Re: Mathematica and Lisp
- To: mathgroup at smc.vnet.net
- Subject: [mg129912] Re: Mathematica and Lisp
- From: Richard Fateman <fateman at cs.berkeley.edu>
- Date: Mon, 25 Feb 2013 02:21:58 -0500 (EST)
- Delivered-to: l-mathgroup@mail-archive0.wolfram.com
- Delivered-to: l-mathgroup@wolfram.com
- Delivered-to: mathgroup-newout@smc.vnet.net
- Delivered-to: mathgroup-newsend@smc.vnet.net
- References: <kcqkv4$lq5$1@smc.vnet.net> <kgc53j$gb8$1@smc.vnet.net>
On 2/23/2013 8:31 PM, Peter Klamser wrote: > Hey, what a useless discussion for different items: > > Mathematica is complex software which uses Lisp like language as a > base system. Um, no. It is written in a dialect of C/C++, Java, and "Mathematica" see http://reference.wolfram.com/mathematica/tutorial/TheSoftwareEngineeringOfMathematica.html I don't know exactly what you mean by a "base" system, but I assume you are talking about implementation. (Not the surface language that has really been the topic of this discussion.) It is useless to open a Lisp interpreter and to try to > emulate Mathematica, because you need 10^10 lines of code to provide > Mathematica like results. This is hardly true. Mathematica (the system) according to the link above, has "several million" lines of code, by which I assume it is something like 2* 10^6, not 10^10. This would make it about as large as (according to one estimate) Microsoft Word. Windows XP, according to a Wikipedia article, has 45 million source lines of code. Mozilla, which is just a web browser, has over 8 million. There are many commercial and military projects of larger size. Strictly judging by lines of code, and counting lines already written in Lisp to do computer algebra, I think that you will find that, in various places, something more than 2* 10^6 lines have already been written. Of course these are not dedicated to emulating Mathematica precisely, and since they are in several competing systems, there is substantial overlap. The impression that one can readily get from reading the Mathematica literature is that the program is one of the most ambitious, complicated, sophisticated and large pieces of software ever written. (see the same link!) As suggested above, this claim is not supported by facts. Nevertheless, this is off the topic (which originally had to do with learning Lisp as a USER writing programs in the Mathematica language). You seem to think that the base language for Mathematica could not possibly be Lisp ... > Nobody tries to write his own Lisp modules to generate a system, that > behaves like AutoCad (which uses AutoLisp...)... Based on what evidence? Do you know that operating systems have been written in Lisp? It is possible to find web browsers, graphics, editors, etc written in Lisp, and certainly computer algebra systems. > > So why is such a nonsense discussed here for some weeks askes > > Peter Presumably because some people find it interesting, and Steve Christensen (the moderator) doesn't find it too offensive :) You have contributed, yourself, in some small way, to the discussion. RJF > > > 2013/2/23 David Bailey <dave at removedbailey.co.uk>: >> On 22/02/2013 06:05, Richard Fateman wrote: >>> >>> Anyway, getting back to Mathematica and Lisp... >>> Since Lisp programs tend to be short, there are fewer opportunities >>> for bugs. Mathematica programs can be short too, but the irregular >>> syntax makes them harder to read. See djmpark's comment about FullForm >>> below. Lisp is like FullForm all the time. >>> >> >> Yes Lisp is rather like FullForm - but who in their right mind would >> write all their Mathematica code in FullForm! >> >> The beauty of FullForm, is that you can render an expression into this >> form to resolve any misconceptions you may have about operator >> precedences etc. >> >> Since practically nobody writes all their Mathematica code in FullForm, >> it follows, does it not, that they must find Mathematica code >> significantly (I would say massively) more readable than Lisp! >> >> David Bailey >> http://www.dbaileyconsultancy.co.uk >> >> >