Minion

[liv]

I have a PhD student! She started yesterday, and she is brill, I am so looking forward to working with her. I'm also a bit nervous, because I've never done this before, and I'm in very large part responsible for her career and quite possibly her happiness.

The PhD system is really weird in how heavily it depends on the relationship between supervisor and student. Modern academia is just starting to put safeguards in place to salvage the situation if the relationship goes wrong, but it's still essentially like a Mediaeval apprenticeship: your supervisor all but owns you and has almost unlimited power over whether you get your PhD, which is the essential and almost the only entry route into an academic career.

So I'm taking a leaf from rachelmanija's book: if you've ever been a PhD student, tell me stories! Tell me something your supervisor did that made things better for you. Tell me something they did that made the soul-killing struggle of getting through a PhD even worse than it should have been. (Like rachelmanija, I don't really need to know about obviously disastrously wrong things like sexually harassing students or completely ignoring them or stealing their work, because I already know I'm not going to do that. But hey, if it's cathartic for you to tell the internet how your supervisor was an evil crook who exploited you, go ahead!)

I have no problem if you want to give me general advice that doesn't come from direct personal experiences, or if you want to chime in with stories about a similar relationship that wasn't specifically a PhD. Also feel free to comment if I don't know you, if you found this by chance eg via Latest Things or Network. Anon comments are allowed but you may have to fill in a Captcha.

Comments

[rochvelleth]

Firstly, congratulations! It's just wonderful, and you must be so excited about it.

Now, my experiences:

My PhD supervisor bogged off to Germany for two of my three years on extended sabbatical/personal leave. Saying that, he was the best supervisor ever. I think part of it is that we're alike in our approaches to material (especially in how cautiously to treat linguistic/epigraphic data) - and he's a complete perfectionist and a total worrier, like me, and also is very responsible and kind and prepared to stick up for you if needed and extremely knowledgeable. Now that I have the BA fellowship, he's agreed to be my mentor, so we get another three years of a similar sort of relationship (but obviously different in various ways). And that's even though he is again off to Germany for a whole year in the middle :)

But the personality experiment is something you can't control of course - I was just very lucky to end up with someone that I ended up adoring. More practically, he was always very receptive but honest when I suggested something he just didn't buy; he kept in touch by email all the time and read through my drafts fastidiously whenever I asked him; he made sure we met up at least once a month or so, even when he was away; he was incredibly sympathetic when I was going through a rough time healthwise; he never put pressure on me or gave me deadlines (maybe in part because he knew how much pressure I was putting on myself - a different type of student would obviously require a different approach); and above all he was generous with his time and ideas, and that's something I think every PhD student needs.

I hope that helps!

[liv]

Wow, that's really lovely to hear, it's so rare for someone to have such a massively positive experience of their relationship with their supervisor. I agree that geographically absent doesn't necessarily have to be a problem; my main PhD supervisor disappeared to Singapore while I was still studying under him and even before that he was fantastically busy being famous and rarely actually in the lab, but he was really helpful remotely.

Personality I agree you can't really control, and you so often end up matched to the person you're working closely with for several years based on little more than a half hour interview. I really like my minion; she's a bit quieter than me, but otherwise as far as I can tell so far we're pretty compatible.

That's a really useful thought about dealing with disagreements. In some ways I have more experience than the minion, but I also intend to take her ideas and suggestions seriously. very receptive but honest is definitely something to strive for. Keeping in touch regularly is very much my plan, and luckily I pretty much live on email, so there should be few issues of minion not being able to get hold of me when she needs to. I've set a weekly meeting for the time being, with the idea that we may reduce the frequency later on once she finds her feet.

Pressure and deadlines is a really tricky one. I don't want to be a slave-driver or cause unnecessary stress, and it does look like I've picked up a student who already has a strong work ethic of her own. But I know when I was a student I somewhat needed a bit of a push from my supervisors. As you say, it does depend on the type of student.

[kaberett]

Part of the reason everything went so wrong last year was that my Master's supervisor:

- was not at all interested in knowing any details of my personal life whatsoever, so she got a bunch of work emailed in at 5am but I did not feel comfortable disclosing that I was struggling or why
- did not set regular meetings, where I really benefit from once-a-week/fortnight regular checkins and clear goals to try to meet
- did not set clear expectations or goals - about four weeks before hand-in, I found out I was supposed to be getting a bunch of SEM data, and so the timetable I was anticipating was really not going to work

Whereas my current supervisor:
- is setting up regular meetings
- is actually giving positive feedback
- is setting clear short-term goals and discussing long-term (i.e. over the course of the three-month project) timetables
- was cool with me disclosing up-front that I wanted to get going early because I expected November to be a disaster zone, and to get at least one serious chest infection during the project

... etc, and astonishgly enough I am not yet having panic attacks over EVERY TIME I SEE HER. (Ditto my SURF supervisor, actually.) So. Yes. That's how I work best; might be worth talking to your supervisee about how she works best?

[liv]

Thank you for this, it's really helpful. I am definitely starting out with regular meetings, weekly to start with and then we'll play it by ear. I have emphasised that the meetings are meant to be discussions, not a way for me to "check up" that Minion is doing enough work.

Giving positive feedback is something I have in mind as a goal, but it's also something I'm still learning to do well. I've got it wrong with my undergrad students sometimes, either putting more emphasis on the criticism (because I am over-keen to help them fix stuff) or being too generic in my praise so they don't find it helpful.

Any supervisor ought to be setting expectations and goals! I am nervous about this one because I have never in fact run a project of this size myself, and I don't always have a clear grasp on what's realistic in terms of time-scales etc. I'll be consulting with my co-supervisor / mentor a whole lot about this one, but good to be reminded that I need to make sure I include my student in planning at this level.

As for personal life and especially health stuff, this is one of the areas I'm most wary about, because even with good will it's easy to get horribly wrong. I hope I'll be approachable and my student will feel comfortable talking about general personal stuff in normal conversation. And that if there's anything that's affecting her work she'll be able to tell me about it and I'll be supportive and helpful rather than punitive. But I am aware that I have rather a lot of power over someone who's doing a PhD with me, and I most certainly don't want to pressure her to disclose things she'd rather keep private. Nor make her uncomfortable by revealing too much about my own personal stuff because however much I may like her, I don't want to presume we're friends.

Talking to her about how she works seems like a really good idea, and it's likely to be an ongoing thing, not just a one-off conversation. She may well not know yet, because she's had relatively little experience of related situations herself.

[mathcathy]

My PhD supervisor (as you probably already know) set me off on a project only tenuously related to his expertise as he wanted his PhD students to further his research career and was somewhat over ambitious with me. He also refused to listen to me when I told him I wanted to work on the theory rather than the practice in the form of constant programming which I was never going to be good enough at to compete at PhD level.

I think he is in a large part responsible for my not completing the PhD.
(He also refused to read the thesis and a fellow PhD student and I proof-read one another's because we couldn't persuade him that that was part of his job).

We're on good speaking terms now, though. I mean, although I didn't enjoy the experience, I liked him and we worked closely together albeit not very successfully,

[liv]

Examples of what not to do are very useful to me, but it's sad that you had to go through a bad experience to have this wisdom to pass on. Thank you.

I am definitely planning to proof-read and generally comment on drafts of any writing my student does, that much I'm sure I can get right! Actually it's something I happen to be reasonably good at.

The advantage of having nearly no money for this project is that I also have almost no strings; there's a lot of flexibility in where we take the project. I've said from the beginning that I'll try to set things up so Minion can pursue her interests rather than having to follow mine. Obviously I am ambitious and do want to further my own career, but it is no help to me if I make her work on stuff she doesn't feel confident with and she ends up not completing. If she wants do things that are very clinical, I hope I can get my friendly surgeons to train her rather than having a "blind leading the blind" situation. My post-doc involved me working on things my supervisor knew little about (he's a yeast geneticist, I do cancer cell biology), and this was a big problem for me, so I need to be very careful about that.

[syllopsium]

I can't offer much other than my congratulations and just the advice to be available, which I'm sure you will be anyway!

Isn't a 'muahahahaha' obligatory when gaining a minion though? ;)

[liv]

Thank you! Being available may be harder than it sounds, because I don't exactly have an office and because I'm going to have to plan my time very carefully to make sure it's not all swallowed up by demanding undergrads. But obviously completely essential for a successful relationship.

You can be sure I'm working on my evil laugh and other suitably tyrannical mannerisms which might come in useful...

[sea_bright]

I second what's been said above about regular meetings, or at least contact. I went through a sticky patch in the third year of my DPhil where I essentially didn't get any research done for about six months (chiefly because my funding had expired and I was suddenly having to try to earn enough money to live on), and neither of my supervisors got in touch to ask what was happening or why they hadn't heard from me. (To be fair to them, having two co-supervisors was a factor here: I think they both assumed I was working with the other one.) This resulted in a bit of a guilt spiral: because I hadn't been in touch for so long, I wanted to have something really impressive to show them when I did get in touch, but as I am the sort of person who is motivated by regular deadlines, achieving that was unlikely without having agreed a deadline with one of them... While the ultimate responsibility for getting the work done was (obviously) mine, a friendly enquiring email after a few weeks might just have provided the nudge I needed to actually start getting stuff done again, rather than perpetuating the spiral and allowing me to give in to the temptation just to keep my head down, because no one seemed to have noticed that I hadn't done any work recently.

More generally, the thing I would really have appreciated during the course of my doctorate was more advice on the things graduate students are supposed to do that aren't actually an official part of the degree. At no point during my time as a graduate student do I remember anyone ever encouraging me to go to conferences, or to think about submitting things for publication. It was at a fairly late stage in my DPhil (when it was really too late to do much about it) that I realized that a lot of other students had been doing far more on this front than I had - something which I'm certain had a substantial impact on my job prospects. It's fine to expect students to be proactive and take the initiative themselves, but even the most motivated students need some pointers in terms of what they ought to be being proactive about! It's also helpful to have some idea of what the general expectations are - what sort of publication/presentation record future employers are likely to be looking for, for example.

Finally, as someone involved in a project designed to promote good research data management, I should really put my work hat on and say something about that. Our experience suggests this is something it's really worth thinking about near the beginning of a project, as getting things right from the start often saves a lot of time and effort later on! There are some useful resources about various aspects of data management (organizing information, documenting data, secure storage and back-up, preparing data for archiving and/or sharing, etc.) here (the PrePARe Project training materials, linked to from that page, provide some of the best brief introductions to the topic that I've seen). The Digital Curation Centre also provide some helpful advice about data management planning, though as their name suggests, their focus is more on how you ensure data is preserved after the end of a project, rather than how you deal with it during it.

[sea_bright]

One other thing that's just occurred to me: a few days before I submitted my thesis, I swang by the university offices to check what the procedure for handing in was. It turned out there was a form I had to get signed by my supervisors before I could submit, which led to some panicky emailing and hasty setting up of meetings. Neither of my supervisors seemed to know anything about the form (something which still puzzles me, as I'm sure they'd supervised doctorates before, and I didn't get the impression the form was a new requirement). If I hadn't gone in to check, I might well have missed the submission deadline. Fortunately I did check, but it would have reduced my stress levels in the final few days considerably if at least one of my supervisors had known in advance what the requirements were...

Congratulations on your minion-acquisition!

[liv]

Yes, paperwork and procedures, absolutely vital. It's a bit in limbo at the moment because we no longer have a central Grad School, so the functions of getting forms signed and hand in procedures and arranging vivas and so on have been devolved to the individual schools. But hopefully by the time we get to the sharp end of this there'll be more of a consensus of how it's supposed to work under the new non-regime.

[liv]

Oh wow, data management. That's a really, really good point and thank you so much for all those resources. I knew we'd need to be systematic about it, but I didn't know about those tools. I'll be delving deeply into those linked resources over the next few months. Really appreciated.

I am planning on keeping in touch regularly, definitely. I know it's going to be tough to hold the balance between making sure there's enough support to keep Minion on track, but without micromanaging her or making her feel pressured. I started in the first meeting by explaining that I'll want to see her whether or not she has achieved anything since the last meeting, so that it's not about checking up, it's about lines of communication. But I'm sure she'll have a phase when she feels horrible because things aren't going well and she'll be scared to tell me so.

And yes, we are both thinking seriously about career development, not just sitting there doing the research. Keele is a lot better than Oxford at actually providing a structure and explicitly both requiring and providing generic / transferrable skills, so I'm a lot better supported in this than someone supervising their first student at one of the more trad universities.

[forthwritten]

I've heard that actual praise is nice. I wouldn't know this though.

My supervisor is utterly terrifying. Even when she's trying to be nice she's utterly terrifying. I showed a friend (also doing a PhD) a draft chapter she'd written on and my friend said that if that was her chapter, she'd cry.

I do kind of adore her because she's tough but fair - she won't let me present bad work or submit a bad thesis. I'd rather it was her saying that I'd written a load of rubbish than my examiners at my viva! She also does care in her own inimitable way and will send me emails if she's worried - she's got a lot better at being supportive over the course of my PhD. I'm her first PhD student so no doubt she's had a lot to learn as well.

[liv]

I am scared that I am in serious danger of turning out like your supervisor. I know I can be terrifying, it keeps coming back on my undergrad teaching evaluations. And I know why it is, it's inexperience combined with massive enthusiasm for fixing all the things combined with a style of interaction that doesn't suit everybody. I get the impression that Minion is rather quieter than me in personality (wouldn't be hard!), though hopefully that's partly because she doesn't know me very well yet. So I have to be really really careful not not overwhelm her. But yes, I should remember to give sincere, constructive praise and not just assume that only the things that need improving are worth commenting on.

[ptc24]

Meetings: during my PhD, a lot of supervision was done by talking informally over my desk. There were a couple of phases where there was a slump into despair, and it seemed that the way out of that was regular, formal meetings (in my supervisor's office, rather than the lab), sending in a page of A4 as a progress report the day before. That sort of arrangement persisted for two or three months until we felt that things were back on track.

The other thing that often happens is that supervision duties often get shared out; during my PhD we had a junior fellow who did most of my day-to-day supervision - I tended to think of him as my de facto supervisor, but it seems that my de jure supervisor was more dutiful in supervising me than many people's. It does seem to be useful to have an informal second supervisor; during my postdoc there was I time I seemed to be an informal third supervisor some of the time - it was one of the better parts of the job. OTOH, there was a formal mentoring scheme in the department where we each got a mentor; we ended up meeting once a term and it was a bit of a waste of time.

[liv]

I think informal just happening to see eachother in passing is very likely to meet that kind of failure mode. It's also particularly tricky for me because I have no base in Life Sciences where Minion is going to be doing most of her lab work; I'm looking into getting a temporary space where I can kind of camp out, and some of the time she's going to be in the lab within the hospital site where I have part of an office, but even in the best case scenario I'm going to be tied up in the Medical School some of the time, so we have to make a specific plan / effort to meet. But yeah, regular meetings, even or perhaps especially when things are not going so well, that seems like a really key point.

Shared supervision duties: I think Minion is going to be particularly well supplied in that department! I am her primary supervisor, de jure if you like, but she also has my collaborator, a biology professor, as second supervisor, and his lab tech to actually show her how to do experiments and where to find things in the lab, and a former post-doc of his, now a lecturer, as her "external" tutor.

I think the thing with artificially setting up a source of social support like that, is that it's a waste of time until it isn't. I had a thesis committee, of which only the chair actually bothered showing up to meetings and they kept reappointing other people in the desperate hope of finding someone vaguely conscientious. And we met once a term and ticked the box to say we'd met, and it was generally a waste of time. But when things did go wrong with the post-doc who was my de facto supervisor, and my de jure supervisor was not prepared or available to deal with that, the thesis committee chair was the person I turned to who was able to set things in motion to get stuff sorted out. Made all the difference.

[(Anonymous)]

I really struggled with motivation and getting work done during my phd.

I used to fail to work all week, then stay up all night before my supervisor meeting trying to doing a tiny bit of work that could be talked up at the meeting, scramble awkwardly through the meeting, and then relax when I'd survived it.

At some point my relationship with my supervisor was transformed when I stopped being terrified of him, realised that I actually wanted to complete the thing and solve the demotivation problem and was quite as annoyed at myself for the endless faffing as my supervisor ever could be if he knew about it.

So firstly, I'd go into the meeting and greet nick and ask how he was without this social connection feeling horribly twined with whether or not I'd succeeded in work, and secondly I admitted to faffing and tried to enlist his help to finding a solution.

He didn't really have one either. I suggested putting all my work into CVS so that he could read the commit statements (+dates) which would make it quite clear how much I had or had not done on each day, but although he agreed to that he never asked to see them if I didn't volunteer them so I didn't feel pressured by the system and didn't use it. Even so just not being scared of him any more was a big transformation.

I don't have the same kind of mind as my supervisor. Any way he framed a project just didn't work for me, and partly because of the motivation thing, I wasn't very good at framing my own, so I didn't really end up finding a project that inspired me. But that's not really his fault. He was quite good at matching his students/researchers up together, so that when I came up with this or that idea he'd direct me off to work with Raj or Georgios or whoever.

I sent him the penultimate draft of my thesis late on a saturday night, in the small hours of a sunday morning. He had the whole shebang read and back on my desk on the monday morning.

I failed to get any papers written during my phd, which is a big problem for thesis submission. He was invited to submit a book chapter and got me to do it, so I'd have a publication. We ended up doing two book chapters by mistake in the end, which was enough publications to scrape by.

[(Anonymous)]

for the first six months or so, I would report on my work during the weekly meetings my group held, which made it easier to hide the fact I hadn't done any. once nick realised that, he decided he and I should have weekly meetings!

[(Anonymous)]

most people in my departement had two supervisors, one 'main' one but then another for extra support. I only had the one for some reason.

[liv]

Thanks for contributing your experience to the discussion. Motivation and getting work done is such a huge thing for nearly every PhD student. Because it's a very long project with no clear deadlines and not really a clear idea of what exactly needs to be completed even by the ultimate deadline of the end of the project. So the onus is on the supervisor to create some kind of structure and act as a manager, which many academics don't have much experience or training to do.

I think I need to put lots of thought into how to provide motivation in a supportive way, without giving the impression I'm punishing my student for not working hard enough. I'm fairly sure we will get to a stage where she struggles to make progress, whether that's coming into the lab to do yet more experiments that don't work, or working another chapter of the thesis when it seems endless.

And yes, reading drafts and coming back with comments quickly is something I ought to be good at, thanks for mentioning that.

[(Anonymous)]

oh yes. all the time I was doing the thing I would be complaining I'd done no work, and other people would agree/say the same, and I'd be quite convinced that they didn't understand that I *really had* done no work and meant something quite different to them, and i was totally special 'done no work' cookie. but in the end - we all finished. and nearly everyone goes through a depression of some kind at some point, beaten down by the weight of the thesis.

I'm a better writer than coder and I actually recovered from mine once the end was in sight - the last four months or so of hammering the thesis together went fairly well for me. Other people (a larger proportion, I think), are more likely okay while they're experimenting and gathering results, and then get suckered by the writing up.

One useful piece of advice which I mostly failed to take but was glad where I had was to keep my bibliography file as I went along. So each paper I read I'd make an entry for and also - importantly - a quick summary of the content. Made it a lot easier to find things again! I guess that's obvious to a seasoned academic though :) In fact this whole comment may be!

I had (alongside motivation failure, etc), an inability to get out of bed in the mornings. I used to try and arrange my supervisor meetings, and also tea-n-chat with friends, for 10am or so, to force me to drag myself into the lab.

[(Anonymous)]

I’m just finishing up my PhD and whilst my scientific relationship with my director was great his social skills were just terrible causing a lot of not needed emotional stress on my part.

Things he did brilliantly:

1. Met with me and his other PhD students once a week for a tutorial. We presented articles we had read in turn or parts of our analyses and this was a great space for learning from each other and talking things out to make sure we understood them ourselves.
2. Consulted me in setting deadlines and always asked if I wanted to break down one big deadline into a series of smaller deadlines.
3. Pushed me to send papers to conferences and therefore to network helping me to break into our academic field and meet my colleagues.
4. Always talked about the national research scene in terms of politics, upcoming research themes, new policies etc. I feel he prepared me for a career not just to finish the thesis. For example, in France, data publication and open-access corpora are buzz words at the moment so my data is structured into a corpus to accompany my thesis and help me during my future recruitment process respond to criteria the labs will put in post profiles because that is where the funding is. He always shared his global vision of the research world and that was enriching.

Things he really needed to work on:

1.I taught full time before my PhD and went part-time in order to take on this project. Despite this, my director demanded that I was in the research lab from 9-5h30 on the four days a week I did not teach. I really appreciated the research-home divide but even if I was five minutes late he asked why or asked my colleagues where I was. At the end I even felt that I could not schedule medical appointments during those times. As a mature student who always met his deadlines I felt he needed to trust me a LOT more.
2. Going over papers I had written he would often tell me ‘no, that’s not right’ or ‘you can’t put that’. I think it would have been more positive to make me tell him first of all the problem areas I had in writing the paper, second to ask me my opinion about the sections he did not deem up to scratch and then to give me his opinion in a constructive way, before planning together the steps needed to make it better. Instead I felt like I had returned to Middle School and would find out where the problems were but never how I should go about solving them.
3. He talked about other colleagues behind their back and what he thought of their research. Not great when you have to work in collaboration with them.
He was very linear with one path from A to B. If you made it to B but by taking a different path he couldn’t cope.
4. And finally, my desk, because of his micro-management issues, became his space too. He had me rename all of the folders one day on my computer so I could find things (according to his logic which meant I couldn’t find anything) and one day screamed at me because my handbag was on my desk. He never asked why -my phone was on silent and I was expecting a call from an estate agent and didn’t want to disturb colleagues. I think the essential is for an advisor to be there to micro manage if needed but to ask the student how they want to work and if they show you that there are doing the work to trust them.

[liv]

Thank you, this is exactly the kind of thing I need! There's probably a fine balance between helping your student to navigate all the complexities and politics of the field, and gossipping inappropriately about colleagues.

I think I am not in too much danger of being controlling and micromanaging like you describe. I regard Minion as a colleague (even though the nickname doesn't really reinforce that). So I most certainly intend to work with her to decide about deadlines, and help her to build her career as well as just doing experiments under my direction. I've already told her I'm not going to be checking up whether she gets in on the dot of nine, and that she's welcome to work from home if she is in a phase where she needs peace to get on with some intellectual work. Apart from anything else, I'm not in the lab from 9 to 5:30 every day myself. When I'm reading stuff she's written, I'm definitely not going to treat her like a lazy schoolkid; I'm reasonably confident that I can give constructive criticism on academic writing.

I am working on setting up a "journal club" for a few colleagues working on related stuff (right now there's only one PhD student, but there's also a technician and a couple of newly appointed academics, and at some point there'll be final year undergrad students doing their lab projects.) But it's good that you note that kind of regular discussion with peers as a positive thing, cos it's something I'm quite committed to.

[ceb]

I had an awesome supervisor. I seriously wouldn't have made it without her. Here follows an assortment of good stuff she did and other good things:

* get your PhD student to write things up as she goes along, ideally for publication. This makes thesis writing less of a giant burden as quite often you can repurpose half-written papers as chapters. Also if any of it does get published that will be extra confidence and CV points for her.
* be enthusiastic. I don't just mean provide praise and encouragement, but every so often have a big joyous chat about how exciting and inspiring your field is and all the places your/her research could go. When bogged down in details of research it's easy to fail to see the wood for the trees.
* encourage her to have a life. PhDs are hard work and if she can find some fun and a support network that will help endlessly.
* she will probably take longer than she has funding for, most people seem to. Find out now about what options are there, then in two years' time you can say useful things like "you should sign up with this society now, and then if you need money next year they'll give you a grant".
* keep tabs on how she's progressing; make meetings frequent and low-stress and let her ask you questions as often as she likes. Don't be afraid to call her out if she's being crap - but be adult, constructive and supportive about it.
* keep her in the loop, and especially let her know when you're e.g. off on holiday/to a conference. There's nothing worse than feeling like your boss has forgotten you!
* encourage her not to get a job until she's written up; when you're supposed to be writing up, pretty much anything looks more attractive, and a job will probably mean it never happens...

[ceb]

Oh and I forgot to add, always do things when you say you will; set realistic deadlines for yourself as well as for them. One of the fastest ways to put distance between yourself and your minion is to make them feel like they have to pester you for everything (or even worse, to have them not pester you and sit there idle and annoyed).

[liv]

Good idea. I need to be setting and keeping deadlines for myself, and it's even more vital for me to be proactive at getting things done now that I have someone else relying on me. Thank you!

[liv]

Wow, it's really lovely to get at least one response of unbridled enthusiasm for an awesome supervisor, I was expecting mostly horror stories!

Writing up as she goes along, check. I need publications anyway, and this project is likely to be my best shot at them, since I don't have any other funding.

Enthusiastic I'm pretty much famous for. But thank you for the reminder that it can be needed when someone gets bogged down.

Encouraging life-having, yes, good plan. I'm really hoping not to be a slave-driver, and hopefully I can model having other commitments apart from doing science all the time, however much I do love science. This is something that someone at my career stage can easily get wrong, because I so desperately need to establish myself as an academic, plus I don't have family commitments which are the obvious reason to keep someone from working all hours.

Have a plan in mind for what happens if funding runs out: that is a really good idea, but there is a lot of institutional pressure for people to complete within the allotted time. As a small, modern university we don't really have a lot of spare funding, and we also get punished by HEFCE if our statistics for completion within four years take a dip. But still, it's true, it's well worth looking out for grants. And you're quite right it's better if at all possible to avoid starting a job in parallel with writing up, so we ought to make sure that doesn't become a financial necessity.

keep tabs on how she's progressing; make meetings frequent and low-stress and let her ask you questions as often as she likes. Don't be afraid to call her out if she's being crap - but be adult, constructive and supportive about it.
This bit is my favourite piece of advice ever! I might print it out and stick it somewhere prominent, because it exactly encapsulates the sort of supervisor I want to be. I need to keep reminding myself that this is what I'm trying to do. Also, yes, she needs to know when I'm unavailable or away, simply disappearing would be really rude!