We can drop 2.4.16 since we have 2.4.18 (I think this is already done). If we add kernel-source-2.4.18 and kernel-source-2.2.20, these are the popularities in decreasing order:
kernel-image-2.2.20 418 kernel-source-2.4.18 150 ---------- we could cut here ---------------- kernel-image-2.4.18-686 29 kernel-source-2.2.20 28 kernel-image-2.4.18-k7 18 kernel-image-2.4.18-k6 12 ---------- or maybe here -------------------- kernel-image-2.4.18-686-smp 10 kernel-pcmcia-modules-2.4.18-686 6 ---------- or maybe here -------------------- kernel-image-2.4.18-586tsc 4 kernel-image-2.4.18-bf2.4 3 kernel-image-2.4.18-386 3 kernel-pcmcia-modules-2.4.18-k7 2 kernel-image-2.2.20-udma100-ext3 2 kernel-image-2.2.20-reiserfs 1 kernel-image-2.2.20-idepci 1 kernel-image-2.2.20-compact 1 kernel-pcmcia-modules-2.4.18-k6 0 kernel-pcmcia-modules-2.4.18-686-smp 0 kernel-pcmcia-modules-2.4.18-586tsc 0 kernel-pcmcia-modules-2.4.18-386 0 The number at the right is the fourth number in the `all-popcon-results.txt' file in http://www.debian.org/~apenwarr/popcon/ (downloaded today) which seems to be the significant one for the kernels. I would start dropping packages from the end of this list upwards until the TeX task fits (btw: tetex-bin is used regularly by 715 entries in the contest). Thanks. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
