4 ms·

I had a discussion with my boss not long ago, he's a tax lawyer with a few decades under his belt. I asked him simply if he thinks he'll keep his job for more d

I had a discussion with my boss not long ago, he's a tax lawyer with a few decades under his belt. I asked him simply if he thinks he'll keep his job for more decades to come. His answer was quite enlightening, mainly because of how concrete it was: The problem with tax law (which is not the only set of laws, and not contract law either, but perhaps automatable nevertheless) is that it's simply come to the point that it is at right now as a cumulative effort of centuries of work. It works additively and builds structures so complex that the title tax lawyer almost becomes meaningless with so many specialties being required; VAT, corporate and personal income taxes, inheritance and property... Any effort to algorithmically digitise the existing law would fail unless it's able to perfectly replicate it in code. All of the loopholes and exceptions, safeguards and declarations, there's immense inertia ahead. What he on the other hand thinks has been revolutionary is free access to law, advanced text search tools, and software to make filings and communication significantly easier. That's the boring part that speeds your work up so you can spend more time devising plans to get your clients to pay less in taxes, and construct stronger defences. There are many low-hanging fruits in this process; I wish these guys the best of luck, but it seems to me that the comparisons drawn (Intuit, Adobe, etc.) are not analogues. Lawyers aren't seeking these tools for it doesn't provide them value, it only takes away billable hours away.

nicoburns 5y ago
It seems like there could be a lot of value in disentangling the web: repealing the original laws and replacing them with a single law that covers everything in one place. Legal refactoring if you like. I can only imagine that would be a humungous task. And presumably the existence of case law makes it much harder still. It would be quite interesting to think about what the legal equivalent of unit tests would look like. Presumably a large collections of sets of circumstances and desired legal outcomes. It seems to me that this might be an interesting output of legislatures in addition to statutes (although whether they would/should be controlled by the legisalture or judiciary is also quite an interesting question, as our existing system seems to rely quite heavily on the judiciary being able reinterpret laws in ways the original legislators most likely didn't intend)
- cbsmith 5y ago
  The idea that simpler laws won't get exploited is... Well, I get it, because that's how one operates in computer security, but the way the law works is necessarily quite different. It's not designed so much to make doing what is illegal impossible as it is to make it possible to prosecute those doing things that we'd like to discourage.
- jacques_chester 5y ago
  > repealing the original laws and replacing them with a single law that covers everything in one place. Legal refactoring if you like. I can only imagine that would be a humungous task. And presumably the existence of case law makes it much harder still. These are called Reform Acts. They are indeed humungous tasks, rarely undertaken. Where I studied law, in the Northern Territory, a Law of Property Act was passed in 2000 and repealed Acts and overruled case law going back to the 12th century in Britain. It was indeed a massive effort. For some areas of law the Act is also a Code (the Act explicitly says that it is the sole, whole body of the relevant law). This is mostly done for criminal law. Codes are rare in common law jurisdictions like the US, UK, Australia, Canada, New Zealand etc. Codes are much more common in civil law countries like France, where the ideal is for all of law to be centralised and rationalised. Both systems require endless tinkering, debate and reinterpretation, which is why I am skeptical of universalist missions to convert the whole of the law into a system of formal symbols.
  - ska 5y ago
    Confusingly Canada has a mix of both civil and common law, at least in Quebec.
    - a3camero 5y ago
      There are other codes in Canada too, even outside of Quebec, like the Criminal Code of Canada.
      - ska 5y ago
        Right, but that in itself doesn't make it a civil system. If I recall correctly Quebec separates by private and "public" law, so if you get divorced or break up a company it's one system, if you steal from someone, the other. Somehow it's all hybridized and works.
munk-a 5y ago
> Lawyers aren't seeking these tools for it doesn't provide them value, it only takes away billable hours away. All billable hours aren't created equally. If I can automate the work that I currently hire fresh grads to do - the work of actually drafting up contract agreements for review and wills - but price my services just under the prices offered by other firms then I can make a killing. You can purchase a hand carved wooden table for your living room and some people still actively manufacture these "bespoke" items, however most folks shop at IKEA and chose between one of a dozen styles available for a dining room table. These two products are mostly functionally identical in terms of supporting food and weathering the years (if you go by depreciation the IKEA table probably loses value slower than a hand carved table - and you can replace it pretty much at will). There will likely always be contract lawyers around to sort out large M&A agreements and other high priced transactions, but wills and simple employment contracts - delivering these in a correct format (i.e. not just a download a sample contract and fill in names possibly resulting in something unenforceable) will be a game changer. Taxes are extremely complicated - but they got that way by design. The US Tax code is the result of one of the most insane applications of regulatory capture in the world and it's there to allow Intuit to continue printing money at the expense of the tax payer - simplifying those (like for real - not the 2017 version) might not end up happening for decades as there is a ton of money entrenched - so that seems like an incredibly poor part of the market to initially target. Contract law though, it's not simple but it's so utterly clear and actually codifying it would likely greatly aide law makers in discovering and resolving any existing loopholes and might even provide some useful insight into how the US could better handle the question of employee vs. contractor by enumerating rights and privileges in a clearer format. And wills, pretty much everyone seeks independent legal council to write one of these at some point - it is highly advisable depending on where you live (in Canada a lack of a clear will causes assets to be tied up by the government for an extended period of time) so you could make a lot of money edging the market out there in particular.
- ska 5y ago
  > If I can automate the work that I currently hire fresh grads to do - the work of actually drafting up contract agreements for review and wills - but price my services just under the prices offered by other firms then I can make a killing. This is probably true, but has a definite knock on effect that breaks the existing training model pretty badly. I suspect the longer term stable solution is relatively fewer working lawyers relying more on tools - but transitioning the industry to that model will be painful.
mengwong 5y ago
Sure, but the intended user for a lot of this stuff isn't the lawyer – it's the end-user who's glad to have an alternative to the lawyer! Like, owning a bicycle means I don't need to call a taxi for some trips. Not all trips, just the local ones, but that's a lot of my trips. Free access to law, and software for filings like afterpattern.com, that's the boring stuff that lawyers don't want to do and I don't want to pay for, so win/win. And that's all part of the larger vision of computational law that Legalese wants to help realize. Thanks for the good wishes and I hope to deliver on the vision, we've been at it for five years and will probably be at it for another ten before we start seeing results percolating into the real world.

nicoburns 5y ago

It seems like there could be a lot of value in disentangling the web: repealing the original laws and replacing them with a single law that covers everything in one place. Legal refactoring if you like. I can only imagine that would be a humungous task. And presumably the existence of case law makes it much harder still. It would be quite interesting to think about what the legal equivalent of unit tests would look like. Presumably a large collections of sets of circumstances and desired legal outcomes. It seems to me that this might be an interesting output of legislatures in addition to statutes (although whether they would/should be controlled by the legisalture or judiciary is also quite an interesting question, as our existing system seems to rely quite heavily on the judiciary being able reinterpret laws in ways the original legislators most likely didn't intend)

cbsmith 5y ago
The idea that simpler laws won't get exploited is... Well, I get it, because that's how one operates in computer security, but the way the law works is necessarily quite different. It's not designed so much to make doing what is illegal impossible as it is to make it possible to prosecute those doing things that we'd like to discourage.
jacques_chester 5y ago
> repealing the original laws and replacing them with a single law that covers everything in one place. Legal refactoring if you like. I can only imagine that would be a humungous task. And presumably the existence of case law makes it much harder still. These are called Reform Acts. They are indeed humungous tasks, rarely undertaken. Where I studied law, in the Northern Territory, a Law of Property Act was passed in 2000 and repealed Acts and overruled case law going back to the 12th century in Britain. It was indeed a massive effort. For some areas of law the Act is also a Code (the Act explicitly says that it is the sole, whole body of the relevant law). This is mostly done for criminal law. Codes are rare in common law jurisdictions like the US, UK, Australia, Canada, New Zealand etc. Codes are much more common in civil law countries like France, where the ideal is for all of law to be centralised and rationalised. Both systems require endless tinkering, debate and reinterpretation, which is why I am skeptical of universalist missions to convert the whole of the law into a system of formal symbols.
- ska 5y ago
  Confusingly Canada has a mix of both civil and common law, at least in Quebec.
  - a3camero 5y ago
    There are other codes in Canada too, even outside of Quebec, like the Criminal Code of Canada.
    - ska 5y ago
      Right, but that in itself doesn't make it a civil system. If I recall correctly Quebec separates by private and "public" law, so if you get divorced or break up a company it's one system, if you steal from someone, the other. Somehow it's all hybridized and works.

munk-a 5y ago

> Lawyers aren't seeking these tools for it doesn't provide them value, it only takes away billable hours away. All billable hours aren't created equally. If I can automate the work that I currently hire fresh grads to do - the work of actually drafting up contract agreements for review and wills - but price my services just under the prices offered by other firms then I can make a killing. You can purchase a hand carved wooden table for your living room and some people still actively manufacture these "bespoke" items, however most folks shop at IKEA and chose between one of a dozen styles available for a dining room table. These two products are mostly functionally identical in terms of supporting food and weathering the years (if you go by depreciation the IKEA table probably loses value slower than a hand carved table - and you can replace it pretty much at will). There will likely always be contract lawyers around to sort out large M&A agreements and other high priced transactions, but wills and simple employment contracts - delivering these in a correct format (i.e. not just a download a sample contract and fill in names possibly resulting in something unenforceable) will be a game changer. Taxes are extremely complicated - but they got that way by design. The US Tax code is the result of one of the most insane applications of regulatory capture in the world and it's there to allow Intuit to continue printing money at the expense of the tax payer - simplifying those (like for real - not the 2017 version) might not end up happening for decades as there is a ton of money entrenched - so that seems like an incredibly poor part of the market to initially target. Contract law though, it's not simple but it's so utterly clear and actually codifying it would likely greatly aide law makers in discovering and resolving any existing loopholes and might even provide some useful insight into how the US could better handle the question of employee vs. contractor by enumerating rights and privileges in a clearer format. And wills, pretty much everyone seeks independent legal council to write one of these at some point - it is highly advisable depending on where you live (in Canada a lack of a clear will causes assets to be tied up by the government for an extended period of time) so you could make a lot of money edging the market out there in particular.

ska 5y ago
> If I can automate the work that I currently hire fresh grads to do - the work of actually drafting up contract agreements for review and wills - but price my services just under the prices offered by other firms then I can make a killing. This is probably true, but has a definite knock on effect that breaks the existing training model pretty badly. I suspect the longer term stable solution is relatively fewer working lawyers relying more on tools - but transitioning the industry to that model will be painful.

mengwong 5y ago

Sure, but the intended user for a lot of this stuff isn't the lawyer – it's the end-user who's glad to have an alternative to the lawyer! Like, owning a bicycle means I don't need to call a taxi for some trips. Not all trips, just the local ones, but that's a lot of my trips. Free access to law, and software for filings like afterpattern.com, that's the boring stuff that lawyers don't want to do and I don't want to pay for, so win/win. And that's all part of the larger vision of computational law that Legalese wants to help realize. Thanks for the good wishes and I hope to deliver on the vision, we've been at it for five years and will probably be at it for another ten before we start seeing results percolating into the real world.