. It’s probably been the most interesting language I’ve studied. The hardest part is not (perhaps surprisingly) the tones, but rather the writing side of things.

Frustrating though this is, I’m not too disheartened. During the course of my studies, I leaarned that even Chinese natives sometimes get so-called character amnesia, mainly due to the proliferation of computerised inputs that people use as opposed to writing by hand.

A few days ago I hinted at the open question of what it means to speak a language. When I first set out to learn, I set myself the goal of being able to order a meal. And that goal I’ve achieved; I’ve been to restaurants in Chinatown and ordered. Had to fall back on English for little bits of it, but I still consider it a success. Once I’m feeling rested, I plan to resume lessons.

Y is for "Yesteryear"

Mon, 29 Apr 2024 00:00:00 +0000

English has some lesser used and (and somewhat archaic) words to describe time relevant to the present. The first of these is representative of today’s letter Y, the word being yesteryear. It means “last year”, analogous to yesterday.

One more word in this category is overmorrow, meaning “the day after tomorrow”. I first encountered the equivalent of this in German when in school, not having realised at the time that English had its own version. The German word is übermorgen, which dissects in exactly the same way the English word does.

Yet another word in this category is ereyesterday, which means “the day before yesterday”.

X for 'Xenophon Zolotas'

Sat, 27 Apr 2024 00:00:00 +0000

For the letter X, today's entry is a head nod to former Greek economist and Prime Minister Xenophon Euthymiou Zolotas.

English has inherited words from multiple language families. Two of the speeches he gave are noteworthy because they contain words mainly of greek origin. Here are the texts of those speeches.

From 1957

I always wished to address this Assembly in Greek, but realized that it would have been indeed "Greek" to all present in this room. I found out, however, that I could make my address in Greek which would still be English to everybody. With your permission, Mr. Chairman, I shall do it now, using with the exception of articles and prepositions, only Greek words. Kyrie, I eulogize the archons of the Panethnic Numismatic Thesaurus and the Ecumenical Trapeza for the orthodoxy of their axioms, methods and policies, although there is an episode of cacophony of the Trapeza with Hellas. With enthusiasm we dialogue and synagonize at the synods of our didymous organizations in which polymorphous economic ideas and dogmas are analyzed and synthesized. Our critical problems such as the numismatic plethora generate some agony and melancholy. This phenomenon is characteristic of our epoch. But, to my thesis, we have the dynamism to program therapeutic practices as a prophylaxis from chaos and catastrophe. In parallel, a Panethnic unhypocritical economic synergy and harmonization in a democratic climate is basic. I apologize for my eccentric monologue. I emphasize my euharistia to you, Kyrie to the eugenic and generous American Ethnos and to the organizers and protagonists of his Amphictyony and the gastronomic symposia.

From 1959

Kyrie, it is Zeus' anathema on our epoch for the dynamism of our economies and the heresy of our economic methods and policies that we should agonize the Scylla of numismatic plethora and the Charybdis of economic anaemia. It is not my idiosyncrasy to be ironic or sarcastic, but my diagnosis would be that politicians are rather cryptoplethorists. Although they emphatically stigmatize numismatic plethora, they energize it through their tactics and practices. Our policies have to be based more on economic and less on political criteria. Our gnomon has to be a metron between political, strategic and philanthropic scopes. Political magic has always been anti-economic. In an epoch characterized by monopolies, oligopolies, monopsonies, monopolistic antagonism and polymorphous inelasticities, our policies have to be more orthological. But this should not be metamorphosed into plethorophobia, which is endemic among academic economists. Numismatic symmetry should not hyper-antagonize economic acme. A greater harmonization between the practices of the economic and numismatic archons is basic. Parallel to this, we have to synchronize and harmonize more and more our economic and numismatic policies panethnically. These scopes are more practicable now, when the prognostics of the political and economic barometer are halcyonic. The history of our didymus organizations in this sphere has been didactic and their gnostic practices will always be a tonic to the polyonymous and idiomorphous ethnical economies. The genesis of the programmed organization will dynamize these policies. Therefore, I sympathize, although not without criticism on one or two themes, with the apostles and the hierarchy of our organs in their zeal to program orthodox economic and numismatic policies, although I have some logomachy with them. I apologize for having tyrannized you with my Hellenic phraseology. In my epilogue, I emphasize my eulogy to the philoxenous autochthons of this cosmopolitan metropolis and my encomium to you, Kyrie, and the stenographers.

I freely admit I need a dictionary for some of these words. Nevertheless, I consider it an impressive feat. Despite my love of language and languages, I've not yet made dedicated efforts to learn Greek (whether modern or ancient).

W is for "Webdesign"

Fri, 26 Apr 2024 00:00:00 +0000

I’ve been in involved in web-adjacent technology for pretty much all of my working life. I’ve worked on some significant large scale websites (both government and retail). Despite that, I’ve done very little web design. Even when working for a Ruby consultancy I was mostly involved in back end components.

Furthermore, when in school my artistic eye never really developed. As a result my web design skills are pretty much non-existent. But within that space I’ve learned plenty in the course of maintaining this blog, especially over the this A-to-Z series. It’s definitely rejuvinated a bit of passion for me in the world of tech.

That said, I’m still mindful of my limits. I’m thankful that the tech space is sufficiently broad that this lack hasn’t held me back. I’ve been able to write backend or system code or look after infrastructure, deliver training, and plenty more.

I’m unlikely to switch directly into web dev but I am at least scratching that itch, the desire to learn.

Whether in technology or not (and whether intentional or not), what facets of your industry have you avoided?

V is for "Vexillology"

Thu, 25 Apr 2024 00:00:00 +0000

Today’s challenge letter is V and the topic I’ve chosen is Vexillology. Wossat? It’s the study of flags. I hesitate to call it an interest of mine, since that feels like an exaggeration but it’s still worthy of mention.

When I was in school (I reckon this was sometime around 1990, based on the room I was in), one of the posters on the wall was flags of the world. I remember noticing that there was one monochromatic flag.

This was Libya, as the flag was then. (It’s since been updated)

Many years later my dad took an interest in sailing. Part of the learning materials was to learn the maritime signalling flags. Being the curious and information-hungry person that I was (and still am!), I took an interest and learned a few of them. At the time, several of them stood out to me due to their shape:

The letter A:

The letter B:

The numerals 0-9. As an example, here is “2”

I can still remember some of them now, along with their meanings. I have never sailed, only ever having been a passenger. Strange, the things one remembers.

Another type of flag is one used for communities. Probably the most well known of these is the Pride flag created by artist Gilbert Baker in 1979.

There have been several variations of this flag over the years, additionally numerous other groups within the queer community have their own flags such as the trans pride flag – enough quer flags exist to write an entire post.

Another community flag (not part of the queer umbrella), and significantly less well known, is this one:

This one is the Esperanto flag – Esperanto being one of the languages I speak.

A couple of years ago I stumbled on the flag of Nepal.

It’s not an overly new flag; it was adopted in 1962. But what very much caught my interest was not only that non-rectangular, nor that it’s ornate, but also that its construction is mathematical in nature and codified in their consitution. If you’re so inclined, get a ruler, a pen, and a piece of paper and give it a go. Or you can use Geogebra.

I’ve mentioned 3 types of flag here: Community, Country and Maritime. On a couple of occasions over the past few years I’ve run a quiz (or sometimes only a round in a quiz) and the questions took the form of: “Is this flag that of a country, a community or a maritime signal?”. Feel free to copy this idea!

U is for "Unfinished"

Wed, 24 Apr 2024 00:00:00 +0000

When I still lived in Manchester a friend asked me to join a small project he’d created. This was a small volunteer organisation which hosted websites for local community organisations at zero cost.

Today’s letter, U, represents the name of this venture: The Unfinished Project. The name is to describe our mindset for the project, that work is never complete, that it’s always unfinished, that there is always something more to do.

It’s typically been 2 or 3 core people running it for the past 20 years but we’ve also taken on young folk interested in technologies adjacent to websites, email, and hosting. It has been great to provide these people with an opportunity to work on real public facing systems and gain experience, and see these people move on to bigger things.

Even though Unfinished stuff takes little of my time these days, this mentoring side of things has long been important to me and I’ve been very lucky to be able to incorporate it into my career for vast chunks of it, in addition to the volunteer capacity I mention here.

T is for "Tracking"

Tue, 23 Apr 2024 00:00:00 +0000

Today’s entry in the A-to-Z of Blogging Challenge is a bit of a rant. The letter is t and the topic is Tracking.

I hinted the other day (when talking about RSS) about my disdain for adverts. Something I dislike equally is the normalisation of site visitor information being collected, and sold on to the highest bidder. It’s this tracking that I’m writing about today.

Often on social network sites there’s a “share” button. These often add a small tracking component. Because many of chat-based apps have link automatic previewers it means that when such a link is shared then the system that the link was sent in (even if different from where the link originated) will issue a request behind the scenes to expand that link. As a result, the site where the link originate will know of the association between the sending party and you the recipient. You typically don’t even have to click on the link for this information to be shared.

For example: https://social.example.com/post/12345678?tr=qwertyu is turned into https://social.example.com/post/12345678

Every site has its own tracking components. It’s also not the case that any string of numbers or letters on the end of a link is necessarily tracking in nature. It means that the method of cleaning a link will differ on a per-site basis.

To help with this, I’ve written a very simple tool and published it on this site at https://peperell.com/stuff/jstoys/linkstrip.html.

If you care about privacy then you can use this tool to clean up a link before sharing it with someone. Note also that if you are comfortable with removing such tracking components yourself, pasting a link into a chat box and then editing it before clicking send is not necessarily sufficient to remove the tracking info - because of the aforementioned automatic link expansion / preview features. You can either do this manually (via a text editor running locally on your machine - not Google docs!) or with this tool.

The current version is very simple, and I intend to improve it over time. It strips out the tracking elements but still links to the desired content.

If you’re a user of a site I’ve not yet taught this tool how to understand, I’d welcome assistance. You can either leave a comment here or instead via Github.

S is for "schema.org"

Mon, 22 Apr 2024 00:00:00 +0000

A few months ago I encountered an open set of guidelines for publishing structured data on websites. This standard is referred to by the name “schema.org” or sometimes by just “schema” and it is published at https://schema.org.

The use is rather niche and likely only of interest to site creators and authors of software which indexes web content, although I’m being rather reductive here.

When writing structured data for the web, especially in prose, a search engine cannot extract the relevant details unless given appropriate hints.

E.g. imagine a heading

<h1>Arrival</h1>

Without any such clues, it’s impossible to know whether this refers to

arrival instructions for some event
arrival times for a train or flight
The 2016 movie Arrival starring (amongst others) Jodie Foster
Possibly something else

But the markup can be changed to include this information without affecting the presentation. In fact, I’ve marked up the movie example in that third bullet point. The markup I used is

<span itemscope itemtype="https://schema.org/Movie">
  The <span itemprop="copyrightYear">2016</span> movie
  <span itemprop="name">Arrival</span> starring (amongst others)
  <span itemprop="actor" itemscope
    itemtype="https://schema.org/Person">
    <span itemprop="name">Jodie Foster</span>
  </span>
</span>

It looks very noisy in raw markup, I’ll admit, but look up again at that third bullet point. The extra information such as the the fact that the number is a year (obvious to a human) can be parsed by compliant tools, but doesn’t clutter the text as seen by the end user.

Another case for where they’re useful is contact details for an organisation. Ever searched for a company’s phone number etc - in a search engine, I mean, not directly on their site? With appropriate hints (such as those provided by schema.org) the search engine can provide this even on the search results page. It will know to extract certain information and ignore the rest. You’ll be unsurprised that there is a suitable schema defintion: Organization which has the telephone property.

In some of the other posts I’ve made in this series I mention various books and I’ve made an effort to mark up those book references with the correct schema type, Book.

I’ve been unable to find recent adoption figures for the initiative, but the Wikipedia article currently shows the 2016 adoption figure at 17% across a subset of marketing agencies and industry adjacent organisations. Not excellent, but perhaps the figure is higher now.

I don’t know of any browser plugins to surface this information, but I can imagine a smart phone feature: being able to extract a company’s phone number when viewing their ccontact page rather than having to copy and paste the number into the dialler app.

The book example I gave above? What if your browser (whether tablet, phone, desktop, laptop or other device) were able to add this to your wishlist from your favourite bookstore? Or reserve it at your local library?

You can likely think of more examples when browing the available schemas.

Wishful thinking, perhaps, but wouldn’t it be nice?

R for 'RSS'

Sat, 20 Apr 2024 00:00:00 +0000

Continuing the A-to-Z Challenge, today’s letter is R and the topic is RSS.

RSS (Really Simple Syndication) is a way that a site can publish its information to facilitate the erading of the content via another app or a separate website which aggregates the content.

If you’ve ever seen this orange icon on on a site then that will typically link to the RSS feed for the site. I have an RSS feed for my site (also linked at the bottom of each post).

The benefit of this is that one can configure an RSS-reader (as they’re called), add the sites which one reads on a regular basis and then have the content all in one place. Depending on the site in question, this can also strip out the adverts (though there are none on my site), and either include or exclude the comments.

One of the most widely used RSS readers was the now-discontinued Google Reader. RSS usage took a huge nosedive when that service was dropped. I suspect, as alluded to above, this was due to the loss of ad revenue on sites because only the content was extracted. Another stronger force is likely to be the proliferation of social network sites which provide similar aggregated feeds of subscribed-to content. Such aggregations, being under control of providers can have their order change, ads inserted, and certain material masked or promoted. RSS aggregation bypasses all that.

For a short while I used The Old Reader (which is still running) but I had exceeded the subscription limit of the free plan when that limit was introduced but I felt like I didn’t read content often enough to warrant any of the paid-for plans.

A little over a year I learned of rssnix which is a self-managed RSS reader written in Go. It had a couple of bugs listed in the issues which looked simple enough(!). I was actively learning Go at the time so I used that as an opportunity to dive in and fixed a couple.

I still use that now, though very much on an ad-hoc basis.

If you publish a blog, do you provide an RSS feed? Do you consume RSS feeds?

Q for 'QR Codes'

Fri, 19 Apr 2024 00:00:00 +0000

Continuing the A-to-Z Challenge, today’s letter is Q and the topic is QR Codes.

First off, what are they? The abbreviation QR code is short for “quick response code”. They are the black and white small pixelated squares that we can “scan” with a dedicate device or (more commonly) our mobile phones.

Here’s one which links to the Wikipedia page for QR codes.

I first encountered QR codes in the early 2010s. At the time they were seen, outside of industry, as niche, obscure and somewhat faddish, perhaps even with a bit of ridicule. But it seems that very quickly they became more widespread and accepted.

Event publicity posters will often contain a QR code linking to the event page so that one does not have to type in the URL manually, an error-prone process. They’re also used for self-printed train tickets in the UK (previously one would collect tickets from a machine, the kind of tickets which have a magnetic strip on the reverse).

In doing research for today’s post, I learned that they were considerably older than I thought - having been invented in 1994. Traditional 1-d barcodes were first used in a retail context in 1974. Small nostalgia story: using this fact helped win me and a housemate a year’s supply of beer in 2004. Rather amusing as he doesn’t drink at all and I don’t drink beer.

Around a month ago I saw a coloured version of something similar, using pixels of cyan, magenta yellow and black. I forget what it was on but that same week (same day, I think) I saw another one on a breakfast cereal box.

As can be seen in the text adjacent to the matrix code, it’s a variant of QR codes with the brand name “Navi Lens”. It’s a Spanish company and the IP is patented technology rather than being an open standard but I still think it’s a great idea because it solves some of the accessibility problems that traditional QR codes have.

As best I can tell, the reader app is free for end users. I wonder what their funding model is; in other words how can the idea last? Perhaps it’ll be done on a licensing basis like QR codes (which are also patented).

Have you used either? I’m particularly interested in the use of NaviLens to learn how the usability and accessibility compare.

P for 'Punctuation'

Thu, 18 Apr 2024 00:00:00 +0000

Today’s letter is P and the theme is punctuation.

In particular I wish to talk about two punctuation marks: The Oxford Comma and the Interrobang.

The Oxford Comma is used to describe the comma which appears before the word ‘and’ in a list of items when written out in prose. Style guides between publishers vary whether it should be used or not. When I was in school I was told that the final comma should be omitted. It was only in my mid-to-late 20s when I learned it was a thing. When I first heard of it I was undecided, but I think I’m shifting to prefering to use it, rather than not.

For example: My pets comprise a cat, a dog, three fish, and an iguana. Without it, the text would be: My pets comprise a cat, a dog, three fish and an iguana. (Side note: I have only a cat.)

The other character I mentioned in the opening is the interrobang. It’s a single character which combines the function of a question mark and an exclamation mark. Why on earth would someone want that‽

Typically it’s used for rhetorical or humourous questions, such as the example above.

Typing the interrobang character in HTML is straightforward - one can use the hex code. But sometimes, for example when writing in plain text, that option is not avaiable. Typically what I do in this situation is to search “whatever character” (or “whatever character code”) and then copy-and-paste from one of the results. But that can be awkward, not least because search engine results typically change over time.

To close, therefore, I’m including here some of the characters (not only punctuation symbols) that I need to type in order to have them together in one place. Likely it’ll be me later referring to my own post, but feel free to bookmark such that you can come back too!

‽ Interrobang, ‽
° Degrees, °
™ Trademar, ™
© Copyright, ©
± Plus/Minus, ±
µ Micro / Mu, µ
¶ Paragraph, ¶
∑ Captial Sigma / Sum, ∑

Which symbols do you type with regularity that can’t be typed with a single keystroke?

O for 'Ologies and Isms'

Wed, 17 Apr 2024 00:00:00 +0000

Today’s letter is O and the associated topic is -ologies and -isms.

I love words and etymology. As a child I used to read the dictionary recreationally, and even asked for a chunky dictionary for my birthday one year, The Chambers Dictionary.

Many years later (after I'd finished University) I discovered another dictionary-type book titled Ologies and Isms and its this book that I'm writing about.

Despite the title, it's not just limited to words containing those two suffixes. It does explain the etymology for many words and phrases which are used in English.

If etymology is your thing you'll probably enjoy it. Also of possible interest is Etymonline which I occasionally use too.

N is for 'New Beginnings'

Tue, 16 Apr 2024 00:00:00 +0000

Today's topic is one that makes me slightly nervous, although that's not today's word. The word (perhaps phrase would be better) is new beginnings.

Having worked for the same company for eight and a half years, in December I quit my job with the intention of having a short break and then starting my own company. In March, I did exactly that. I've started a technical consultancy offering services in my areas of expertise (software development, cloud solutions, and delivering technical training).

I've called my company mlem blep Ltd. Continuing the cat theme, the company logo is a drawing of my cat, Freddie, with his tongue out (having just been interrupted washing himself). There are a couple more pictures of him in my post earlier in this series in which I wrote about cats.

You can see the site at https://www.mlemblep.com/. For the moment it's just me, but I am open to the idea of growing if, as, and when it feels right.

M is for 'Mathematics'

Mon, 15 Apr 2024 00:00:00 +0000

Today's letter is M and the associated word is Mathematics.

I've enjoyed numbers and maths for as long as I can remember. In school numbers and mathematics were taught to me as a natural pairing, perhaps close to synonymous. But as I've learned more not only has my interest grown but I also realised that that pairing is hugely reductionary.

One Christmas in the early 90s I received the book Chaos by James Gleick (ISBN 9780749386061). The book has very little numerical data in it; rather it's a treatise about the research and development of mathematics in the mid 1900s of a branch of mathematics called Chaos theory. Very loosely speaking, it's about how tiny changes in a system can result in huge consequences. It's a very accessible read, even to non-mathematicians (though it would be less salient to those with zero interest in the field).

Way back in early 2009 I was trying to decide what to do for my 30th birthday. My working premise was along the lines of "How many 30th birthdays does one go to? And how many does one remember - perhaps one's own, and maybe one other. I want mine to be the one that people remember that wasn't their own!". The idea I settled on was to hire the Royal Insitution, and a guest speaker to talk about something sciency or technological. I reached out to several people and settled on one ("Myths in Hollywood Science") which was hugely entertaining. But as a result of having reached out to a handful of science communicators I stayed on their mailling lists. And it's through one such contact that I learned of the event MathsJam (which I mentioned a few days ago).

Though I enjoy games and socialising, maths is something I can do alone during those occasions in life when one doesn't want to be around people. Maths is very much an interest which has been with me for as long as I can remember.

L is for "Languages"

Sat, 13 Apr 2024 00:00:00 +0000

Today’s L-word is “Languages”.

I’ve had an interest in languages for as long as I can remember. My mum is good at languages too, so perhaps some of her interest influenced me.

In school I studied French, German and later Latin. I wasn’t overly good at Latin but I did enjoy it. Studying the language taught me plenty about English too, and I’ve found this to be the case with subsequent languages I’ve learned - whether formally or with self-study.

In a previous job (in 2015-6 ish), a colleague taught a series of lunchtime lessons in Spanish. I learned a little, but didn’t really continue. That said, I do remember one lunchtime towards the end of the course when I was buying lunch from a nearby takeaway and one of the servers had a Spanish flag on his name badge. I greeted him and explained that I’d been learning at work for a few weeks. I got given some free food (I think it was a cake, but I no longer remember)

Slightly longer ago (in late 2010) I learned Esperanto. Through that language I’ve made many friends across the world and travelled to many new countries. Often at Esperanto events there are taster courses for the language of the host country. As a result, I’ve toyed with Hungarian (the grammar and phonology of which I love) but haven’t continued it beyond some very basic phrases. (Barely enough to even understand responses!)

Many of the events have been in Poland and Slovakia. Despite having taken several of these taster courses, I’ve not really progressed. It’s weird; despite progressing (with mixed results) in other languages, this family is particularly troublesome for me. The language structure and phonology seems to not stick.

In late 2019 I went to Japan for the Polygot Conference but extended the trip to spend almost 2 weeks there. Prior to the event I’d planned ahead by having a few lessons in Japanese. This helped a little but it was only at a very elementary level. I’ve not continued my Japanese studies.

But the language Esperanto has been a continued theme in my life. I use it almost every day, even if outside of event this is limited to only a handful of short messages sent to friends online.

A recent language that I started studying in January 2022 is Mandarin, one of the languages of China. For almost 2 years I had a couple of lessons per week though a month ago I decided that I needed a break. As is often the case with life, energy levels go up and they go down. I felt like I needed a break. My goal when I set out was to be able to order food - which I can now do. I’m also able to read very basic children stories. I intend to resume lessons when the time is right.

When I speak with people about my interest in languages, a common question is “How many languages do you speak?”. That answer varies depending on what it means to speak a language. For European languages there is the CEFR. Using this a a metric I’d say I have (in addition to my native English) one at C2 (Esperanto), and two at A2 (French, German - years ago these were B1 perhaps bordering on B2) but one forgets through disuse.

Thinking about it, though I’ve had a number of passing interests, languages (along with bellringing and computers) have been pretty constant.

What language(s) do you speak, and to what level?

And unrelated speifically to languages but merely because my last-but-one paragraph touched on the subjet, do you find that your interests are mainly constant, or do they change over time?

K is for "Knitting"

Fri, 12 Apr 2024 00:00:00 +0000

One of the hobbies I have (though I’m less active in it at the moment) is knitting - mainly by hand though I’ve done a tiny bit of machine knitting when I had access to one. I first learned by hand in the late 80s or early 90s from my maternal grandmother. I didn’t have much staying power though, so never made anything larger than swatches or practice sqaures at that time.

Around 2006 I had a housemate who knitted and she encouraged me to start up again. So I did, and I’ve since made all sorts of things: hats, gloves, scarves and socks. A pair of socks is my favourite thing, though I’ve made several hats too.

A pair of socks that I’ve made

On the tech side of things, around 15 years ago I came across KnitML, an XML DTD to describe knitting patterns than could then be rendered into prose (in English or any other human language) and even simulated into graphics form to assist with resizing existing patterns and design of new ones. Unfortunately that project is no longer active but it was fun to read from afar whilst it was still in existence.

In recent years I’ve run a handful of workshops where I’ve taught knitting. As with any skills, people’s ability and speed with which they learn can vary tremendously but it’s hugely fulfilling when someone takes to it and enjoys it.

To the crafters who come across this post, do you know about the site Ravelry, for knitters, crocheters and fibre makers?

J is for "Jekyll"

Thu, 11 Apr 2024 00:00:00 +0000

This website is simple - both in terms of functionality and design. This is intentional. I want to be able to focus on the content. Other than the comment system (which is hosted separately), there isn’t even a database. I write the posts in a plain text editor.

I use the site-generator Jekyll, and this is the “J” mentioned in the title of today’s post. I’m aware of a similar piece of software called Hugo. Both work in this space, but Jekyll has existed for a little longer (since 2008) than Hugo (since 2013).

I write in markdown the majority of the time (with .html in a couple of cases), and Jekyll does the transformation into static files. I then upload the result to my webserver.

There’s therefore no user accounts, no password to be guessed resulting in a possible breach.

Those of you with blogs, do you use a fully fledged CMS or a lightweight templating system. Is it one you wrote or is it an existing product?

I is for "Integrity"

Wed, 10 Apr 2024 00:00:00 +0000

My sense of integrity is important to me, including in the workplace. I’ve a few stories but two scenarios in particular come to mind.

The first is: I was working for a government agency which provides funding to various organisations. I was speaking with a family member about this work and was told “Oh, yes, my employer received money from them”. I didn’t work anywhere near the funding committee or decision process; I was working on software at the time. But I still immediately contacted my superiors and alerted them to this possible conflict of interest. It was all cleared without problem.

The second story is from over a decade ago. I was employed, but looking for a new job. I found a recruiter and they described a role which seemed interesting. As they gave more information, I had a suspicion who the client might be. I queried with the recruiter and my guess was correct. I explained “I’m going to have to withdraw because they’re a client”. I wasn’t actively working on that client project. Additionally I vaguely recalled that my contract didn’t have any prohibitive clauses, but it still could have been a possibly awkward situation later on when already established in that potential new company.

A colleague at a more recent employer used to refer to me as his ethical canary: “What would Matt think about decision X?”. I’m sad that we no longer work together but I’m also hugely thankful as he probably shaped me the most of all of my employers.

H is for "Handwriting"

Tue, 09 Apr 2024 00:00:00 +0000

Like many of my peers in tech, my handwriting is pretty awful. For many years whilst I was in school I had private handwriting lessons. It did help a little but not until rather late. Until that time I’d been using a laptop. My reading was unaffected, as was my speech and reading comprehension. Indeed, I was to some exgent hyperlexic and had a broad vocabulary. I remember, vaguely, the first time I beat my mum at scrabble.

Changing school or moving years was sometimes difficult because each new set of teachers had to be told about my accommodations (supported by the local education authority). I remember with frustration one teacher who not only initially denied this accommodation but also wrote (and I quote): “This is hopeless. Re-write this legibly”. It took a bit of sorting out but it naturally soured the relationship.

I do write now (in addition to using a keybaord) and my writing is mostly legible (at least to me). But a few techniques help me to me more legible when I’m writing for others. The first is to print rather than write joined up (cursive, for the US folk). The second is that I will use more than one colour (e.g. if presenting information in a table then I’ll draw the table in, say, black, and then write the table entries in, say, blue. A final technique, though one I use sparingly, is to write in all caps. Typically I’ll do this for a couple of words, as it is difficult to read a block text in all caps.

Though I carry a phone around with me (who doesn’t‽), I also often carry around a notebook and use this in preference to my phone for text of more than a few words. I’m sure this will surprise my former school teachers were they to learn this. (I’m not in touch with any – I was in school almost 30 years ago)

I may have been slow on the up-take but I’m glad that I was both supported in alternatives and that I am capable as an adult.

With the proliferation of keyboard use in the modern age, how is your handwriting?

CSS Naked Day

Tue, 09 Apr 2024 00:00:00 +0000

According to this, April 9th is CSS Naked Day! Quoting from the website: “The idea behind CSS Naked Day is to promote web standards. Plain and simple.”

Today, therefore, I’ve removed the CSS declarations from this site (with one tiny exception that appears on one page. Can you find it?)

A bit of silliness. Why not‽

G is for 'Games'

Mon, 08 Apr 2024 00:00:00 +0000

I like to play games - of the boardgame and cardgame variety.

A few years ago I came across a flowchart which can help one to decide what game to play within a group based on different factors and preferences. I’m not sure of the exact origin, but the earliest source I’ve found is from January 2015: https://www.reddit.com/r/gaming/comments/2s4355/choosing_the_perfect_board_game/

The games I play aren’t limited to what’s on this flow chart. I typically prefer the strategy and team based games, in addition to the co-operative where players have a common shared goal rather than in competition against one another. Naturally, the games I like to play change over time but some of my current favourites are:

Hanabi (2010)

A co-op game where players must play as many cards in sequence. This game is unusual in that one plays without seeing one’s own cards; instead players hold their hands with cards facing outwards.

Pictures (2019)

I have very little artistic skill but somehow this game somehow captures my interest. There are several photos and each player has different materials with which they have to create a representation of one of the pictures. The other players must guess which piece is being created.

Turing Machine (2022)

Unrelated to Alan Turing, this game is a pure logic game. It can be played co-operatively, competitively or solo. The goal is to work out a 3 digit secret code using clue cards (e.g. “Is the first digit even”) and through making guesses to the code and performing these tests, the code can eventually be deduced.

Bridge (1925?)

The exact date of origin is hard to pin down. This is the famous 4-player game originally invented in the UK but now played worldwide. A few years ago (before Covid) I used to play a lot of bridge but I don’t play much now. I never really got on with formal bridge clubs - I’ve tried a couple. Time is a finite resource but I have been thinking recently that I’d like to start playing again.

Hare and Tortoise (1973)

This is a race game, but rather than based on luck of a die, one can choose how quickly or slowly to move. Faster moves are more expensive and slower moves are cheaper. Move too quickly and one runs the risk of running out of the game’s currency (carrots).

And you?

What games are your favourites?

F is for 'Felines'

Sat, 06 Apr 2024 00:00:00 +0000

Low effort post since I'm visiting family today. Today's letter is 'F' and the word is "Felines".

As a child I grew up with cats but lived in rented accommodation from when I was at university until 2006. A few years later (2010) I was thinking that I'd like to have a cat again and, being in my own place, this was now a possibility.

I went to my local cat shelter and after a home check adoption was approved. Freddie has been with me ever since - he's now 14.

Ginger cat outstretched on a grey sofa with belly exposed

Photo taken from above of a ginger cat sitting on a grey sofa, looking directly up into the camera

Ginger cat looking into the camera with his tongue sticking out. A classic mlem

A ginger cat rubbing his nose on the nose of a bespectacled human wearing can-style headphones. There is slight motion blur

I've been working in IT for in excess of 20 years. I'm still happy in the industry but know that it won't be for ever. Even if in a volunteer capacity after retirment, I want at some point to work in a big cat sanctuary looking after cheetahs, pumas, lions etc.

It feels odd to be thinking about plans 20 years in the future. But even if I don't do this, I know I won't be bored. What's the most distant future plan you have?

E is for 'Events'

Fri, 05 Apr 2024 00:00:00 +0000

The series “A-to-Z Blogging Challenge” continues with today’s blog entry on the topic of Events. I have a number of varied hobbies and it’s fun to spend time with people to share in these interests.

Two events worthy of mention are MatjsJam and EMF.

MathsJam is a medium-sized event (a little over 100 people) filled with talks and games on mathematics and taking place every year. Some of it serious, much of it less so but pretty much all engaging. There’s even a baking competition. Thinking to yesterday’s post on desserts I should probably take part - so far I’ve not got around to it. Being a smaller event compared to some of the others I go to, this is also the event where I gave some of my earliest talks. So it has a warm place in my heart for that reason too.

What about EMF? The name is a pun - with the letters standing for Electro Magnetic Field. The event is a geek festival taking place over a long weekend. 3000-odd people camping in a field with letures, workshops, presentations etc. The official EMF Camp website gives more information.

It takes place every other year (on the even years) and this year is in Eastnor, Herefordshire. It has sold out, so unless you’ve got tickets already (or decide to submit a talk and that talk is accepted) you’ll need to wait a couple of years to take part.

I suspect at least one my readers is the kind of person to go to one or even both of these. But for those of you with different interests, what events do you like to go to?

D is for 'Dessert'

Thu, 04 Apr 2024 00:00:00 +0000

Living alone I don’t bake desserts all that often because I’ll be tempted to eat them at an unhealthy rate! But I do enjoy baking for a social event, whether I’m hosting or attending.

I find it frustrating that recipe sites often contain information extraneous to the recipe because when I’m working to a recipe, I’m typically using my phone which has a small display and so screen space is at a premium. That said, I do understand why it’s done this way. (Broadly speaking: Recipes can’t be copyrighted but the associated narrative can be). I recently discovered a couple of tools which scan recipe sites and extract the relevant sections. They are Just The Recipe and cooked wiki.

I’ve a few recipes that I like to cook semi-often, some of which are my own and some of which are recipes from elsewhere. Focussing on today’s theme, I’ve selected some desserts to showcase here.

Chocolate orange cake

Not my recipe, but this one is probably my favourite. It is a huge labour of love, mainly because I prepare it over a couple of evenings. The first evening I’ll candy the orange peel and then the next evening is for the cake itself. But the benefit is that I then have candied orange peel to snack on.

The chocolate sauce

The candied orange peel drying

The candied orange peel ready to eat

The finished cake

Chewy chocolate chip cookies This one is also not mine, again from the BBC Good Food site.

Also on my website is an uncategorised list of my own recipes . These are written without the fluff I lamented above. There’ll be no need to use one of the stripping tools I mentioned. One in particular I’d like to highlight is a lemon meringue pie. Here’s some I made during the first lockdown. It’s a good way to re-use those Gu pots.

What’s your favourite dish to cook? (Whether dessert or not)

C is for 'Caddy'

Wed, 03 Apr 2024 00:00:00 +0000

A few months ago a friend introduced me to Caddy, a webserver. Why yet another one? Well, ultimately this one fits my needs better.

The huge benefit, compared to other webservers such as Apache httpd or nginx is that it supports native auto generation and renewal of SSL certs (either self-signed or via Let’s Encrypt). I know that LE certs can be automatically generated by other means, but having it done in-band is the big bonus for me.

My own site (the one you’re reading right now) has been hosted via Caddy since January 2023. The SSL site of things is so automated by Caddy, that the only configuration I need is

peperell.com, peperell.co.uk {
  log {
    output file /var/log/caddy/blog.json {
      role_keep 10
    }
    format json
  }

  handle {
    root * /var/www/blog/
    file_server
  }
}

The mere existence of a domain name (or several domains) to configure a site is enough for caddy to know to generate (and auto-renew) its SSL cert. No extra cronjobs needed.

Other than that, it does pretty much what one would expect from a webserver, including ability to proxy on to lower tiers (e.g. APIs, unix sockets, HA proxy etc). One extra thing that it provides but which I don’t use is an API, via which one can add new site configuration at run time. My raspberry pi running this site (and a couple of internal house-related services) does not need this functionality but it’s good to know it exists should I encounter a commercial project of a suitable size and shape for this to be relevant.

Have you ever use Caddy, or does this post tempt you to give it a try?

B is for 'Bellringing'

Tue, 02 Apr 2024 00:00:00 +0000

One of the numerous hobbies I have is bellringing. The -ology word for it is campanology but virtually no one in the community calls it that. A rather obscure name for it, and one seldom used, is The Excercise. But what’s it all about?

It’s mostly an English thing, but does exist elsewhere in places where there’s been Church of England influence. Bells are used, primarily as a call to worship. But my interest lies in the musical and mathematical theory side of things, along with team work and mental endurance. It requires physical motion, yes, but not an undue amount of effort. The mechanics of wheels and pulleys makes it possible for even a person of my limited strength to run bells exceeding one ton.

Musically, the bells are pretty much always tuned to a major scale starting with the highest note and ending with the lowest note. In each unit of time (which we usually call a row), each bell will ring once and once only. For mechanical reasons (conservation of momentum, for those who care) from one row to the next a bell can be in the same position, one position earlier, or one position later.

We change the order in a pre-determined pattern that must be committed to memory. These patterns can be simple or complex, and it’s this variety and the challenge of memorising the patterns (which we call methods) that keeps my interest.

What got me into it? New Year’s Day 1996 I was helping friend clean out his loft and got distracted, having found one of his theory books. I expressed an interest and he invited me to come along to the practice night, which happened to be the following evening.

When I first moved to North London, I heard through the community that the bells of the local church were not being regularly rung and the church had members of the congregation who were interested in learning. So I made myself known and have been teaching and ringing there for the last 18 years.

If you’re interested to find out more and you live near a church then turn up on a practice night and they’ll almost certainly welcome you. It’s a very open communinity and one does not need to be a person of faith to take part, though one must always keep in mind the reason for ringing existing, is a religous one.

A is for 'Alphabet'

Mon, 01 Apr 2024 00:00:00 +0000

I was thinking recently that I've let my blog languish for a long time (having checked, it's over two years). Apt in timing, yesterday I discovered the A-Z Blogging Challenge. It feels like the sort of thing I can use to help pick up activity. But then again, I don't want it to be forced - so who knows how long I'll keep it up.

As many of my readers know, I like words and language, and foreign lanuages too. But today's post is on the English alphabet. Or rather on a book themed on the English Alphabet. The title of this book is The Ultimate Alphabet. Though the book contains prose, the main content is a series hand-painted pictures each themed on a single letter of the alphabet. The wonderous thing about this book is that it is suitable for both children and adults. There are simple words such as "angel" (findable by even young kids), flags for countries beginning with A and for a real subtlety there is the maritime signal flag for the letter A.

I first encountered this book as a child in the late 80s (soon after it was published) and have bought a copy of it myself since and as I've learned more complexities about the world, there are still relevant things to be found and I get plenty of use out of it. I also occasionally buy it as a present for people.

The same artist drew each of the pictures - his name being Mike Wilks. The small amount of prose serves as an explanation for how the pictures were developed, along with a per-picture introduction giving sample words and how many items are to be found.

Here is the picture for the letter 'A'. Can you find an angel, an aquarium, the Australian flag, an awning, and others? There are 306 things to find.

For convenience, here's a link to the book's details: https://isbnsearch.org/isbn/9780805000764

Some wordle variants I've encountered

Sun, 20 Feb 2022 00:00:00 +0000

Over the past few weeks a word puzzle called Wordle has become popular on social media, or at least on Twitter/X, the only social media forum in which I have any kind of active presence.

For those not familiar with the game, the goal is to determine the 5-letter target word by making guesses at that word and being told which of those letters appear in the target and of those letters which are in the same place as the guess. Sounds simlpe? There’s a limit of 6 guesses. It’s not just a case of being lucky.

I played several variants before I tried the original. In this post, I’ll share a few variants that I’ve encountered. They mostly have the same gameplay as the original but I’ll point out some differences where relevant.

Lewdle

Like the original, but the target 5-letter word is in some way rude, lewd or offensive. N.B. Some of the words are slurs.

Sweardle

As above, but 4-letter words.

Hello Wordl

Play the original game but with a word length of anywhere between 4 and 11 letters.

Worldle

Try to identify the country based on its shape. The clue given is the distance and direction to the target.

Mathler

Find the hidden calculation that equals the target number. Uses digits 0-9 and the operators +, -, / and *.

If you can solve this then there’s a harder version to try.

Nerdle

A harder version of Mathler - find the mathematical equation. You don’t even know how many digits are to be found in the answer.

Dordle

Like the original but try to solve two words at once.

Quordle

Like Dordle but try to find four words at once.

Passwordle

A much harder version of the game. The possible characters a not limited to letters but include numbers and symbols too. At least the letters are case-insensitive!

Passwordle (a different one)

The first one I consider to be genuinely infeasible for a human to solve. Enter the password and be shown its SHA256 hash. You’re then shown which of those hash digits are found in the target word’s hash. (A hash, amongst other things, can be used to store passwords securely). More details can be found in the Wikipedia article on hashing but for a simple explanation: A tiny change in the input (the guess made) results in a huge change in the output (the ‘clues’ given).

Vortjo

Perhaps the most obvious adaptation is to have a foreign language dictionary. This variant has an Esperanto wordlist. Small disclaimer: Vortlo is written by a friend of mine.

A version in Google Docs, compatible with Excel

Made by Twitter user akshathkilla.

Numble

A version made in the Geometry modelling software Geogebra.

Absurdle

The first variant I found. This is a contrarian version of the game; it continually changes the target word to avoid your guesses (whilst still giving non-contradictory clues). This one doesn’t have a fixed word on a given day so you can replay it to your heart’s content.

Chordle

Try to determine the correct form of a musical chord. It’ll be some kind of 7th and the challenge is to work out the notes and what inversion the chord is in.

Any more?

Have you encountered any that I’ve not mentioned here? Foreign language variants are equally welcome as gameplay variants.

How to avoid a low bus number

Thu, 16 Dec 2021 00:00:00 +0000

This post is nothing to do with public transportation. Instead I refer to the concept called the Bus Number, which is a way to quantify organisation risk. It refers to the minimum number of people in an organisation that need to be hit by a bus for there to be lost knowledge.

A less jarring version of this term is the lottery factor; referring to people who would leave the organisation were they to win the lottery. (The risk would also apply when people take holidays or go on parental leave).

In a perfect world, all knowledge would be captured in written form such that it wouldn’t be lost in the event of people moving on but reality doesn’t work that way. I remember once place I worked which had a large ish team but also had a large variety of duties. In the early stages that worked okay, but as the team and variety increased it didn’t scale. Our bus number was seldom 1, except perhaps during the initial stages of taking on that duty. As a result we became adept at distributing this knowledge within the team. We soon realised that trying to share knowledge of all tasks to all members wouldn’t work. Instead, I proposed that we would have each member of the team be the subject matter expert for a different component. They would not only be in charge of discovering and documenting the relevant intricacies, but also ensuring that their particular knowledge area does not have a bus factor of 1. So they’d also run training sessions to share the relevant information.

The team was made up of 3 senior engineers (me being one) but was junior heavy. By distributing the workload, this not only freed me up for managerial duties, but also empowered and engaged other team members. What better way than to be told “I trust you to make this happen”‽

A second technique, relevant to the documentation side of things¹, was that we’d have one person write it, but then asked that the documentation be validated by having a newcomer attempt to use it. The SME would answer questions, but take a passive role such that the documentation – and not the SME – would be the guide. The newcomer can then amend documentation filling in gaps and ambiguities. Having the ability (and permission!) to amend procedural documentation was key here. ²

All in all at the time of my departure we had in the order of 20 topic areas spread across 10 SMEs, though basic knowledge of the majority of tasks was able to percolate to most of the team. But 100% across 100% would be an impossible and demotivating goal. Don’t set yourself or your team up for failure. Success is possible but don’t equate this with perfection.

I have a post in the back of my mind on the topic of documentation too. ↩
Hm. Maybe this would be a good addition for my Questions to ask a company list; how painful is the changing of procedures and/or documentation? ↩

Tracking personal expenses

Wed, 15 Dec 2021 00:00:00 +0000

For the past year or two I’ve been tracking my finances using a Plain text system called beancount. Following on from yesterday’s post on online grocery shopping, I was thinking about the sorts of graphs I can extract from my Beancount data.

For the graphs below I’ve left the dates present but removed the numbers. The shape is what tells the story.

My grocery expenditure, with bars for each quarter of the past almost 2 years:

Relatively stable over the past year or so, with a tail-off in Q3 of this year. I can’t think of a reason why this might have changed; if anything I’d expect it to have increased a little (see my explanation for the next graph)

Deliveroo, grouped by month from March of this year

It’s interesting to see my mood reflected in this graph. May was a rather difficult time for me for a few reasons(thankfully okay now). So I was cooking for myself a lot less and instead ordering food in. And in July I recall cooking a lot of my own food. Although as I write this,I recall that I also had some deliveroo vouchers as gift. So that wasn’t a direct expense to me and therefore isn’t represented in this graph.

Council Tax, grouped by month from several years ago.

I was aware of this change (because of the annual statements the council send me) but it’s pleasing to see one bill reducing, in a world which seems to have every other bill increasing.

Pet expenses, grouped by year

This is the sharpest rising expense. Partly it’s insurance, and partly it’s the out-of-insurance expenses incurred as he has more health problems. But the main increase is the specialist food he needs.

I’ll finish with an obligatory picture of Freddie. He’s definitely worth it!

Thoughts after a year of online grocery shopping

Tue, 14 Dec 2021 00:00:00 +0000

Last October I started grocery shopping online. This worked very well for around 9 months. Product accuracy was high and it was very easy to request a refund (at point of delivery) for unwanted substitutions or for items extremely close to expiry. But even still, the vast majority of items were in stock and so things were rarely a problem.

But around August of this year, things took a downturn. Since then almost every single order I’ve placed has had items missing due to low stock levels. Almost without exception, these items have all been in the same category: prepared meals. Not the greatest culinary choice, sure, but I value the convenience (and besides, though price is of some concern, I’m not aiming to optimise for cost).

As a result it means I’ve had to make alternative arrangements for 5-ish meals per week. I mind less if this is something I would have eaten in the evening, but when working from home it does have an impact on lunch time arrangements.

I am aware that the orders are picked by hand on the day of delivery and therefore subject to local store stock levels. Furthermore, missing items are refunded. But it’s very poor from a user experience point of view.

Since the pandemic, I have occasionally walked to the supermarket (typically whilst needing to run other errands such as visiting a bank) and have noticed the same items missing from shelves. So to be clear: the items missing from my orders do appear to be due to genuine stock issues. I also understand that it’s an unreasonable expectation that the pickers go to other stores in order to minimise the number of missing items. But what’s less clear is why this has been the case for almost 4 months, and why this has changed when it was fine before this period.

A little over 10 years ago I worked for an ecommerce company and that had stock-level awareness. It would give warnings when items were low in stock, not merely when the item was completely out. Technically this would be possible because they know what my delviery address is and so can estimate what the local picking store would be. But that said, I don’t know what use it would be to raise a complaint. This isn’t a problem with a specific order; it’s a systemic issue and thus harder to fix but also it seems rather self-centered. The process as a whole seems to work; perhaps it’s just my choice of product. I don’t know how common my usage pattern is.

About a month ago there was an interesting problem that happened one evening when my order was delivered. The crate presented to me contained none of the items I normally ordered; it was someone else’s. The delivery driver was on the verge of re-funding the entire order and in our conversation he made an off-hand comment that he had only 5 orders remaining to deliver. I made the suggetion of checking whether it had been switched with someone else’s. The driver checked the van and found another crate which had the same identifying number. It seems unlikely that crates are numbered on a per-van basis, but with this being the case the chance of the same number being issued and then ending up on the same van, seems very low. But this is what happened, and my items were located. (Presumably the other household, being one of the remaining 5, correctly received their crate which had first been presented to me). So this issue too was solved

My supermarket of choice was Sainsbury’s but I’ve been thinking over recent weeks that I might give a differnt brand a try, to see how it compares with missing items. Do any of you get groceries online reguarly, and if so how common are subsitutions or out-of-stock issues?

Train operating companies typically publish punctuality and reliability figures, as a percentage of on-time or services ran. I wonder what supermarket percentages showing “Orders met without subsitution”, both aggregated (either company wide or perhaps per store) and for me as a user, might be. There’s bound to be a couple of data geeks who read this. Can you think of any other figures that might be interesting, were supermarkets to collect them? (and ideally publish them too)

Oneword - Writing exercises

Mon, 13 Dec 2021 00:00:00 +0000

Another nostaligiac post!

In the past I used to write a lot of prose. My dayjob has occasional call for it now but my time is normally spent writing code and mentoring people. That lack is part of my motivation for wanting to kickstart this blog. I’ve no idea how long I’ll keep this up. Maybe just until the 25th.

Most of what I write (and wrote, for that matter) is report or tutorial in nature with little inventive writing. I had a memory last night of a writing prompt technique that I encountered perhaps 15 or 20 years ago. This being “one word” - a single word is presented and the task is to write, unrefined, unplanned and unedited on that topic for 60 seconds.

I was curious and searched online and found a website called, precitably enough, OneWord which offers such suggestions. A new word is offered every few days. A fun thing about this site (and any others like it) are that users’ submissions are collated so you can read what others wrote (whether or not you make a submission for that particular word).

I don’t have the right sort of creative skills to write stories with plot and character, but this is the closest I get to creative writing.

Science-themed video calls with kids

Sun, 12 Dec 2021 00:00:00 +0000

My niblings are aged 5 and 7. Due to COVID I’ve not been travelling to see broader family much but I have had a few videocalls with them. Because they both have an interest in science and doing things, I’ve come up with a few ideas for activities. On one such occasion I talked them through making a lemon drizzle cake (my own recipe, available here for those interested). On another occasion I demonstrated to them Subtractive colour mixing though I didn’t call it that. I think I used probably the wording “making new colours”. That afternoon is what this post is about.

I started the call wearing a labcoat and protective eyewear¹ which definitely got their attention! [I don’t have a photo of this available, I shall try to obtain one and update this post] I started them off by first getting them to find 3 big glasses and 3 small glasses, and some red, yellow and blue liquid food dye (which I’d prep’d the parents to buy a few days beforehand). Exactly what size you use doesn’t matter; it’s merely a convenience.

In case you can’t guess where this is going, I had them almost fill each of the 3 big glasses with water. Then into one glass they’d add a few drops of blue food dye and mix it up, yellow into the next and stir that, and red into the last big glass.

It looks like the following

I then asked them what colour they thought we’d get if we mixed the red and the blue. The guess doesn’t matter - it’s to get them thinking about colours. So we take one of the little empty glasses and pout into it a little of the red water and a little of the blue water!

We then did the same again but this time with the yellow and the blue.

You can use other combinations, and also different ratios.

If you’re reading this then feel free to replicate this with your own kids, or do it with kids over a video call as I did.

Not overly taxing, but very fun and very engaging.

A word of advice though: Make sure that they have careful supervision. I’m not responsible if they start trying to handle the dyes directly instead of using the diluted colours!

Am I unusual in having my own such PPE at home? ↩

Introducing the microCOVID

Sat, 11 Dec 2021 00:00:00 +0000

A couple of months ago I discovered the website for the so-named microCOVID project. What’s a microcovid? It’s a unit by which one can estimate risk of contracting covid based on one’s lifestyle - things such as how one travels to work, with how many people, how far apart and so on. One can enter one’s risk appetite and the website will show whether the indicated activities put someone above or below that threshold. It’s still a new project and the science is still imperfect but I’ve learned a bit from it about relative risks.

That website made me think of af few other unusual units (by which I mean non-SI) -derived) that I’ve encountered. The first one that comes to mind (and from which I suspect the microCOVID was named) is the micromort, a measure of mortality risk for certain events, such as travelling (by whatever mode of transport), smoking, skydiving, living in a city etc.

Next up there’s the Helen (with the milli-Helen being the most common multiple), based on the character from Homer’s Iliad, Helen of Troy whose face is said to be so beatiful that it launched a thousand ships.

Another one relevant to the pandemic (or perhaps more to the lockdowns) is the beard-second, which as its name suggests, is how fast the average beard grows in one second. The exact value differs (it’s not an SI unit. afterall) depending on which defintion ones uses, but 5nm and 10nm are the typical values.

There’s a whole load of them on Wikipedia falling into two categories Humorous and Unusual. There’s little point in me reproducing them all here but it makes for a fun distraction - if a foray into metrology is your sort of thing.

Lamenting the downfall of rot13

Fri, 10 Dec 2021 00:00:00 +0000

When I first gained access to the Internet in the mid 90s, the world was a very different place. The web was still in its infancy. Usenet was still very popular. Online culture was very much focussed on sharing of information rather than the skew towards collection of data that seems to proliferate today.

One of the things I remember is ROT13 - a simple way that text would be hidden, but done so in an intentionally simple way. It was simple enough that with a little practice one could even decode it by eye, though it was rare for people to actually do that; typically email clients and newsclients would have ROT13 functionality built in.

It worked as follows: All non-letter characters (such as punctuation, spaces, new lines) would be unchanged. But every letter would be advanced 13 spaces in the alphabet, remaining as upper case or lower case, depending on what they were originally.

The real elegance of ROT13 is that the same calculation can be used to both “encode” and “decode” the text. In other words, there’s no need to know whether a text segment is coded or in plaintext before applying the transformation. It was simple to implement.

Some examples: A would become N, B would become O, C would become P etc, but also note that N would become A (since there isn’t a letter after Z). The same thing would apply for the lower case letters.

Click on this text to see the effect. Guvf grkg jnf uvqqra ohg vf abj erirnyrq.

Here’s another example. The English alphabet in both upper case and lower case:

ABCDE FGHIJ KLMNO PQRST UVWXYZ
abcde fghij klmno pqrst uvwxyz

Why was such a thing used? The common use cases were

to hide the punchline of a joke
to prevent giving a spoiler about a TV episode or movie
to hide the answer to a puzzle

It seems as though ROT13 has fallen into disuse these days. It also never really became widely used on the web. I’m not sure why either of these might be so, but I still have a soft spot for ROT13. One possibility that as Javascript became more widely adopted on the web, designers used more aesthetically pleasing techniques such as blurring or black bars, or perhaps having the solution behind second page load.

I’m not trying to make a point here. Instead, I merely felt indulging in a little nostalgia.

I’ll finish with a joke: When is a car no longer a car?

Jura vg gheaf vagb n fvqr fgerrg.

How to interview a company

Thu, 09 Dec 2021 00:00:00 +0000

I see lots of advice about how for candidates on how to prepare for interviews, covering both algorithm specifics (e.g. “how to crack the technical interview”), and how to prepare for common non-technical questions (“Why do you want to work here?”, “Where do you see yourself in five years time?” and the like). But interviews are a two way street. They’re not solely for the company to decide whether they want to hire you or not – they’re also to help you decide whether you want to work for the company. I see much less advice on how the candidate can address that second aspect.

I’m hugely lucky to be in a job I enjoy. This hasn’t always been the case but I think what really helped me to learn how to interview potiential employers was a bumpy patch in my employment history in 2014. Around and since that time, I’ve collected some questions that have really helped me to gain insight into what working for a company might be like.

Here are the questions, along with some thoughts on why these are useful

When top performers leave this position, why do they leave, and where do they go?

If a company doesn’t know, it likely means that they’re not conducting exit interviews, meaning that they have little interest in working to retain talent. Another possibility is that people are leaving because they’re disgruntled¹, though it’s understandable that the company won’t admit this, and thus an answer of “We don’t know” does not necessarily mean something negative. But if exit interviews are conducted then there’s often something the company can learn from that.

What one thing missing from my resume concerns you the most?

I use this one as a barometer to help gauge how well I’m faring compared to other candidates seen so far. I’ve worked with in excess of 40 different languages and syntaxes in my working life² so I’m able to demonstrate that I can learn new ones. This question, therefore, is useful to see how the company views differing skill sets and their willingness to offer training opportunities.

Conversely, with there being so many other good candidates, what made you seek me out?

This is the other side of the coin for the question above, but it can be of use.

Where have the best ideas come from in the last two years?

This is a great one to learn about company culture as set by the executive suite (or C-levels, or whatever term is used for the organisation in question). If all ideas come from The Powers That Be and none have come from lower tiers, including Individual Contributers, then it’s possible that the organisation is an inflexible or dogmatic one.

With learning curves in IT typically steep, productivity from day 1 is rare. If I’m hired, what is one thing I’ll need to deliver in the first 90 days in order to be considered successful?

This is a great one that I use to get an inkling into what it might be like working for that person. A list of projects or tasks is useful, but as the candidate I can also get a sense of whether the expectations are realistic or not.

Burnout is a common problem in IT - how do you help staff not to?

More than once I’ve quit my job before having another one lined up, and once I didn’t but probably should have. I realise this is hugely privileged of me to be able to do so. IT has a reputation for providing a good income, and I’m thankful for that. But workplaces shouldn’t make us feel bad. Money should not need to be a sweetener to be subject to duress for 40-odd hours a week.

When was your last production outage and what was it? How green is the monitoring and how common are false alerts? What is your oncall rota like?

I use these questions to help gain an insight into how high-pressured the workplace is. If there’s always something breaking and it needs to be fixed immediately, then the chance of burn out increases and staff churn increases. Plus due to lack of rest (because of gruelling on-call), tired people are at increased risk of making mistakes, and this too leads to more outages.

What’s the diversity of the engineering team? (not just that of the company in general)

Tech has an appalling reputation for diversity with women, trans people and non-white people woefully under represented. I’ve had some very lazy answers to this question when asking it but the best answer I received was supplemented with activities that the company is undertaking to combat the disparity. That’s the place I accepted the offer from almost 7 years ago, and the place I still work at today. [Edit in 2024: I no longer work at this place]

I’d be curious to know whether you have any of your own questions of this kind. I’m particularly interested on hearing from people who work in tech-adjacent or even unrelated fields.

due to working conditions, pay, lack of progression, or whatever else ↩
I’m not claiming to be an expert in all of them! ↩

Invisible characters

Wed, 08 Dec 2021 00:00:00 +0000

A few days ago I wrote about an issue which was particularly tricky to debug. Here’s another such story from a few months ago.

Context

At this client, we support the websites used by members of the public when applying for or renewing passports online, along with the associated applications, systems, and third-party providers needed for these.

Problem

A specific customer’s passport application was being rejected (due to failure of automated validation checks) and appeared in one of our error queues with the message notifying us that the phone number was missing. We’re used to that; it happens once in a while. The engineer working on the problem asked the customer service team to obtain the applicant’s phone number, and then manually updated the record by copying and pasting it from the email containing it received from the customer service team.

The engineer then, as was established procedure for amending records in this way, resubmitted the application from the error queue. It was again rejected and landed back in the error queue. This time the message notified us that the phone number was invalid.

An initial visual inspection showed nothing out of the ordinary.

Investigation

After performing basic investigation (i.e. a visual inspection of the phone number), our team’s next port of call is typically the application developers. They confirmed to us that there is a regular expression on the front end when users make applications in order to prevent mal-formed phone numbers from being introduced.

At this point, the engineer sought advice from a senior engineer - and this is where I joined the investigation.

Inspecting visually (i.e. not using the website) I again confirmed that the phone number matched the regular expression. A second set of eyes helps, plus performing these checks helps the senior engineer to get up to speed with the details of the problem.

I then clarified the origin of the phone number and realised that it never came from the website (being instead entered manually), and formed the conclusion that the website regular expression would never play a role. It was at this point I suspected Unicode – somewhere between luck and suspicion borne from experience.

To prove or disprove this hypothesis, I then used the tool hexdump (with the -C option) to view the byte-level representation of the number. My expectation was that all bytes would be in the hex range 30-39 (digits 0-9), or 2b (a plus, +, for international numbers), and 20 (a space). Instead we saw

echo +6421333333​ | hexdump -C
00000000  2b 36 34 32 31 33 33 33  33 33 33 e2 80 8b 0a     |+6421333333....|
0000000f

(number masked for privacy reasons)

So there’s a stray character. It would be a simple matter of deleting the erroneous character and re-amending the record, but having gone this far it’s both prudent and intellectually/forensically interesting to identify the nature of the malformation.

My initial interpretation was 808b, which is an invalid byte sequence in UTF-8 encoding. After a bit of circling around I then noticed I’d missed a byte out and the full value was e2808b. A quick websearch for this sequence identified this as being a zero-width space character.

Conclusion - this was indeed a funky Unicode character.

Summary of findings

The format of the phone number was mangled in a way invisible to the eye: it contained a Unicode character, specifically a zero-width space. This was brought along when the phone number was copy&pasted from the original email from the customer service team. To be clear: this is no fault of the original engineer, nor of the customer service team who sent us the document in this format. It’s an artifact of how office-based text editors such as Word sometimes present text, and end users should not be expected to know about this behaviour.

Recommended actions

I forget the reason why there was a missing phone number despite the website having validation. Perhaps the validation had only recently been added. That detail isn’t important to the story but I felt I should cover it here in case anyone wonders. But with that validation in place, no further application updates are needed because the issue should not occur again.

Manual updates to records are rare; we see approximately 2 a month. We already perform input validation in the form of a regular expression when we get the phone number directly from the user via the website. The receiving systems to which we send applications (e.g. for identity checking or for referee validation) also have error detection - this being the reason that periodically requests end up in the error queue for manual investigation. The effort required to make an application change to detect these and give a more specific error is probably not counteracted with the amount of time it would save, based on this low occurence rate.

There might be some merit in teaching the application support team about some likely culprits when Unicode extended characters cause problems. (Non-breaking spaces and smart quotes also being occasionally encountered)

One could make the suggestion of typing the number by hand rather than risk introducing formatting errors such as this, but the risk of introducing a typographical, transpositional or some other error is far too great.

Addendum: But what is a zero-width space?

It’s a way to indicate that a line-break may appear at a given point in a chunk of text, but that if the column of text is sufficiently wide that the line break is not necessary.

To demonstrate the effect, in this paragraph of text I’ve writte out numbers in English but separated with ZWSPs. Resize this window by making it narrow to see it in action.

OneTwoThreeFourFiveSixSevenEightNineTenElevenTwelveThirteenFourteenFifteenSixteenSeventeenEighteenNineteenTwenty.

It doesn’t make sense to have linebreaks in a phone number, but the document that was sent to us was a text document, and the text editor used (Word, I think) doesn’t know what it means to be a phone number, so it didn’t know not to add one. ZWSPs are invisible to the naked eye, on screen, and when printed so it only matters in very subtle cases such as this – far too nuanced for something like Word to care about having intelligent heurestics to make the determination of whether to add one or not!

Something old, something new

Tue, 07 Dec 2021 00:00:00 +0000

Don’t name things new or old!

Systems are replaced in IT often (though perhaps not as often as we might sometimes like or that they deserve!). Some of the time, these old and new systems will co-exist. Given enough time, such a system might even be replaced a second time, maybe with some of the same staff members present and even by the same staff members.

When in the middle of such a transition, how can the old and new systems¹ be distinguished?

The example I’ll focus on here is a print spooler but it need not be limited to that.

A separate name might well be needed if someone consuming these print services (or whatever) cannot automatically select the correct destination. Calling the print queue printserver makes sense. But when replacing it, giving the new system the name newprintserver though in the short term it allows the two to be distinguished, will eventually lead to confusion once the old one gets removed.

Let’s assume for the moment that the name newprintserver has been decided upon, and that it’s sufficiently far in the future that the old one has been removed. Will the newprintserver system be renamed? If so then this can cause another series of disruption whilst configuration is updated.

Maybe it won’t be renamed from newprintserver once the old system has been removed, but this has a couple of drawbacks:

newcomers(!) joining the organisation might wonder whether the old one is still in service (for example as a backup) or maybe even that the new one is still in a trial period and therefore not ready for full use
What happens when the newprintserver system is itself replaced? Will the next one be called newnewprintserver? This is hyperbole, I admit, but hopefully this makes the point.

The other approach is to use the name printserver for the new one and to rename the old one to be oldprintserver. Because when new systems are introduced they often have teething problems, it’s usually the case that the two systems co-exist. But by using the prefix old- might make people think it’s already due to be phased out and is deprecated. But it also causes problems because in the event of the new system not yet fully working, people are guaranteed to have to make a configuration change to go back to the original printer which they knew to be working.

With the goal being to migrate people onto the new system, this is perhaps the least disruptive version of the previous scenerio, but still might result in the double renaming if there are teething problems with the new system which necessitates people to use the old one for a while.

Better to have the one rename, and to do so once it’s been proven. As much as one tries to perform testing before launching a new service, some things will be missed. But you can help to cut down on this impact by inviting a few users to be beta testers, perhaps who promise to be flexible and who can give you useful feedback and are willing to help diagnose issues.

Aside from system replacement, having a few users with whom you have a such a good relationship will make easier your life as a person who provides services to end users.

tl;dr

Don’t name things new or old.

But what name to use? Cute names might be fun, but they are also opaque. So names such as marypoppins, hyperion, saturn are best avoided. Descriptive names are better. So when replacing a system called printserver use a name such as printer or printspooler.

Note that I’m making a distinction here between systems and machines. Naming of machines is a different debate, some of which falls under the “cattle vs pets” mentality although as a side note I will admit to not entirely liking that nomenclature. Sometimes a service is provided by a single machine in which case there will be an overlap, but in general for this article I’m focussing on the naming of systems. ↩

No agenda, no attenda

Mon, 06 Dec 2021 00:00:00 +0000

Meetings are a necessary part of pretty much any office-based job. (I’ve only minimal experience in other work sectors so cannot meaningfully comment on them). People often feel both that meetings are unproductive and that they take time away from work itself getting done. There may be an element of truth in this but that doesn’t mean that meetings are forgoable. I say that well-run meetings can act like a lubricant, helping work to be more effective, that information can be shared with relevant parties etc. So how can one bridge the gap between these two viewpoints?

I’ve come up with a few ideas based on my own observations and from conversations with colleagues over the decades.

Have an agenda

How often have you gone to a meeting not knowing what it was about and then been called to comment on something and felt woefully underprepared? Publishing an agenda before the meeting (e.g. in the invitation) allows participants to prepare. The other benefit of having an agenda is that it sets a clear expectation of when the meeting is over. (Otherwise, how do you know when you’re done?)

Despite what the title of this article suggests, I’m not recommending to automatically refuse a meeting if there is no agenda published. Doing so is likely to make waves in your organisation and sully your reputation. But a followup request for an agenda prior to the meeting can help to establish the cultural habit in your company of including an agenda.

Include suitable breaks

Human concentration wanes after an hour, perhaps 90 minutes. This could be through boredom, maybe they need a coffee/cigarette/toilet break. If your meeting is longer than this, be sure to include breaks.

Consider the list of attendees

Indicate in the invitation whether attendane is mandatory or recommended. This may differ on a per invitee basis but it’s still relevant. It might be clear from the subject and the agenda (you are including one, right‽) but you can’t always rely on this; new starters within the organisation might not have the cultural knowledge to make this determination. They could, for example, attend everything thinking it necessary or perhaps useful to help them gain organisational knowledge, or they could do the opposite and deselect themselves due to thinking them too new to be able to contribute.

Make sure, also, that you are including relevant people, and not including people who don’t need to be there. For example, imagine that a team believes the company needs to purchase some software. A person in chgarge of signing off expenses would likely need to be present. But one of the graphics design team is unlikely to be needed in a meeting to decide whether to renew the contract with the office cleaning company. These are extreme example, sure, but even within teams there can be meetings which require some but not all members.

Publish a summary

Some people might have missed the meeting - either due to being in another meeting(!), declining due to other duties, or perhaps being on holiday or absent due to illness. Publishing a summary of decisions made and agreed actions along with task owners and due dates can help to set expections.

The recipients of this summary should be the original invitees plus any additional people who took part. The benefit of including the original invitees rather than just the attendees is that it benefits those who missed the meeting, regardless of reason.

Try to avoid lunch times

"I've booked a lunch time meeting because that's the only time people were free".

"Yes, that's why I was free"

Really, try to avoid booking meetings which occur during lunch time. Though this isn’t always possible, people need to eat lunch in order to be able to concentrate. If you’re going to book a lunch time meeting then provide food.

Have suitable padding

In this context, by the word padding, I mean times between the other meetings of the attendees. If one meeting ends at 1000, this doesn’t mean that attendees will be free at 1000 and that your meeting can start at that time. What if the previous meeting overruns? Although on this note, it’s also rude to the next group who might use that meeting room if you’re not out at the time when their meeting is due to start. By finishing on time, people don’t have to leave early plus this gives time to tidy up the room, re-arrange any chairs, and let the air circulate. (It’s unpleasant to go into a meeting room still with the aroma of the previous users)

How much padding to add? A suitable padding is in the order of 5-10 minutes before the half-hour mark, depending on how long the metting is. Possibilities include 1000–1050 and 1330–1355, instead of 1000–1100 and 1300–1400. This gives opportunity for the aforementioned room-reset.

Providing gaps between meetings also allows people to get a coffee or visit the toilet rather than having to join their next meeting late.

Finishing thoughts

I’m trying hard not to come across as misanthropic in this post. I’ve been to well-run meetings, and I’ve been to poorly run meetings. I know which I’d rather go to. It is hopefully clear that having attendees be engaged is much more useful than having ones who are not, and having a reputation of running effective meetings will serve you well professionally.

I’d welcome input of any additional ideas you might have.

Adventures with puzzle websites

Sun, 05 Dec 2021 00:00:00 +0000

I’ve been writing code for a very long time. I like to think I’m good at it, but know also that I can be better. My current job, as much as I enjoy it, has limited scope for writing software; it’s typically limited to writing Infrastructure as Code which is close, but not entirely the same.

To scratch that itch, and keep skills honed, I like to make use of puzzle websites. These are websites which contain a series of challenges. Typically they start off simple and get more complex but this isn’t necessarily the case. Those that do, are good for people learning a language because simpler problems will likely use less complex code constructs, but they also provide the opportunity to challenge more seasoned developers.

Once in a while I’ll use one of these sites to experiment with a language other than the one(s) I use most regularly at the time; those at present being Ruby and Python. I like to compare the type of approach that one uses in one language with that which would be used in another. Sometimes I find that my use of one language will even impact how I write another. (This happens to me in spoken languages too.)

These puzzle websites usually have a discussion forum which is hidden from view until a puzzle has been solved. I have found this wonderfully instructive because there have been occasions where I’ve solved the problem in a convoluted way but read some of the solutions submitted by others and can then learn something.

Here are some of the puzzle websites I’ve encountered

Project Euler

I think this is the first puzzle-based website I started using with any regularity or perseverance. I discovered this website in 2008 ish. I started using it to improve my mathematical skills rather than programming, though I do usually write code to solve them. (There have been a couple I’ve solved by hand with pen and paper)

This website doesn’t take code as a submission; instead it asks for the answer which is almost always a number but is occasionally a word or series of letters.

As much as I enjoy maths, I’m finding the limits of my mathematical ability. The later ones are well beyond my reach but there’s lots left of the early ones for me to solve still. I’ve completed 114 out of 775.

A new puzzle is published approximately every week. It’s suggested on the site, but I repeat it here for the sake of this article: Don’t try for a brute-force approach, It may work for the early puzzles, but it definitely won’t for the later ones.

shelly.dev

I discovered this one only a couple of months ago. The goal for this website is to draw shapes in as few moves as possible using a dialect of Logo. There are 33 puzzles (at the time of writing) but there’s a sandbox so you can have a play with the language too.

I don’t use Logo as part of my job but I have feelings of nostalgia towards the language because it’s one of the earliest languages I used. I first encountered Logo in the 80s at school on the BBC Micro (through which I also learned BASIC)

With this website you can also create your own challanges and offer them to other site participants.

Solutions are not shared even once solved, but the website does have a recommended number of steps with which to solve each puzzle and awards a bonus if your solution uses fewer steps than this recommendation.

microcorruption.com

The theme for this website is an electronic lock. The challenges take the form of using carefully constructed usernames and passwords to cause the lock to open without knowing the genuine credentials. The firmware to the locks is made available, and is written in machine code for the MSP430 processor. I’d not encountered this processor before discovering this website. My experiences with processors thus far has been limted to ARM, 6502, 68000, Z80 and MIPS

I don’t do much electronics these days, but this website is a fun way to keep a toe in the water.

No user-contributed challanges and no sharing of solutions. But definitely one of my favourites.

CodeWars

This is a very comprehensive site. It has a series of challenges in lots of different languages and as a participant in the site you can chose in which languages you’d like to see the puzzles.

The challenges take the for of a text description and a series of automated tests (written in code). Your goal is to write code to get these tests to pass and then your submission will be subjected to further more rigorous testing.

User contributed puzzles are welcomed; indeed this is how the site thrives. You can see other people’s submissions and compare them to yours. On this site in particular, I am frequently impressed by the variety of approaches that people take compared to my own.

PostgreSQL Exercises

I’ve only had a small play with this one. But it is a good place to practice SQL skills. It shows sample tables (both structure and their content) and then shows what data the site wishes to be extracted. The challenge is to write the SQL query to retrieve this data.

SQL has dialectal variations between different DBMSs but this site is focussed on Postgres

AdventOfCode

Topical for this time of year! This website has two closely-related puzzles per day for the first 25 days of December, and has been running every year since 2015.

The challenges are usually based around text manipulation though there are a few algorithmic related ones, and the puzzles from year 2019 featured several in which an emulator for a fictitious CPU is developed.

As well as a site-wide league table, this website allows you to create a league table amongst a group of friends/colleagues.

Odd and ends

There was also a couple of puzzle sites written in Flash that I used to play 10-ish years ago. But with Flash having discontinued in 2020 they are no longer available in their original forms.

The first one I’m thinking of is Lightbot, in which one pilots a robot around a tiled arena trying to light up certain squares. It’s a bit like Logo. This game is now available as a mobile game (still called Lightbot) though I’ve never tried it in this form.

The second Flash game was called Manufactoria. The game play for this was a finite-state machine, which was run by a robot which would run around an arena containing directional conveyor belts bridges etc, and the goal was to turn one input pattern into another, or to accept/reject certain input patterns based on criteria specific to the puzzle.

There’s also the website exercism which I discovered whilst writing this article, though I’ve never used it. At first glance, it looks a bit like CodeWars so is good if you wish to work on several languages.

Finally, there’s a collection of links in this Reddit thread. Some of them are mentioned here, but I’ve not checkout out all of them.

Closing notes

As part of my job, I occasionally run training courses in Python. I usually suggest this sort of website to give ideas and practice drills to students for content after the course.

Do you know of any others? Do you run one, even? (Are you involved with ones of the sites I’ve mentioned? Please say hello if so)

The difficulty of debugging without logs

Sat, 04 Dec 2021 00:00:00 +0000

According to a colleague I mentored around 10 years ago, “What do the logs say?” is one of my mantras. I guess that’s fair. It probably still applies now. [Current colleagues, feel free to pass comment!] Logs are designed to be a useful source of diagnostic information and they’re often the first thing I check, after any message immediately output in stderr or stdout.

But what happens when logs don’t contain useful info? This story covers the investigation I took when trying to track down the cause of a problem at a previous client.

Context

One of the functions my employer provides at this client is the building of Ubuntu laptops used by developers and civil servants. This is not at the hardware level; we use a hardware reseller for that. Rather, we perform software installation/updates and machine hardening such that they are allowed on production networks and to access customer data.

The build environment takes the form of a network segregated from the main one used by workstations on a day-to-day basis. We’d relax the boot options of laptops to permit a network build, peform that network build, and than re-restrict the boot options to prevent unauthorised boot media from being used. The build server has several components, the primary ones being: DHCP, TFTP, Cobbler, and Apache.

The problem

I had updated the build server with a newer kernel. But attempting to boot from this newer kernel image was unsuccessful. Normally the visible effect on the laptop that was being built was a series of dots, which act as a progress bar, except that there was no completion figure given; instead it would uncompress and execute the kernal. (I think it’s probably 1 dot per kilobyte, but that detail isn’t important). The effect when using this problematic boot menu option is that no progress dots were shown.

Previous boot menu options remained working.

Investigation and findings

Checking the system logs on the boot sever showed that of the two files required for each of our boot menu options, for this problematic menu option only the first file was being requested by TFTP. Other boot menu options continued to work unimpeded. We knew that the network cabling was functional and correct because the to-be-built machine was able to perform the network boot to acquire an IP address and load the menu of available options.

A tcpdump showed the TFTP request coming in and (at least a partial) response going out. On the cobbler server, an md5sum comparison between the new image and an older one showed that the kernels were identical, and that the initial ramdisk images were also identical. We could see the first file being requested (for the problematic option) via TFTP in the logs but two things stood out: one that the response was short (and there was no followup traffic), and two, that the second file was not requested.

Eventually I had the idea to run tcpdump in verbose mode and then debugged the individual bytes in the response. It was here I determined that the short response was a denial. (There was no easy way to get the client to log more verbosely because it was still in the pre-execution phase; no OS had yet been started by the client). I then looked more closely at the file on the boot server and found that root was the owner and the filemode was 0640. The TFTP service ran as a different user (I think nobody).

Summary of cause, along with highlighted suggestions for improvements

The kernel (and initial ramdisk) images had been given too restrictive permissions. Relaxing the file mode to 644 was enough to resolve the problem.

The “simple” diagnostic approach would have been to run tftpd in verbose mode, to make it be more chatty about permission errors. But there’s also the so-called principe of least surprise: one would normally expect an application to log when it cannot fullfil a request, at the default log level. Apache and nginx do, for example. The “complex” answer would be to submit a patch to tftp such that it logs inability to accede a request at the base log level. This approach also requires that patch to be accepted which may or may not be a simple battle, depending on the software in question.

How do I write good code?

Fri, 03 Dec 2021 00:00:00 +0000

“How do I write good code?” was a question put to me by one of my direct reports a little over a year ago, in one of their one-to-ones. I was able to give a brief answer, but the question has been ticking over in my mind. A few days ago I launched this blog, and so I thought it would be a good opportunity to put some of those digested thoughts into written form.

There isn’t a simple answer this question, but it is still answerable. To get closer to an answer, we need to take into account a number of factors. We can do this by first seeking clarification: What do we mean by good? We can assess code on several fronts: ease of maintenance, speed, bugs and error-handling, and comprehensiveness. Let’s look at these in turn.

Ease of maintenance

Sometimes code is written once, to be thrown away. More often - the vast majority of the time, even - code is preserved to be run again repeatedly.

As business needs change, such code will need to be modified. Consequentially, it’s usually the case that more time is spent reading code than actively writing or changing it. Therefore it is better to have code that is easy to understand than it is to have code which uses quirky or obscure behaviour. Although such tricks are sometimes necessary, for example they might be faster; see the section on Speed below.

Speed

Code doesn’t always have to run quickly. Is this a one-off process or will it be repeated? Will there be a human waiting for the code to complete or will the code run unattended. Imagine a scenario when a monthly mailshot is sent out and that this script takes 60 minutes to run. If it needs to be launched manually, and runs locally on someone’s machine, then that person can’t shut their machine down whilst the script is running. But if it completes in 60 seconds then this is less of an imposition. Better would be for such a regular task to be on dedicated infrastructure, though this is perhaps a separate factor from that of speed. But returning to the issue of ‘dead time’: if a script takes 5 minutes then that’s a useful amount of time: one can make a coffee, go to the loo, etc. But 30 minutes is an awkward amount of time because it’s too short to work on anything else, because that task would need to be interrupted to resume the task being done that necessitated the script in the first place (the mailshot, in this example).

Let’s take another example. Imagine a website - perhaps it’s an online retail website. Numerous studies indicate that visitors will abandon a website if it is slow to respond. So there’s a financial and reputation impact to the organisation to have performant software.

Bugs

Except perhaps in subversive contexts it is rare to intentionally add bugs to code. It’s generally agreed that once identified bugs should ideally be fixed, and if not immediately fixable then the bug should be recorded in a tracker, somewhere central, and ideally also linked to the code repository such that future people making use of the code can find the list of defects easily.

Whether a bug is high or low priority depends on its severity and/or impact, and how often it comes up.

A bug with minimal fallout that doesn’t happen very often (e.g. “This script does not work on any leap day”) might easily be postponed if it is non-trivial to fix. But a bug such as “The payment processor will abort the entire batch if any of the recipients have an apostrophe (') or hyphen (-) in their name¹ is much more critical and would likely need to be prioritised.

Error handling and reporting

This is closely related to the Bugs section but it’s worthwhile discussing separately. Let’s take the previous example of hyphenated names. It could be that the whole batch fails with a cryptic error “Batch failed”. Or it could be that the whole batch fails with a more useful error “Record 437 failed due to unrecognised character in name Malcolm Wynn-Jones”². Hopefully it’s clear why this kind of error is preferable.

Another kind of failure mode would be for the other records in the batch could succeed and solely this problematic record go into a queue for human attention, for manual processing. This is, perhaps, the preferable result, so long as the fact that there are problematic records is somehow made known, rather than them falling silently into the void.

A truly awful outcome, which would leave a lot of manual cleanup, would be if in the event of a hyphenated name aborting the batch processor, the records up to the problematic record were correctly processed and furthermore that the error reported be the less useful “Batch failed”. Why? Because the natural response (to someone who doesn’t know this script’s cursed behaviour) would be to run the entire batch again – because they have no indication that some of them did, in fact, succeed. This would lead to load of double payments. Problems such as this can be mitigated with staff training, sure, but people will always leave and join so it’s better to have error messages be as useful as possible. Plus in the event of a rare error, the specifics of a given script’s foibles are easily forgotten so having the error message be descriptive is a good thing.

Comprehensiveness

Imagine a process with 5 steps. Does this script do all 5 steps, or does it do the first 3 and still leave the final 2 to be done by hand? It might be the case that the script is new and hasn’t been finished.

But the downside of a script not being complete is that it’s much more likely for a step to be omitted or carried out wrongly. This risk diminishes (well, bugs notwithstanding) if the script covers all steps.

In the absence of a script automating all of the required steps, a useful safety net is to have a document with a well defined procedure, including a checklist. Designing a checklist is itself an art; NASA have even conducted researched into the human factors of what makes a good (or in their words “normal”) checklist.

Closing notes

A thousand or so words is more than I was expecting to write! Writing perfect code is almost impossible to achieve. Writing good code? Sure, that’s doable. But first let’s decide on what we mean by “Good”.

There is a excellent article which expands further on this topic: https://www.kalzumeus.com/2010/06/17/falsehoods-programmers-believe-about-names/ ↩
This is a character’s name from the BBC Series Spooks. That was a fantastic series, worth a watch if you enjoy BBC Drama/Action series. I might buy myself the box set. ↩

On recognising people face-to-face vs online

Thu, 02 Dec 2021 00:00:00 +0000

Like much of those in systems/development and sys/dev-adjacent roles (and many other industries, for that matter), I’ve been working from home for the vast majority of the past 2 years. I have only occasionally travelled in to either the office or to client sites.

During this time, my employer has grown from 80ish employees to 130ish employees. Some people I’ve not met in person. And for a long time, this included my (then) delivery manager. During the summer she ran an in-person general knowledge quiz and seeing her from afar in the office that day I at first I didn’t recognise her. It made me think; we take different cues based on the context in which we encounter people.

On camera, though people have facial expressions, features such as body language, build, and height are camouflaged. But in-person we do get these cues, plus, perhaps a person’s gait.

I’m reminded also of a time when I still drove and was living in Manchester. I was driving to a place that I’d only ever walked to, though had done so many times. On my first time driving there, I missed the turning. Again, this comes down to the cues. When driving, one tends to think about which lane to be in, how wide or tight to take a turn, etc. When walking one might notice the house with a broken fence, the bush that overhangs the pavement, or where the texture of the pavement changes etc.

I’m not a sociologist or pyschologist but I’d be interested to read people’s thoughts on this, and whether any studies or theses have been written on the topic of context-based cues and recollection. The pandemic has shut down much of the world (to varying degrees) but it has raised a number of curiosities such as this.

Musings on markup of deeply nested tags

Wed, 01 Dec 2021 00:00:00 +0000

This post started off as a conversation with a Twitter mutual (also called Matt), in which we lamented some of the pain-points of HTML and came up with a couple of proto-ideas of solutions.

Markup, particularly in HTML and its relatives, can often have deeply nested tags. Sometimes, when writing a closing tag, the hierarchical context of its opening countpart can be off-screen. This is particularly the case if one of the tags contains a lot of content.

Take a look at the following example:

<html>
    <body>
        <div class="article">
            <section class="introduction">
            <h1>Introduction</h1>
            <p>This article demonstrates the use of placeholder text. It
            contains pseudo-latin text which starts with the wording
                <span lang="latin"> Lorem ipsum dolor sit amet</span>
            and </p>
        </div>
    </body>
</html>

This example is short so the off-screen effect does not apply so strongly. But one of the ideas was to have implicit closing tags, borne out of the idea that though HTML tags can be nested, they may not overlap. Using this suggestion, the previous fragment could be represented as:

<html>
    <body>
        <div class="article">
            <section class="introduction">
            <h1>Introduction</h1>
            <p>This article demonstrates the use of placeholder text. It
            contains pseudo-latin text which starts with the wording
                <span lang="latin"> Lorem ipsum dolor sit amet</span>
            and
</html>

Notice the absence of the </p>, </div> and </body> tags. Since the </html> tag cannot appear until these previous three tags have themselves be closed, why not have the </html> tag close them implicitly? As I write this, I’ve come up with an extra idea - that even the </html> tag, in this example, can be omitted if it were to appear at the end of a stream.

There are a few benefits to this:

less chance of making a typo in a closing tag’s name,
less screen real-estate used,
less need to keep track of 100% of context in mental state.

It’s also slightly future-proof; if further opening tags are added then they do not necessarily need their closing tags to be added.

A second proposal that we discussed, again based on the non-permisibility of overlapping tags, is to have closing tags of the following form

<html>
    <body>
        <div class="article">
            <section class="introduction">
            <h1>Introduction</h1>
            <p>This article demonstrates the use of placeholder text. It
            contains pseudo-latin text which starts with the wording
                <span lang="latin"> Lorem ipsum dolor sit amet</span>
            and
<////>

See the 4 / symbols? Although the HTML standard requires it, there is no need for the tag name to appear after the closing symbol. So this <////> tag closes 4 tags. It doesn’t have all of the benefits that the previous idea does, but it is still lightweight. And a small extension to this idea is to have an integer count:

<html>
    <body>
        <div class="article">
            <section class="introduction">
            <h1>Introduction</h1>
            <p>This article demonstrates the use of placeholder text. It
            contains pseudo-latin text which starts with the wording
                <span lang="latin"> Lorem ipsum dolor sit amet</span>
            and
</4/>

Note that the number refers to the number of closing tags, not the number of omitted / symbols. So </4/> means close 4 tags. Labouring the point, perhaps, but it’s easy to miscount (or mistype) the the number of closings to be performed by the tag </////////////>? It’s much kinder on the eyes/fingers/brain to see </13/>.

With the use of minimisers (such as might be performed by CDNs and other layer 7 technologies), the emphasis presented by use of indentation would be lost, and so use of the <////> and </4/> forms harder to debug than named closing tags.

Neither me-Matt nor other-Matt are members of the W3C. I’m also not anywhere near claiming that any of these proposals are ready for submission or adoption by the W3C, but it’s interesting to muse on what might be appropriate.

Perhaps a lighter-weight version of this would be to use a filter or wrapper language (such as jinja2, SASS, LESS, or Cheetah) and have the processing be done on the remote end. This also improves adoption because only servers need updating, rather than the plethora of browsers, screen readers, etc that exist in the wild. I’ve not yet written anything that does this, but I’d be interested to hear your thoughts, and especially if you go so far as to implement a filter.