WikiLaw (3)

From Meta, a Wikimedia project coordination wiki
Jump to navigation Jump to search
This is a proposal for a new WMF sister project.
WikiLaw
The logo for the project, if you already have one
Status of the proposal
Statusunder discussion
Details of the proposal
Project descriptionLaws are all mostly already in the public domain. The court cases and decisions are in the public domain too. But its all text, not fit for machines. It should be organised and broken down into a Wikidata-like structure to make it machine readable. Required for research in legal fields using advanced computing solutions.
Is it a multilingual wiki?One multilingual wiki like wikidata.
Potential number of languagesMultilingual
Proposed taglineFree legal repository
Technical requirements
New features to requireA mixture of Data of Commons (nosql structure part) and wikidata (ui and editing simplicity part).
Development wikiMJL Testing Wiki
Interested participants
List of project participants

For legal research all the laws, regulations and legal decisions need to be indexed to make them machine readable.

Proposed by[edit]

Alternative names[edit]

Related projects/proposals[edit]

Domain names[edit]

Mailing list links[edit]

Demos[edit]

People interested[edit]

  • Ie.sanm (talk) 13:55, 30 August 2019 (UTC)
  • Capankajsmilyo (talk) 16:40, 11 May 2019 (UTC)
  • Arep Ticous 16:43, 15 May 2019 (UTC)
  • Very interested. –MJLTalk 23:26, 30 May 2019 (UTC)
  • Fungster (talk) 06:42, 22 June 2019 (UTC)
  • When I first heard about this I thought "meh", but after reading the proposal this could fill a useful, albeit small area of need. Would be very useful for legal research for lawyers/paralegals, government, and private individuals if done well. – Ajraddatz (talk) 17:29, 1 July 2019 (UTC)
  • Interested. Making this globally-available would be a challenge though. Altoria 2016 (talk) 07:21, 19 July 2019 (UTC)
  • James Martin - struggling to get a login to Wiki (I have sent an "unblock Ticket Request System") - There is massive potential for a territory/Country Application that logically codes Laws to eradicate "contract frustration" in Laws. A critical point could be reached whereby all related laws are coded and decentralised legal resolution is possible between 2 parties - independently of the "State" and the Judiciary for limited legal issues at virtually zero cost as resolution would be on its own blockchain. 02:15, 6 August 2019
  • Xover (talk) 07:01, 2 September 2019 (UTC)
  • --Rubbish computer (Talk: Contribs) 16:26, 20 October 2019 (UTC)
  • Supertrinko (talk) 03:06, 19 December 2019 (UTC)
  • J947: 05:19, 14 March 2020 (UTC)
  • Agree that this would be pretty useful --DannyS712 (talk) 05:24, 19 March 2020 (UTC)
  • Yes, we need the global law database. Фред-Продавец звёзд (talk) 19:36, 28 June 2020 (UTC)

Discussion[edit]

  • @Capankajsmilyo:. To who would this be useful to? and how would this be useful to the society? ArepTicous talkcontribs 15:53, 15 May 2019 (UTC)
This would be useful to law students, researchers, lawyers, machine learning community, and many others. This will help in analysing law through history, mapping changes with change in society and communities. This will help in understanding similarities and differences in legal procedures among various countries. This will help in making law readable and accessible to common people, publishers, reporters and other users as most of legislations are either not indexed, or not readable, or not in languages, or not available online. India alone has thousands of acts but at max hundreds can be accessed. Even they are either pdf or photograph, or if they are in text, its not in structured format. Further, as listed above there are 3 similar proposals already, so those can be subsumed in this one as well. Capankajsmilyo (talk) 16:13, 15 May 2019 (UTC)
  • @Capankajsmilyo: Well...I am happy to say that i Support Support this. It's good that you are renewing a long lost proposal. Arep Ticous 16:45, 15 May 2019 (UTC)
  • Support Support Assuming that the process of getting legal information and adding it to a wiki is practical. SelfieCity (talk) 17:09, 8 June 2019 (UTC)
    @SelfieCity: very practical! Almost all laws are written down somewhere. The problem is that they are constantly changing and aren't a good fit for Wikisource. –MJLTalk 00:59, 9 June 2019 (UTC)
  • Support Support - So few suggested projects actually could be beneficial, but I think this is one that could be. There is a major set-up like this atm...but it's expensive. So it's a good area to branch into. Nosebagbear (talk) 20:27, 19 June 2019 (UTC)
  • As a long-time author of articles on legal matters on German Wikipedia, I gather that legal research with quantitative methods has only just begun, and I would like to stress that only official sources will qualify for this. What we need is more open data, and, indeed, structured open data, but it needs to be official data to be used for scientific purposes, cf. the recent article by Coupette, Corinna and Andreas M. Fleckner. 2018. Quantitative Rechtswissenschaft: Sammlung, Analyse und Kommunikation juristischer Daten. JuristenZeitung 73, Nr. 8 (April): 379–389. doi:10.1628/jz-2018-0020 along with the accompanying dataset at https://www.quantitative-rechtswissenschaft.de/ . So, I am sorry to say that this proposal is indeed interesting, but it is of little practical worth only. Linguistic analysis of laws and court decisions are already possible today. Other data are just not available because they are stuck in the records and have never been digitised and, thus, cannot be published and analysed anyway. What's more, this would be a strictly scientific project which means that not everyone would be able to contribute. Only validated data can be used for that which means that you would need very strict and tight QA from the very beginning. If you would like to realise a kind of Wikidata for legal sources, however, you are still able to do this within Wikidata.--Aschmidt (talk) 18:07, 5 July 2019 (UTC)
  • Oppose Oppose: Most legal codes and case law are not public domain; while you'd be able to get all US Federal cases, statutes, and regulations, a large amount of state material would not be useable. I doubt much law from outside the US is public domain as well. Moreover, I don't understand what this "wikidata like structure" is supposed to achieve. The only thing you might be able to do is create a citator (like Lexis' Shepard's and Westlaw's KeyCite), but there are loads and loads of editorial judgment calls in deciding things like whether and how one case distinguishes another, or whether and how one case abrogates another, or whether and how a statutory enactment supersedes a particular case. I can foresee significant problems with data quality and annotation quality as well, Aschmidt touches on. Even keeping your regulatory dataset up-to-date would be a monumental undertaking. Just go to regulations.gov and look at how large each day's issue of the Federal Register is. Each one of those issues every single day has things that may be regulations that may need to be inserted in your Code of Federal Regulations structure (as well as a large amount of material that is not a final rule). —/Mendaliv//Δ's/ 05:31, 6 July 2019 (UTC)
    @Mendaliv: I know what you are saying, the law won’t apply in all archived laws, plus current laws are unfit on Wikisource. 92.8.113.109 10:27, 18 July 2019 (UTC)
    @Mendaliv: We have c:Template:PD-EdictGov which pretty much says that no government law can be copyrighted. Since the servers are located in the United States, all it would take is a foundation statement (which would be required anyways to launch the project) saying PD-EdictGov can be applied universally. That's pretty much been de facto the case anyways on Wikisource. –MJLTalk 03:14, 27 July 2019 (UTC)
  • Support Support - Expanding on the idea of a Decentralised Legal Resolution at virtually zero cost (I have been dabbling in crypto currencies, and I am very interested in the capabilities of the technology - which may not be for current Wiki platforms). If a blockchain can be produced for one territory/country and tested, this could be rolled out to all possible territories/countries and left to evolve. An incentive would be that the transcriber of the law to the blockchain and the coder of the Legal Articles, share a percentage of capped fees, if eventual decentralised legal resolution is enacted in the future via 2 parties (with all resolution recorded on the blockchain (anonymity default for individuals). There would be a year or two or maybe more until a significant number of laws are transcribed and coded (there are loads of legal students who could participate...) Coding the articles in the laws will be incredibly difficult, but if the software is developed this should not be to onerous. (I have experience of complicated legal agreements, such as Highway Maintenance agreements (spent 3 years writing), and currently redrafting the Tenant Management Organisation Modular Management Agreement 2012 which is an almost unusable piece of UK legislation. I am migrating TMOs from the previous 2005 and 1994 versions (took me months of research to find the official pdf on the government archives and I put it on this Wiki Page https://en.wikipedia.org/wiki/Tenant_management_organisation)). The subjects that a law applies to, and the cross references to other laws are all critical. Laws would have to be coded in a cognitive bias free way - again this would have to be peer reviewed or programmed into the software. James Martin - awaiting an Unblock Ticket Request System so I can create an account 02:24, August 2019
  • Support Support with Comment Comment if you were to create such a source... how would someone use it... in a more brief way... how would someone locate the files and do they have any other site like this which already exists... 43.250.242.189 02:04, 7 September 2019 (UTC)
  • @Capankajsmilyo: vandalism may be a problem here, and someone may vandalise a page and then it is not detected. What if the vandalism involves changing the outcome of the court resolution to a fake one? Nigos (talk) 07:50, 22 September 2019 (UTC)
  • @James_Martin001 Without going into too much technical detail, Blockchains are secured by calculations and nodes which is difficult to unwind the calculations via a 50% node attack. Basically this would be very expensive via targeted computing power to do for the Bitcoin Blockchain, which is secured by mining. There are non mining Blockchains and Blockchains for Smart contracts such as on the Ethereum Platform or Tezos which apparently has amendable governance. A Blockchain tech will present itself in the the next 10 years which can not be out forked. This means that there are no other coding evolutions that can branch off and be better than the original branch. There is a long way to go on this. But this does not stop the Legal Community to come up with the logic protocols to code laws. As I said in my first post, 'Contract Frustration' in the legacy legal system is unworkable. 18:16, 14 March 2020 (UTC)
  • @James_Martin001 What year in the 21st century will a 'decentralised legal resolution' solution overtake government legislatures and basically make them irrelevant? This is probably the single most important question for humanity, otherwise politicians carrying on in a biased way will probably lead to our extinction. A sub-text to this is that AI - Artificial Intelligence, needs to be stopped from using the legal logic for it's theoretical own ends (personally don't think it is possible to make AI self aware, as we only give the impression that it is self aware... stories of blind people getting their sight back late in life was so disturbing that they wanted to be blind again causes me to be cautious about AI potential). If we all individually own our data for example on facebook, we could set the cost for reusing or viewing our data. Likewise every "AI computer unit" must have an individual who is legally responsible for its coding and outputs. Micropayments are now possible for data ownership royalties via cryptocurrency wallets and addresses, or legal fees incurred. My original support post alluded to micropayment incentives for editing a law blockchain. For example if I coded the Equalities Act 2010 (UK) 100%, I could get an author fee every time it is used in legal resolution. Clearly this could be amended over the years or completely replaced and my percentage would be reduced to reflect this. The economics needs to incentivise the realtime maintenance of the laws, for the benefit of everyone. Blockchains solutions could be for a private company, or open source such as Bitcoin which no one owns. Satoshi Nakamoto the anonymous pseudonym for the creator or creators of Bitcoin, was very intentional to avoid jurisdictions for legacy resolution. Legal Professions have not caught up with the implications of a borderless protocol yet, the original Bitcoin White paper was in January 2008! 18:48, 14 March 2020 (UTC)
  • Support Support - This would be a monumental task and would never be completely up to date on all international law, however the bigger it gets, the more useful it would be. I myself have interest in comparing similar Statutes from different jurisdictions and a resource such as this would be useful. There exists the Free Access to Law Movement (FALM) which started with the creation of the Legal Information Institute (LII), which aims to provide free access to legal information. You'll find there's similar namesake movements in many countries/regions that attempt to provide the same in their own jurisdictions. The proposed WikiLaw could be a way to combine the efforts of all these institutes and create a single repository of this knowledge, presented in a much more powerful way, allowing users to make use of the service in ways we wouldn't be able to anticipate yet. Supertrinko (talk) 03:51, 18 December 2019 (UTC)
    • Ah yes, "combine the efforts". https://xkcd.com/927/ Nemo 10:46, 18 December 2019 (UTC)
      • One of my favourite xkcds and works well when there are competing standards that someone is trying to make universal. But in this instance, the various LIIs do not compete with each other. In fact they frequently work together and link to each other, they just stick to legislation within their own jurisdiction so that they have a maintainable scope for their resources and funding prospects. They're based on each other and follow similar philosophies. In addition, many of these services don't have the capacity to transcribe legislation like WikiLaw would attempt to do. Many of them simply find legislation wherever it is available, and scan it onto their website as PDFs. Much of the work I'm doing right now is on historical social security legislation, comparing social security changes year-on-year. Unfortunately all I can do is scan through PDFs each year via the New Zealand LII website. It's been useful, but having a more powerful database that can compare revisions to legislation year-on-year would be fantastic. Right now WikiSource seems the only place to upload legislative documents, and indeed there are hundreds on there, but as an example, just my local Social Security Act has hundreds of revisions, amendments, and re-issues. All of which have slight changes that would need to be documented. That'd be a pain on WikiSource. Having a dedicated project that made this sort of thing easier to manage would be great. Supertrinko (talk) 02:21, 19 December 2019 (UTC)
  • Oppose Oppose. You can read the laws in the official publications. BoldLuis (talk) 00:33, 2 May 2020 (UTC)