I've only been playing (sorry USING) Linux for a few months. For an application which only has a small (250MB) compact flash card as a hard disk my colleagues and I created a small Linux installation with just the minimum number of tools installed. Then an application came along which needed us to be able to have a more graphical front end so we decided we would port XFree onto our system. We down loaded binaries for XFree4.4.0 for Linux running on an X86 processor with GlibC2.2 because we knew that was what we were going to be running on. Then what do we find. The install script starts to try and use tools which we don't have installed (grep). So we printed out the install script to see what is going on. 28 pages later we are not really any better off. So we searched the XFree website and found a document on doing a manual install. WOW the install is only a dozen or so commands. We ran all of the commands as per the document and 5 minutes or so later we had a working version of X installed on our system. Since then I've spent a few bleary eyed evenings pouring over the install script and have a simple question. Why if I have gone to the trouble of downloading a specific set of binaries for a specific processor (X86) a specific operating system (Linux) and a specific version of GLibC (2.2) does the install script need to go and check everything. Surely it is not beyond the wit of man (or developers) to produce a specific script for the specific set of binaries that have been downloaded. I realise that this would involve a little work and that several different versions of the script would need to be maintained BUT do these scripts really change that much between releases. The script for 4.3.0 look identical to the script for 4.4.0 apart from the version number. Kevin Benstead Senior Engineer R&D Division Thermo Electron Corporation Radiation Measurement & Protection Beenham, UK _______________________________________________ XFree86 mailing list XFree86@xxxxxxxxxxx http://XFree86.Org/mailman/listinfo/xfree86
