 |
||
Telephone: 07725 735255 |
||
Note: Some platforms warn you about “active content” that could access your computer. If you don’t trust me not to give your computer a virus, you can browse this website without allowing the active content. The only difference is that the buttons below would not be highlighted when the cursor moves over them.
My greatest achievement to date has been the development of a computer simulation language called SDML – which stands for ‘Strictly Declarative Modelling Language’. The language’s official home page is at http://sdml.cfpm.org and I have set up a ‘strictly-declarative-modelling-language’ discussion group that can be used to discuss the language (but it has generally been used as a repository for my left-wing messages on a range of subjects).
SDML is a modelling language with the following features:
I was the sole implementer of SDML, which I wrote in Smalltalk. Other members of the Centre for Policy Modelling at Manchester Metropolitan University were involved in designing the language, particularly my boss Scott Moss and logician and Artificial Intelligence (AI) programmer Bruce Edmonds.
SDML is quite a difficult language to learn with limited documentation. I strongly recommend learning Prolog before attempting SDML. Prolog is the original logic programming language and the backward chaining aspects of SDML are very similar to Prolog, but without the imperative facilities (such as the “cut”) that undermine Prolog’s relationship with logic. I put a lot of effort into finding efficient ways of doing things declaratively, for both forward and backward chaining, to avoid such compromises.
For some advice on textbooks to use to learn Prolog, click here. I think I learnt Prolog via an earlier edition of Clocksin and Mellish’s Programming in Prolog.
There is a tutorial on the SDML website. Getting through it and understanding the concepts that are introduced requires a lot of concentration and I recommend putting aside a serious amount of time (a few days) before attempting it.
Until now (I am updating this page at the end of 2007), I have hidden away the flaws in SDML on this website to try to avoid bad people (who want an unethical society to continue) finding out what they are, because this could have helped them get round the problems and improved their modelling of the world (particularly if they had access to the source code as some do). I have considered trying to rectify these problems myself by getting a job at the Artificial Intelligence Applications Institute in Edinburgh. I am now confident that good people are in a strong enough position in the world, for me to provide advice on those who may want to improve SDML, to both good and bad people, but I would prefer it if computers did not get too powerful until we have a much more ethical world so I have no plans to improve the language myself in the near future.
The main limitations of SDML are as follows:
For some advice on textbooks to use to learn Prolog, click here. I think I learnt Prolog via an earlier edition of Clocksin and Mellish’s Programming in Prolog.
I see society as a struggle between good and bad conspiratorial organisations and individuals (who sometimes act conspiratorially on their own but are generally more open about their true views and intentions). Good people are in favour of creating a more ethical world and bad people want to preserve the status quo or make things even worse.
In the past, bad conspiratorial organisations, including government agencies like MI5 and the CIA, had access to the best technology and I see my foray into AI as a way of providing good conspirators with cutting edge software that could be used (and I suspect has been used but to what extent is unclear) to change the world. Despite the flaws mentioned above, SDML is the best language of its kind that I know of (similar software may have been secretly developed taking my research into account), and I have publicised it for several years on my socialist website (at the old address www.stevewallis.org before it moved to www.socialiststeve.me.uk). I also told people on what looked like a left-wing hackers mailing list, called “hactivism”, about SDML.
Of course, I provided my enemies with the software too. The early academic papers on SDML, primarily written by Scott Moss, all appeared in business/economics/marketing journals so tended to reach right-wing conspirators more than left-wing ones. I never managed to get a technical article on the techniques I used to develop the software published in a refereed AI publication – when I tried at first, my use of different terminology put off the reviewers. but a rewritten article was rejected too on the basis that it was not original (even though it was and the supposedly similar software was imperative and didn’t even work). Never mind, in these days of the internet, my research centre, the Centre for Policy Modelling (CPM), has been able to publish articles itself; a selection of the most relevant articles on SDML written when I was there are mentioned on the SDML home page (which I wrote) and others are accessible from the CPM website.
I have sometimes overestimated the power of SDML and AI generally, thinking that a complete model of society has been developed on computers that is capable of predicting precisely what will happen and planning a revolution (or planning to stop one). The free will of individuals, including myself, makes that infeasible. Nevertheless, I have no doubt that parts of the problem of changing the world have indeed been modelled on computers, to supplement the modelling that goes on within human minds.
I have sometimes been of the opinion that there are robots in society, or people being wholly controlled remotely by a form of mind control, perhaps using mobile phone masts, genetically modified food and drink (although this is only applicable to affecting masses of people because it cannot be directed to individuals), and/or nanotechnology with mini robots moving in our blood stream as suggested in the comedy science fiction TV programme Red Dwarf. If such mind control has been used, both software and people interacting with it have jointly affected what controlled people have done. Maybe I should admit to have not completely rejected these ideas, although I think I have exaggerated the power of computers. Whatever the truth of such conspiratorial ideas, I have come across what appears to be an AI program on the internet masquerading as a human; visit the Rosa Lichtenstein page on this website for details.
For more information about SDML, visit the language’s official home page at http://sdml.cfpm.org or the ‘strictly-declarative-modelling-language’ discussion group that can be used to discuss the language (but it has generally been used as a repository for my left-wing messages on a range of subjects). You may also like to visit my Computer Science page, which contains some information about my development of SDML.
For more information about AI, visit the moderated comp.ai newsgroup, or unmoderated groups such as comp.ai.shells, comp.lang.prolog and comp.lang.functional (which you can do by going to Google, choosing Groups and searching for the newsgroup name).
