The LSE Rexx interface - cryptic commands, insufficient use of return codes.

See evaluations for the evaluation criteria.

Not well defined

The documentation for the LSE ARexx interface is very skimpy. It just lists each command, and what action that command takes. There is no discussion of how the commands behave in extraordinary conditions - for instance, what does "next line" do if there is no next line? Can "use characters" be used to enter commands, or just text? None if this is explained in the manual.

Complete for the application, not for Intuition

Even though every command may not exist in the LSE ARexx command set, they can be accessed through the menus, which are available from ARexx. Exactly how isn't clearly documented, but it is apparently the case.

Consist with themselves, if nothing else

The LSE ARexx commands are consistent with themselves. However, they don't look like ARexx commands. They don't manipulate textual data as a ARexx command would be expected to, they don't use the ARexx return codes or the RVI at all.

Not readable in any way

The LSE ARexx commands are two-character sequences. This makes it hard to remember what to write to issue a command, and nearly impossible to read. Some commands require a subcommand, and you can apparently issue multiple LSE commands in a single ARexx command (again, this isn't clearly documented). This leads to scripts being at best difficult to debug.

On to the MFF+ evaluation, back to the AmigaVision evaluation, or up to evalutions.