It's great to see women translating the classics of European literature, and getting a good audience for their work (such as Emily Wilson's translation of the Odyssey). I had hoped, however, for something other than what I got. I share Samira Nadkarni's reservations, and have my own (some of which overlap), and a bro-tastic translation of Beowulf isn't for me. And I'm disappointed, because I felt that the translator was so close to getting the point, but ended up overshooting. In the end, it renders the translation rather flat and one-note, and misses out on communicating the subtlety of the Beowulf poet.

I admit that I'm not up-to-date with Beowulf scholarship (mind, even if you're employed as a scholar of Old English, keeping up with Beowulf scholarship is a full-time job); I studied Beowulf as a seminar class in my finals 20 years ago. Even so, I can spot a few unfortunate gaps in the engagement with the literature. The biggest gap is in her engagement with Beowulf's monstrous nature. I was fortunate enough to be a classmate of Alaric Hall, whose work on the gradations between "human" and "monster" helped put Beowulf (and elves) into context. The Beowulf poet uses the same word (āglǣca) to describe Beowulf, Grendel and the dragon, as Headley (and many others) have noted. But that repetition doesn't turn up in the translation, obscuring the commonality of violence and exceptionalism shared by these characters. She even notes that most translators don't translate them the same - and then reproduces that choice.

In the introduction, Headley says that the poem "reads like and Old English fairytale". This, to me, neglects the poetic form, structure, and repetitions used by the Beowulf poet. Sure, there's a moral core, but the nature of how vernacular poetry survived from before the Norman Conquest renders this pretty much inevitable. The Beowulf poet used repetition to provide structure: "oþþæt ān ongan", "þæt wæs gōd cyning" - those repetitions and how they're used have stayed with me for 20 years. It doesn't feel like a fairytale to me.

Headley notes that "the poem feels populist" - well, it's got to be popular and/or noteworthy enough for it to be worth writing down and copying over the centuries. That doesn't preclude it being complex and subtle - Shakespeare's plays are all of those things. Headley also misses the point of "the periodic recaps" - this is less a poem about action than about reflection on action and its consequences. Repeats aren't there primarily for those who've lost the plot. There is a structure and a purpose to it, to loop around and around the action and its meaning in different contexts.

We see another missing of the point when shy says, "the poem, and especially the monologues of the men ... feels to me like the sort of competitive conversations between men". This is so close to getting it - indeed, that really is the point of the conversations between Beowulf and Unferþ - they're competitive, edgy. But it's missing out on the need Beowulf has to justify himself, the words he uses in different contexts to describe his action. It's all flattened into one bro brawling match, with no sense of the subtly different contexts the speeches happen in. Beowulf has to justify his violent nature - he's only useful when there's a threat; he's no peacemaker. And thus in this translation we miss the way that the Beowulf poet is undermining it all - worldly fame and gold are no good once you die. But because the translation flattens it, we miss all the nuance of the differences in the story when it's being retold.

I can't write a review of Beowulf without mentioning the first word: Hwæt. Obviously every translator needs to provide their own essay on how to translate this word. Our seminar leader suggested, "Yo!" (Can you tell that I was studying in the late 1990s? Yes, you can. Way to date yourself.) To stay slightly closer to the Old English, you could dip into AAVE and go for "Wassup?" Which choice could lead you in a very different direction. Headley goes for "bro", which immediately frames her whole translation, full of frat house culture (the boozing sections also lean into this very heavily). In a bold way, she reuses the word throughout the poem, to try to give some structure. That might work for some readers - it threw me out completely because it reframes the narration and speeches into a different mode that's not really done in the original poem. (It partly is, through the use of maþelode, but this isn't a direct mapping, so it's not a consistent usage either.) It didn't work for me; I felt it pushed everything into one register, in a way that obscures what the original poet was doing. And it heavily gendered a neutral opening statement.

The translation isn't that close; there are some elided passages, and some mildly expanded ones. Now, this is a fair tradition in translating (I like to compare and contrast two tellings of the Tristram story - one from mediæval Iceland, the other from France. They're separated by a few decades, and a massive cultural gulf. The French instance goes on for pages of monologues from all the main characters, saying how tragic they are, and how woeful it is to be in such a terrible situation. The Icelandic version cuts out most interiority and gets straight to the fighting and the tragedy.) I think I'd be happier about the choices if they were consistent. Sometimes Headley turns a religious phrase into something completely secular; at other times she retains the religious sense. This gives a rather uneven effect - we lose some of the sense of the poet reframing a story that involved pagan gods into a Christian context. And that weakens the sense we have of the religious import of the poem, and the reasons why it could be preserved. It also lacks the opportunity to really lean into the frat-house interpretation and see how Evangelical Christianity shapes that scene.

The big issue for me is that a lot of the repetition is gone. "That was a good king" survives (though if it didn't, I'd be wondering what the hell the translator was doing). But our Fierce Assailant āglǣca gets translated as "fiend", "enemy", "intruder", thus reproducing the issue Headley found with earlier translators. There's also some bizarre gendering choices. Headley leans in to the ship in the opening passages being ice-clad to say it was an "ice-maiden", which is pushing the "ships are she" thing a bit far, in my opinion. We also see her use "that bitch" to translate "feond" - which has a perfectly good English cognate that isn't gendered at all, and still shows what you thought of the person. This effectively manufactures more misogyny than there was in a poem that was probably written 1300 years ago. The women are already badly-treated by life (which the poet makes perfectly clear, by undercutting all our assumptions); there's no need to add to that. Headley draws attention to this in the introduction, and then fails by flattening everything into one-note misogyny.

Nonetheless, the moment when Beowulf goes out to fight the dragon (c. line 2540) evoked genuine emotion. It's pretty hard to ruin Beowulf; I just feel that it was so full of missed chances.

Spoiler: this is a quick rewrite and consolidation of what's in Wikipedia on these topics, so there is some more detail there. I've just synthesised it, with a side order of this being one of the major topics for GCSE history, though there it came with a heavy dose of "and this is how Britain led the way in workers' rights", though you could just as well interpret it as "Britain industrialised early and led the way in exploiting their workforce until forcibly stopped". Obviously there are a ton of academic articles arguing about the nuance. Wikipedia will link to some, JSTOR will give you a lot more if you have access to it.

The tl;dr of it all is: Trade Unions (TUs) have been intimately involved with (male) suffrage. TUs were mostly male dominated. They're a primarily industrial phenomenon.

This is primarily focused on the UK, primarily England. Some of that is that there was undoubtedly a lot of activity there. Some of that is also that

There are two things feeding in to the Trades Union movement: one: Friendly Societies; two: the massive social and economic changes caused by the French Revolution and Napoleonic Wars, and the Industrial Revolution. One gives us a principle for organising groups; the other gives the impetus to form those groups.

The Friendly Societies originated in the mid-late 18th century, and have some things in common with the mediaeval guilds (mostly around organisational structure, and the notion of some benefits for members. Some of these societies were specifically Benefit Societies (e.g. for insurance, mutual aid, and so on. Like credit unions or building societies - both of which evolve from this).

The Wikipedia article on Benefit Societies (accessed today), says they are characterised by:

• Members having equivalent opportunity for a say in the organisation
• Members having potentially equal benefits
• Aid goes to those in need (strong helping weak)
• collection fund for payment of benefits
• educating others about a group's interest
• preserving cultural traditions
• mutual deference.

All of those things characterise modern unions. The path to those modern unions was a very difficult one, and as new industries evolve, workers aren't necessarily covered by existing trade unions (see the recent creation of the https://iwgb.org.uk/ the Independent Workers Union of Great Britain, which is trying to cover gig workers that aren't eligible to join other unions). This is a history of changing industries (starting with the great shift from agricultural workers + apprentices, to fewer agri workers, still apprentices, and the growth of factory workers) and economic difficulty and low representation.

From the mid-18th century onwards, new jobs were being created in cities. These weren't the agricultural jobs of the past (where labour was limited in winter, because it was too damn dark, with some seasons of intense work - like harvest). These were jobs using machinery such as the Spinning Jenny. The Napoleonic Wars took a lot of men out of the workforce, then dumped them back in (plus a lot of newly-disabled people - this also triggered a growth in workhouses, and started the fight for state benefits). Agricultural wages fell, and the price of corn went way up.  Mysteriously, this made people pretty pissed off.

First of all, unionising was difficult. The Combination Acts of 1799 and 1800 restricted "combining" or gathering together for a common purpose. These laws were strongly motivated by fear of the French Revolution - well, the fear that it might give people ideas.

They got ideas anyway. There were the Luddites (ca. 1805), who were textile workers protesting about their jobs being taken by lower waged workers using machines. The Luddites smashed machines, and they were arrested, put on trial (often these were show trials), and sentenced to transportation. Their protests were disrupted by military force, and new laws were written to discourage similar equipment sabotage.

There were the Factory Acts of 1802 and 1819, which started to restrict the use of very young children as factory workers. This didn't spring directly from trades unions. The first Factory Act came about because of the spread of an infectious disease, and this was the main motivation. Workers' rights were not really involved. The later Factory Act came about because of the work of Robert Owen, who was a factory owner, and a strong believer in socalist enterprise. It was he who coined the phrase: "Eight hours labour, eight hours recreation, eight hours rest". He coined this in 1817, and in modified form, it's been used by many campaigners since.

At the time of the second factory act (1819), there were already some trade unions, as the Manchester General Union of Trades formed in 1818. Trade Unions early on tended to be either regional, or to cover specific trades (or both). A lot of trade unions and political movements of the nineteenth century relied on increasing literacy. This was mostly because of the Sunday school movement, which arose in the mid-18th century, which wished to educate poor children (mostly boys) to read the Bible. "Ragged schools", with no or very low fees, arose from around 1818. This meant that there was enough literacy for there to be some readers in most groups of workers. The Industrial Revolution also reduced the cost of printing, so that it was feasible to distribute leaflets and journals.

Some early attempts at unionisation were harshly punished. The Merthyr Rising was a miners' strike in 1831 over low wages, unemployment, and the treatment of debtors. It was the first time a red flag was flown by workers. It was dispersed by the army, and some protestors were killed, many were imprisoned and transported, and one was hanged. The Tolpuddle Martyrs formed a Friendly Society of Agricultural Labourers in 1834, to complain about the lowering of wages. They were initially sentenced to transportation (for "swearing false oaths" of membership), and later pardoned.

There were also some early attempts to get a National TU movement going, starting with the National Association for the Protection of Labour in 1830 (this mostly covered the textile industry), and the Grand National Consolidated Trades Union in 1834, which was focused on the building trade, and riven with disagreement. The New Model Unions were also forming around this time, but they tended to be for specific trades, generally ones which required higher literacy (e.g. Engineers), and they tended to focus on negotiation and education, rather than strike action.

Not long after (in 1838), the Chartist movement began. They weren't a workers' rights organisation, as such, but a group looking for male suffrage, a secret ballot, and fairer constituency boundaries (the boundaries had been drawn up for a primarily rural population, and didn't take into account growing cities such as Manchester). This rapidly tied into the Trade Union movement during the 1842 General Strike, where the demands focused on increased wages, and implementing the demands of the Chartlsts. The strike began when a mine owner broke the law about giving notice for pay reductions, but swiftly spread through the Potteries and into South Lancashire. Many Chartists were imprisoned as the strike was broken up.

As far as I can tell, everyone took the 1850s off, or at least they were doing things that didn't make it into history lessons or Wikipedia. Though that may have been a result of post-1848 panic, as the UK government got a bit twitchy after a bunch of European revolutions.

The 1860s saw a lot of consolidatory activity. The International Workingmen's Association was formed in 1864, which was an international alliance of socialists, communists, anarchists, and TUs. It suffered a lot from internal conflicts, and disbanded in 1876. Its heir was the Second International, founded in 1889. Marxism and Communism became an important part of the Trade Union movement. The UK Alliance of Organised Trade (1866) briefly existed; the Sheffield Outrages (explosions and murder) caused its end. The Trades Union Congress (which survives today) was formed in 1868, and was greatly aided by the eventual legalisation of unions in 1872. (Legalisation of unions elsewhere tended to come later on in the 18th century - in France in 1884, Germany 1897, and in the US from ~1880 onwards a series of court decisions made unions more permissible.)

The TUC was formed primarily from Northern unions from Manchester and Salford, and was trying to deal with a perceived London bias), and also some of the New Model Unions. Their ability to share information was probably aided by the 1870 Elementary Education Act, which permitted local government to set up schools to fill in gaps in provision, which led to increased literacy amongst the workforce. The TUC organised a congress in 1899, "to establish a voice for working people within Parliament". This led very directly to the modern Labour Party.

The first candidates weren't Labour Party candidates per se, they were candidates endorsed by the Liberal Party and sponsored by Trade Unions. The first candidate was in 1870. The TUC and socialist groups such as the Independent Labour Party, the Scottish Labour Party, and the Fabian Society gathered together to promote more candidates, and eventually we ended up with the Labour Party, with its very strong union roots.

You'll notice an almost complete absence of mention of women here. The TUs were mostly male led, though there were plenty of women employed in the textile industry, and they were more prominent in those unions and strikes. The Factory Acts in the mid-19th century reduced the working hours of women and children, thus categorising women as children, and not as adult workers. And I've not been able to find out how white the TU movement was in the 19th century, though I'm going to guess that it was pretty damn white, even if the workers themselves weren't. Though I'd expect wide regional demographics - London and Liverpool and Newcastle, as big ports, probably weren't so white, but the Potteries almost certainly were very white indeed. Sorry. My history education was somewhat silent on these matters, and spending a few hours reading isn't a substitute for having this embedded in a course of study. There is some material on 20th century union history that covers this: unionhistory.info/britainatwork/narrativedisplay.php, but there is a strong colonialist and racist legacy here.

The union movement was very successful; the UN Declaration of Human Rights has clauses in about the right to work (and not be discriminated against), the right to just remuneration, the right to join trade unions, and the right to rest and leisure. Those rights stem from the demands of Trade Unions across the world.

I'm participating in the UCU strike on pay, pensions and working conditions. I intend to tell you why. This will also expose some inaccuracies and generalisations in a recent Guardian article about inequality in Cambridge: https://www.theguardian.com/uk-news/2020/jan/12/beyond-cambridge-spires-most-unequal-city-tackles-poverty?CMP=share_btn_tw

The article has the usual difficulty in understanding the gap between colleges and the University, and also doesn't even mention that colleges are not all equally endowed. I strongly support the campaign to get the colleges to pay the Living Wage, and I hope that this can be prioritised when the colleges do their small annual redistribution exercise.

But the thing I really want to take issue with is this:

"Academics are protected from the worst financial pressures of living in Cambridge, benefiting from central, subsidised college accommodation, free meals and access to a cheap, university-backed shared equity mortgage scheme."

This is basically not true for most employees of the University. The biggest employment group in the University is Researcher. Then there are the admin staff, and academic-related staff, like me. We don't get college accommodation (neither do married fellows - I'm not sure which fellows are eligible for it these days; also, having a room in college for teaching is not the same as having subsidised accommodation). We're not eligible to be affiliated with a college for the most part. The equity-backed mortgage scheme is only open to new lecturers who have moved to Cambridge, so if you'd done a post-doc in Cambridge and then earned your lectureship, you're out of luck.

Living in Cambridge is expensive. Average house prices are north of £400,000. Average rents are high - you can realistically expect to spend from £500 to £1000 a month on rent. The lower end is only if you're prepared to live in a house share, or are lucky enough to qualify for one of the University-owned flats. This turns my salary (~£36,000) from a very nice salary in most parts of the country to one that's not really adequate if one wants to buy a house or have children. And good luck saving for a deposit with those rents. And if I'm priced out of the housing market by all the tech employees, what of the assistant staff who are paid much less well than I am? (I should know, I was first employed as assistant staff. It's miserable.)

Like many people, I've got a contract for as long as we can win the grant money. Fortunately, we're on quite a long grant at the moment, so I've got nearly 30 months before I need to worry about my job security. I'll point out that I'm 40. There are plenty of researchers in my building who are around my age and who are on the same kind of contracts as I am. The people with the secure jobs are all over 50. Hope you didn't want kids, or that you're prepared to disrupt their schooling as you throw yourself around the world (or the country) in search of your next job. If you don't want to do that, you have to leave academia. This is how tech siphons off ever more of the best and brightest - they may get to do original research, but it's in service to Google. We used to be able to offer good pensions, but the pension provision is being hollowed out, and our contributions are going up, That latter negates our pay rise for the year.

Pay rises. Yes. We've had 10 years of below inflation pay rises, resulting in about a 20% loss in real earnings. This year's pay rise was 1.8%, which is 0.1% above one of the measures of inflation (but below others). It's way below the increase in e.g. rail fares or housing costs. I'll note that this hits hardest on assistant staff who aren't covered by UCU - I think that's wrong, but pay rises won by those of us on higher salary points will translate into raises for the lowest paid, because of fairness (well, because some of us believe in fairness, and management are averse to the kind of bad publicity that will result).

My grant-funded full time position is actually one of the better outcomes. Lots of people are employed to teach at 0.25FTE. They have to work multiple jobs to make ends meet. They often don't get paid across the summer (no students). In Cambridge, most of those people are employed by the colleges. Some are PhD students and Postdocs getting experience and a little extra cash at the same time. Some are basically peripatetic teachers. In Cambridge, these people mostly aren't covered by the strikes, because they're employed by the colleges. Well, not employed as such. They're technically running their own small business, and the college is not their employer. Locally, we can work to get them taken on by departments as Affiliated Lecturers, so they can have more security, sick pay, and a pension. Nationally, we can work towards getting people longer-term and more substantial contracts. There are definitely lots of students available, and they'll definitely need teaching. We can't use an AI to teach them, so we should make sure that the people teaching them are able to do so to the best of their ability, not desperately stressed because they have to reapply for their current job in 6 months. Or desperately stressed because they have to put in extra hours to do the job properly, so they're actually averaging less than minimum wage.

This brings us to workload and conditions. A colleague of mine is seriously considering not applying for a lectureship, because all the lecturers they know are desperately overworked, and have very little time for research, especially at first. Even at the level of researchers, most people are seriously overworked, because there's not enough time to do good research, write the papers, do outreach or serve on committees or serve as conference organisers or do lots of peer review and so on - all of which are needed to try to apply for more secure jobs. And the example set by most of the people above you is that you do long hours, otherwise you can't get the job done. That's a terrible message for everyone, and this goes double for disabled people and people with caring responsibilities (still disproportionately women).

And there's the gender pay gap. Cambridge recently published its annual report. My grade is fairly evenly balanced. It's basically the bottom grade on the academic/academic-related track. Below me, there are more women than men; this is mostly admin grades. Above me, there are steadily fewer and fewer women. It's probably better not to ask about the pay differentials for BAME and disabled staff. It's probably grim reading, if only because the declared number of people in each category is exceedingly small and gets smaller as you go up the hierarchy.

So we're on strike, to try to get the Universities to commit to safeguarding our pensions, to help improve national pay scales, to get universities to commit to real action on pay inequalities, to move to more secure contracts, and to do something real about managing workloads. Platitudes are not enough. Pilates is not enough (though please don't cut it!)

It's not a case of our diamond shoes being too tight. I'm doing sort of OK, but I'm doing this for my colleagues who are effectively getting less than minimum wage. And I definitely support underpaid and overworked members of other industries to go out on strike. Please do pay your paramedics and your teachers more. Please do raise the minimum wage. The government can maybe support small businesses and the NHS to do this by e.g. taxing Amazon and Google and Facebook properly. There's a lot of money out there. Perhaps we can reroute it to people who are stuck in the gig economy.

Academia isn't the ideal ivory tower that most people imagine. It's full of a few people who have secure jobs and high salaries (though most of those at the professorial level are shockingly overworked), slightly more people who have secure jobs and good salaries, and a lot of people who have insecure jobs and OK salaries, and some others who have insecure jobs and poor salaries. If we want to keep academia as a place where people can learn things, and research things that can change how we look at things or how we do things, we need to improve pay, pensions and conditions.

When conscripted workers returned to Cambridge after the Second World War, some of them brought with them knowledge of some of the technological advances in Radar and computing.

The histories of the Labs have focussed on the dynamic men who directed the work: Watson & Crick (and later Aaron Klug) at the MRC unit (initially part of the Cavendish), Mott in High Energy Physics, (HEP), Martin Ryle in the Astrophysics Group (AP), and Maurice Wilkes, Director of the Mathematical Laboratory. All of the scientists won Nobel Prizes, and much of their work was facilitated by computation, first done by people, later by computers.

 

And these computers were often women, hired as "Computors", or "Technical Officers"* if they were lucky, first to do calculations by hand, and later to write computer programs. There were other women employed in technical roles, as "Scanners" for looking at photographic plates for HEP, looking for the traces left by sub-atomic particles, as draughtswomen, drawing contour maps and other technical diagrams, and also some research students.

 

* This job title still exists in the University; I'm employed as a Technical Officer.

 

The employment of women as Computors was common: at NASA, human computors calculated orbital trajectories for the early space program*; the UK Met Office employed computers, as evidenced by this guide from 1916: https://twitter.com/verityallan/status/1164850561214296065 , some of whom were probably women; at the Royal Observatory in Edinburgh, computers calculated the position of stars from photographic plates.

 

* As made famous by the book and film, Hidden Figures.

 

In the Cavendish, three main groups hired Computors: crystallography, the MRC (before it was spun off into a separate organisation), and astrophysics. All three groups were doing work (x-ray crystallography or interferometry) that involved Fourier Transforms. Later, they programmed the computers to do data analysis using the Fast Fourier Transform. These computers were the computers designed by the (then) Mathematical Laboratory, and included EDSAC II and TITAN.

 

However, despite their essential contribution to data analysis, the Computors were not likely to feature on papers. Much of this was the convention of the time, and some of this was down to a group culture (in Astrophysics at least) that did not encourage elucidation of the computational methods used. Since many of the scientific "computers" were also women, there may also have been some level of gender bias.

 

Whatever the cause, Computors did not get authorships on papers, and only ralrely are listed in the Acknowledgements section. We can look at a paper from 1962 as an example: Ryle, M. and Neville, Ann C., A Radio Survey of the North Polar Region with a 4.5 Minute of Arc Pencil-beam System, Monthly Notices of the Royal Astronomical Society 1962: https://ui.adsabs.harvard.edu/link_gateway/1962MNRAS.125...39R/ADS_PDF

This paper is a seminal paper on earth-rotation aperture synthesis. It is a full practical demonstration of the technique, using the 178MHz telescope to collect the data, and the EDSAC II computer to process the data.

 

The electronic computer was essential - it would have taken too long to do the calculations by hand. Some of the program was written by Ann Neville, one of Martin Ryle's research students, and co-author on the paper. However, the paper also contains what may be the first reference to a practical Fast Fourier Transform, devised by David Wheeler of the Mathematical Laboratory.* This feat doesn't earn him an author credit, though; he just gets mentioned in the acknowledgements. This is something that today would certainly earn him a co-author credit, and then a ton of references when his work was used for other applications.

 

Also mentioned in the Acknowledgements was Mrs E Waldram, for the "photographic output". This actually involved some early work on computer graphics, to output the data on a CRT (Cathode Ray Tube) and then photograph that.

 

Not mentioned in the acknowledgements at all is the team of people known as "the girls in the attic". They did the work to draw the fantastic map of the North Polar region, joining several photographic plates together into a hand-drawn map. Until good printing of computer graphics was possible, the "girls in the attic" routinely produced contour maps and other technical drawings. Very rarely did they get credit in the acknowledgements. (They did get credited in at least one PhD thesis in the Astrophysics group, by a man who will remain nameless.)

 

* I'm trying to confirm that this is a version of the Fast Fourier Transform, rather than something similar but distinct. This is still a work in progress.

 

I have looked at papers primarily from the 1950s to 1970s - by the 1980s, the convention was changing, and Computers and Technical Officers were more likely to have substantial contributions recognised on papers. I have a list on ADS of the early papers from the Astrophysics Group, which I have used as a reference: https://ui.adsabs.harvard.edu/public-libraries/KkZ1AAqVRsakG7N87AKV2A

 

I am keen to find out more about these women, and the techniques they employed, especially the computing techniques used. I work on the Square Kilometre Array, which will produce huge amounts of data, and memory management is going to be important again. We may be able to reuse some of the insights from the limitations of early computers to improve software for cutting-edge science in the 2020s. As well as the computers, I'm looking at the female research students in the Radio Astronomy group, as women have been working in the field for a long time. We just don't know about most of them.

Since my first year report is due in by tomorrow, I thought this was a good time to reflect on my experience as a part-time science PhD student. (Actually, I submitted my report before Christmas, because I had better things to do with my life than write my first year report while on holiday.)

First, a reminder that being a PhD student is So. Freaking. Hard. I mean, I'm finding it easier than the first time round, but that's because I'm 15 years older, have written a ton more words, and, having experienced two major depressive episodes, my brainweasel wrestling is world-class. Even so, it's very tiring, especially when the brainweasels come out to bite.

Second, quite a lot of brainweasels came out to bite over the first year report. For two reasons: one, my first attempt at my first first year report (Oxford, 2002) was sufficiently bad that my supervisor told me I couldn't submit it. (He was quite right. It was shit. Turns out you shouldn't do the bit of the project you hate while having your first major breakup from your first proper love affair. Being 22 sucks.) Second, when I did submit a proper first year report, the examiner asked a lot of pointed questions, which my supervisor brushed off. Those pointed questions surfaced again nearly three years later, in my PhD viva, which I failed. Fortunately, my supervisor this time was quite happy with my draft, and I shall be playing bloody close attention to what the examiners of my report say. I can take a fucking hint. However, I can't say that I could blithely ignore my history. It was certainly playing at the back of my mind, and took energy to deal with. That's mostly what I hate about the brainweasels. It's the wasted energy. And they often don't go away for long. Each individual weasel is fine. A constant stream of them is incredibly distracting. I mean, I'm really good at weasel-wrangling now. I'd just prefer not to have to do it so often.

Third, for me, doing a science PhD has some elements that are waaaay harder than doing an arts PhD. Mostly, it's the sense of near-constant failure. I'm getting this through programming, though I understand that anything where you have to do experiments also is really good at causing this. Throughout my undergrad, computing was unbelievably easy. I hardly ever wrote programs that didn't work first time. If they didn't, it was usually a trivial error (whoops, where's that semi-colon), and fixing that would cause it to compile, and then run Just Fine. I hardly ever made logic errors, because they led me through the exercises so gently. I only ever failed to write a working program for an assignment once. And before I got the marked copy back, I'd fixed my bug.

Now, I have to write Real Programs (TM). Obviously, I use StackOverflow, like everyone else, to find snippets that other people have written, and try to join these into a program that does what I want. However, I have to define what I want, and, if it doesn't seem to be happening, work out if that's a logic error, a problem with me not understanding my tools, or what.  I've very rarely failed at things I've cared about. I've either not cared, or I've not failed. I still deal terribly with failure. Intellectually, I know this is a learning process, and that I will get better at it (and indeed I am). But it means I'm confronted with failure most days in my PhD. Quite a lot of my programmer friends got familiar with this aspect of programming before they even hit puberty, or before they left school.

This is really new to me. In my previous attempt at a PhD, I was taking techniques that I knew already, and applying them to something obscure that no-one gave a shit about. When I did apply techniques new to me, they were pretty simple pieces of mathematics, well-understood and explained to zillions of students in other disciplines. And no-one who read my thesis understood them anyway. Now, I'm using tools that are under active development, so they're not always fully documented. And, while I know a lot already, i can quickly program those bits, so that I can get more fundamentally stuck on the other stuff.  So it feels like I'm constantly failing, even though what this really means is that I'm learning a lot.

Finally, some things are just quite hard and slow to do. I can read papers pretty fast. Apart from ones with lots of maths in. I'm still just much slower at those, because reading a maths paper is a fundamentally different discipline from reading anything else. Fortunately, I already knew from Terence Tao's blog that there's a particular skill to doing this, and it's one you have to learn as a grad student.

All of this is much easier to cope with now that I'm not in my early 20s, and know that it's not me, it's that I'm choosing to do something challenging. But it's not entirely unexpected that quite a lot of PhD students have mental health problems. Scaling the slopes of knowledge is hard, hard work.

One of the future bits of the Square Kilometre Array will hopefully be a Mid-Frequency Aperture Array telescope. We have some prototype designs in Cambridge, one of which I have photographed: Obviously, what you want to do is build one out of gingerbread or similar, because you can.

LemonBread/Gingerbread/Cinnamonstoat

75g butter

150g plain flour

0.5 teaspoon xanthan gum (if you are using gf flour)

60g soft brown sugar

2 lemons, or 4 heaped teaspoons of ginger or other flavouring

1 egg

Icing sugar

Water

Pans & Oven:

Gas Mark 5/190C/375F

1 baking tray, lined with greased greaseproof paper

things to zest and juice lemons

rolling pin

Instructions:

Put your flour (and xanthan gum, if using) into a bowl. Cut the butter up into little pieces into the flour.

Make pastry - rub the butter into the flour until it’s like breadcrumbs.

Grate the zest of your lemons into the mixture (or add your ginger).

Stir in the sugar.

Add the egg, and stir it in.

Juice your lemons, and stir in the lemon juice until it forms a soft dough. (If it’s not forming a dough, add a little water - or a lot of water, if you’re not using lemons.)

Flour your working surface and rolling pin. Roll out your dough until it’s around 5mm thick.

Take the Useful Shapes provided with your recipe, and cut out four copies of each shape. Put the shapes on the greased baking tray stuff.

Bake for around 15-20 minutes, until they’re starting to brown.

Leave them to cool.

Avengers, assemble! Mix up icing sugar with a very very little water, until it forms a firm paste.

Put a blob on a plate. Use this to anchor the central pillar (see illustrated assembled antenna that is totally an antenna, not a mistake at Stoathenge), and assemble the pillar, putting icing sugar paste on each edge.

Quite quickly, put some paste on the largest trapezoid, and splat it against the central pillar. You’ll want the longest edge against the plate (this would totally wreck your antenna, so it’s a good job you’re making a Stoathenge).

Then add the other large trapezoids, making sure that they’re pointing the same way.

Assemble the next layer, pointing the trapezoids the other way (still with the longest edge towards the plate). Then the final layer, pointing your smallest trapezoids the same way as the largest set.

Congratulations! You can pick up Stoat Medium Wave.

 Mushroom soup:

This is vegetarian, can trivially be vegan, and it can be easy to make if you have problems chopping stuff.

Ingredients:

1 leek
garlic (I usually use about 1/3 of a tube of garlic puree)
a lot of mushrooms. I'd usually use 1-1.5kg of mushrooms, a mix of ordinary, shiitake, chestnut, forestiere - basically strongly-flavoured mushrooms. Sometimes I rehydrate a pack of dried mixed expensive mushrooms as well, if I remember.
some oil (I usually use olive, but I suspect you can't notice)
stock (I use Marigold Swiss vegetable bouillon, which is gluten-free.)
a bottle of cheap red wine (optional)
cream or creme fraiche (optional)
seasoning: I usually use pepper, paprika, and mixed herbs (or fresh basil, if I have any)

If you're using dried mushrooms, rehydrate them now. When you drain them, save the water.
Chop your leek. 
Massacre your mushrooms. I usually chop some (for texture), and food process the others. However, if chopping is a problem, food process the lot. If you want varying sized chunks, perhaps use button mushrooms, which you can just wash and put in whole.
Fry the leek, mushrooms, and garlic in the oil in a 2-3 litre saucepan/pot/thing. I usually start frying after I've attacked the first batch of mushrooms, so that they start rendering down, and make it less heavy to stir. I don't usually stir much - just enough to stop it sticking.
Put the kettle on, with around 1litre of water in it.
Once the last lot of mushrooms have been stirred in, I usually pour in a bottle of red wine (if you're not using red wine, then add 750ml more water). Then I season generously with pepper, paprika and herbs. I don't usually make the bouillon up properly - I just put in enough bouillon for around 1litre of water, and then empty the kettle on top, and stir it all up - this usually isn't hard work as there's so much liquid.
Cover, and leave to simmer for at least 30 minutes.

At this point, you can either serve it, freeze it, or stir in some cream/creme fraiche and then serve it. It will serve 6-8 depending on whether it's a full meal or not, and on appetite. You can add cheese if you want more protein.

 I tweeted earlier today about three people whose work I admire. I thought I'd write a bit more this evening, and add a fourth.

My mother, former Senior Lecturer at City University, London, modeller at Rolls Royce and Associates 

She was working in tech in the late 60s and early 70s, including that mathematical modelling stint at RR&A. A child-related career break (OK, she took time off to have me), led to a career change, which eventually led to her lecturing in medical statistics. She co-authored an awful lot of papers in a short space of time. It's an excellent concrete example of how you can change what you do, and how you can be successful.

Dr Laura James, a founding director of makespace.org, co-CEO of okfn.org/

Laura was my manager in my very first job in tech. She's a visionary who can turn that vision into a sensible, achievable plan. Having had other managers since then, I've started to see how very, very rare that is. Makespace is an incredibly cool project, and I'm really starting to warm to the Open Knowledge Foundation after reading her blog posts. 

Jennie Fletcher, lead developer for cudl.lib.cam.ac.uk/

I've worked with Jennie for the past 18 months. I've been QA-ing her code, and I've been impressed with the consistent quality of the code she's produced. I'm also impressed with how the site works - I've had cause to use several sites which put scanned manuscripts etc. online, and, while CUDL's not perfect, it's definitely the best one I've used so far. That's at least in part because of the work Jennie's done on big, structural things. If I were composing a list of top 5 programmers whose work I've been involved in testing, Jennie would be on that list. I find that quite inspiring.

Professor Dame Athene Donald, professor at the Cavendish Lab, shortly to become Master of Churchill College Cambridge

She's shown me that women who are at the top of their field still experience everyday sexism, still feel uncertainty and a lack of confidence, still run across those unwritten rules. This makes me feel that I'm part of a community of STEM people, that I might be able to achieve something. It also reminds me of the equality that we should be striving to achieve.