From ... Path: archiver1.google.com!newsfeed.google.com!newsfeed.stanford.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.icl.net!lnewspeer00.lnd.ops.eu.uu.net!emea.uu.net!npeer.kpnqwest.net!reader3.kpnqwest.net.POSTED!not-for-mail Newsgroups: comp.lang.lisp Subject: Re: size of executables? References: <1hqun9.nv.ln@hermit.athome> <3209635698009992@naggum.net> <6tv2o9.5n.ln@hermit.athome> Mail-Copies-To: never From: Erik Naggum Message-ID: <3209668982240409@naggum.net> Organization: Naggum Software, Oslo, Norway Lines: 49 User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 16 Sep 2001 22:43:02 GMT X-Complaints-To: newsmaster@Norway.EU.net X-Trace: reader3.kpnqwest.net 1000680182 193.90.206.213 (Mon, 17 Sep 2001 00:43:02 MET DST) NNTP-Posting-Date: Mon, 17 Sep 2001 00:43:02 MET DST Xref: archiver1.google.com comp.lang.lisp:16334 * Aleksandr Skobelev <2269@mail.ru> > You are kidding. Need I explain you meaning of "standalone executable" > term in my messages? Yes. Because if you compare a hello, world program that uses massively optimized operating system facilities and shared libraries and all sorts of other things that work very well to reduce the size of "executables", you are looking at the wrong end of the stick. Look at what that small executable is attached to: A huge set of libraries and support functions from the operating system. I mean, I could write a hello, world program in MACRO-10 for TOPS-10 that consisted of three machine instructions and the string literal. That was done at a time when small _actual_ executables were worth real money, but these days, small executables are not small in memory imprint, any longer. Therefore, the _concept_ "standalone executable" is no longer valid. > Computers programs are intendent to be using but not debugging by other > people. This is so broad as to be utterly false by the very nature of the claim. Please limit your statements to the assumptions that support them, then always evaluate the assumptions for validity for broader claims. Otherwise, you will always be stuck in using your observations from a very limited context in a much broader context, and that never leads to anything good. > So if your programm will be able to run only in CL enviroment with all > its really nice development features which are not required by program's > semantics, then you just shift off a cost of these features on an user. How did you arrive at this strange conclusion? > Sometimes such features have to be in program, sometimes have not. I > like freedom to choose use or not such features in a distributed program. Sometimes, we do not get the freedoms we want, but instead may other ones we did not know we wanted because we never asked for them. You are way too quick for my taste to dismiss what you do not know, yet. I find this very annoying, in fact, and suspect that you will not want to invest the time it takes to learn a new language. I suspect you will only learn C++ with a new syntax if you learn Common Lisp in your present state of mind. _Please_ look at what you get, not at what you want. Somehow, you _did_ do that when you came to C++ and found "small executable" a valid concern. You have to be equall open-minded with _every_ language you approach, not just the first. ///