Writing a pager isn't that hard.

You might want to use sys* commands to avoid Perl buffering:

sysopen your file

sysseek to the position

Read a fixed length block and look for a line ending

Show a line if you find a break, read another block if there is no line ending left

Repeat the last two steps until as much lines are shown as you like

You may want to save the positions of the last line beginnings using systell to directly sysseek to lines above and below the current screen.

Writing a complete less-like pager may be require more features, but a basic one may also do the job if you're only viewing logfiles.

I try to use the head and tail Linux commands to limit the number of lines when working with huge logfiles, for example "tail -n 25000 httpd/access.log" to get a day on a medium busy webserver. grep is also good if you're searching for specific information, less and it's coloring will do a good job when viewing the results.