CompuServe Messages

1 – An enormous number of people used the "old" speech technology. 2 – For that 1$, they could have left off quite a bit of *garbage* that is still being shipped with the system. Examples from 2.1: "edit", which no one but a very rare unix dinosaur will use (and as you point out, they already HAVE it). Like: "MagTape" which could just as easily come with an HD install disk. You get the idea. Speech was more valuable, on a byte per byte basis, than all of these taken together, in my opinion – which I think represents the majority here. It's also not clear that the speech "cost" an extra disk, not at all in fact. 3 – Lame or not, it *worked* and it was very handy. Amiga users weren't auditioning for the Toastmasters; we were using it to get information from programs which were running on hidden screens or with no other output, and a substantial number of handicapped people were also using it… the potential for more was there too, until the Big Stupid decision was taken. 4 – new speech technology that uses a DSP is a joke – why? Because there IS no DSP available. As per usual, CBM hasn't come up with the promised hardware, period. When and if it ever shows its face, it very likely will not work in some or all of the installed base. Software-only (or at least, software plus *basic* Amiga hardware) based speech has the value that it promotes continuity among all the machines. That was a value in and of itself; as a developer, one could count on the narrator etc "being there" and you *knew* you could use the thing. The more the Amiga market diverges, the less constant the user base is and the more strain is felt by developers who try and reach into the dusty corners and get as many users as possible to use a particular product. Advances – like AGA – are easier to swallow; and DSP based speech _if there was such a thing_, might reduce this to a quieter level of grumbling. But AGA machines are *very* backwards compatible – if a developer decides to write to 6-bit HAM displays, it'll work on a 1200 or 4000/xxx. If a developer decides to write for the (mythical) DSP-based speech, it'll only work on specific machines – and here, my opinion says that almost no one will write anything at all with it in mind, if and when it ever sees the light of day which I quite frankly have serious doubts about given the current publicly known situation at CBM. I know we have NO intention of doing anything in particular with DSP based speech from CBM. Yeah – it's a pain. CBM is a pain. To say the least.