My experience then (this was back before "Attention Is All You Need", I hadn't met the output of generative models) was that students tended to produce work that did not have a proper thread of reasoning in it. There was a tendency to repeat things they had read but rehashed in various ways.

Reviewing some of their texts it was clear that much of the writing - by law tutors - was of the same kind. Much was incorrect. The fact that someone at some time had said a particular case was a proposition for something, meant that got repeated from book to book. Many authors simply didn't read their sources or check their references. Students repeated what they had been told incuriously.

Note: this was a graduate level course. Not wet about the ears undergraduates.

The worst material was little potted notes produced for law students. Utterly awful material in most cases.

Anyway, when LLM's became a thing, a lot of what did not feel right about their output and many of their error patterns, reminded me of the experience of teaching masters' students.

One of the saving graces of English court room practice (when I did that sort of thing) was that judges would say to you "where does it say that?" in a case you cited. You had better have them all at your fingertips and know exactly where you had cited. That avoided a lot of hallucination.

Just a random remark which might be of interest.

https://equal-earth.com/images/home/Slide_Show2-6.jpg

It is the most obvious place to look (being on the front page of the website).

Growing up, I did not see Mercator at all often.

Of course video is more easily available now, so there are some aspects of study that may not need the ability to write things down.

Post employment restrictions fall under the doctrine of restraint of trade - which explains why we use "gardening leave". While you are employed you and your employer have a relationship which includes a duty of loyalty by you to your employer. They have duties such as a duty to pay you for your work. By default they should also be offering you work, so a specific "gardening leave" clause is required to keep you at home, bound by restrictive covenants but without work and being paid.

After employment, it is much harder to make restrictions that will bite. Typically an employer must show there is some legitimate interest they are protecting and that it is reasonable to do so.

Hence: non-solicitation of clients tends to be easier to justify, though only if an employee actually had some kind of relationship with or knowledge of those clients. Whereas a non-compete is much harder to justify. It is also harder to make reasonable, hence time/space/sectoral limitations.

Looking at confidentiality: the default established in the amusingly named Faccenda Chicken v Fowler (Mr Fowler was a frozen chicken sales lead). It is that you are bound by the usual duties of confidentiality that arise when anyone has confidential information but that you cannot be prevented from using information gained while working that is part of your normal skill acquired as part of the job unless that is so secret as to amount to a "trade secret".

In this case, Mr Fowler knew where to sell chicken (from having done so). While this was confidential, he was able to use that information (not being a trade secret) in setting up a competing frozen chicken sale network.

The core idea is that you can't stop someone plying their normal trade.

That's the default. Obviously if you sign an agreement that imposes post-contract confidentiality obligations it can go further than Faccenda Chicken, but if it went too far, it would also be vulnerable to the rule against restraint of trade.

In short:this sort of "you cannot work in the industry" idea is very unlikely to work in England and Wales (and I suspect the rest of the UK - though I am strictly only an English lawyer).

That said: there is still value in reading your contract of employment carefully and making sure that you are happy with it.

What happens, he suggests, is occasionally you have significant events which cause one, or sometimes several closely located, languages to spread widely and then you can see a family tree over that period of time. For example: Indo-European. What you can't hope to do is to go back tens of thousands of years.

Obviously he sees things through the filter of Australian linguistics, but he did some excellent fieldwork (which is also very interesting, though it may be out of print) and produced a wonderful monograph on Dyirbal (of "women, fire and dangerous things" fame).

You can "solve" this in Second Order logic, because you have a more powerful induction axiom, but how exactly you define that logic is tricky. There's no proof system that defines it completely, so you have to do this via a semantics that relies on knowing which models are or are not OK.

I don't think it solves the problem that you can't define all the "truths" (as most logicians would put it) of Peano Arithmetic.

Being an ex-proof theorist, I'm a bit dubious of not having one.

You do meet opponents, but in some fields, not often enough to really know.

In crime, your client would be in the cells much of the time and you could sit at the back of the court and watch lots of counsel do a good or bad job.

Of course even the civil bar do get to know each other in the way that commercial transactional lawyers - which is what I do now - don't have as much opportunity to do. I am typically dealing with commercial organizations all over the world. Too many lawyers to get to know.

i.e. even where it isn't opaque, it is surprising how many people mess up.

(Of course that includes lawyers)

People often ask me for recommendations of a lawyer outside my field, or indeed in my field if I can't help them because my firm is conflicted or for some other reason. The thing is I don't personally use lawyers, so I have only the vaguest idea. I can tell some things from firms' reputations and their websites, but even for me it is a hard call.

And the law firms I have most experience of in practice tend to be the ones which I would go out of my way to disrecommend :-).

So I agree, that is a problem. Particularly for litigation. Unless you are doing litigation a lot - which is likely to be bad news for you - you can't get experience of a lawyer and decide that you like them. Word of mouth is also rather less useful.

A specific problem in this scenario is tracked changes. Google has a history/version control but it does not map particularly well onto Word's tracked changes, which the other party will understand and is likely to want to use. Passing things into and out of Google will often result in loss of useful information like that.

Personally: Word is the absolute best piece of software for dealing with numbered lists that is easily available. In many ways it is terrible of course, but it is less terrible than anything else. Getting numbering right is important.

Google has gotten better. It used to be very bad at larger and more complicated documents. But it still doesn't have all I need to write a really good contract (at least by my standards of "really good").

Worse, the ounce, as a unit of mass, has two different meanings in English law, one of which is so obscure and differs so very little from the internationally agreed quantity, that it is hard to believe that it could ever matter, but strictly if you are thinking about the law of food labelling - and I do sometimes - then you always have a slight twinge at the thought.

However, in contradiction to least one remark about the universality of cups, they are not. British cups are, like US cups, half a pint, but British pints are larger. A British cup is 10 fluid ounces while a US cup is 8 (or thereabouts, I am not sure how standardised it is).

So I have to remember to adjust by a factor of 20% when working from a US recipe in the UK, or just convert to weight.

Where I live now a "cup" is typically 200ml unless I am measuring rice into a rice cooker.